[{"data":1,"prerenderedAt":6200},["ShallowReactive",2],{"blog-posts":3},[4,272,507,808,1067,1294,1574,1855,2077,2295,2658,2904,3306,3515,3743,3948,4279,4478,4613,4719,5123,5239,5370,5488,5843,5938],{"id":5,"title":6,"author":7,"body":8,"date":257,"description":258,"draft":259,"extension":260,"image":261,"meta":262,"navigation":263,"path":264,"seo":265,"stem":266,"tags":267,"updated":261,"__hash__":271},"blog\u002Fblog\u002Fai-that-sounds-like-you.md","Getting AI to write in your voice (and why \"write like me\" fails)","SelfSM",{"type":9,"value":10,"toc":246},"minimark",[11,15,18,23,26,50,53,67,76,80,83,89,95,101,104,108,111,117,128,134,140,144,147,150,153,161,165,168,176,183,187,190,201,207,211,217,223,229,240],[12,13,14],"p",{},"You paste three of your best posts into ChatGPT, add \"write in this voice\", and get back something that is unmistakably not you. It has your topic and none of your speech. Add more examples and it gets more average, not more accurate.",[12,16,17],{},"This is not a prompting skill issue. The instruction \"write like me\" asks the model to infer a voice from evidence and then apply it, and the inference step is where it fails. What the model extracts from three posts is not your voice, it is the genre those posts belong to.",[19,20,22],"h2",{"id":21},"what-the-model-actually-learns-from-your-examples","What the model actually learns from your examples",[12,24,25],{},"Give a model five LinkedIn posts and it will reliably pick up:",[27,28,29,38,44],"ul",{},[30,31,32,33,37],"li",{},"The ",[34,35,36],"strong",{},"format",": line breaks, list use, length, whether you open with a one-liner.",[30,39,32,40,43],{},[34,41,42],{},"register",": casual or formal, first person or third.",[30,45,32,46,49],{},[34,47,48],{},"topic domain",": SaaS, recruiting, design.",[12,51,52],{},"Those are the visible layer, and copying them produces something that looks like your posts from a distance. What it will not pick up from five examples:",[27,54,55,58,61,64],{},[30,56,57],{},"Which words you never use.",[30,59,60],{},"Which claims you would refuse to make.",[30,62,63],{},"What you find embarrassing.",[30,65,66],{},"The specific things you have actually done, which are the only source of material that could not have come from anyone else.",[12,68,69,70,75],{},"So the output lands in a predictable place: correct shape, generic substance. It reads like a competent stranger doing an impression, and readers detect it immediately even when they cannot name what is wrong. That gap is ",[71,72,74],"a",{"href":73},"\u002Fblog\u002Fwhy-ai-social-posts-get-ignored","why AI-written posts get ignored"," rather than disliked. Nothing about them is bad. There is just nobody in them.",[19,77,79],{"id":78},"a-voice-is-three-separate-things","A voice is three separate things",[12,81,82],{},"Treating \"voice\" as one setting is what makes it unfixable. It is at least three, and they need different inputs.",[12,84,85,88],{},[34,86,87],{},"Style."," Sentence length, punctuation habits, whether you use questions, how you open and close. This is the layer examples genuinely teach. It is also the least important, because it is the layer readers notice last.",[12,90,91,94],{},[34,92,93],{},"Vocabulary boundaries."," The words that are yours and the words that are disqualifying. Most people cannot list these from memory but recognise them instantly: \"leverage\", \"unlock\", \"in today's landscape\", \"game-changer\". A negative list works better than a positive one, and it is much easier to write.",[12,96,97,100],{},[34,98,99],{},"Substance."," The facts, numbers, opinions and stories you are allowed to draw on. This is the layer that makes a post identifiably yours, and no amount of style transfer produces it. If the model does not have your material, it will invent something plausible in your shape, which is the worst of both.",[12,102,103],{},"Almost everyone tries to solve a substance problem with style instructions. That is why the loop feels endless.",[19,105,107],{"id":106},"what-to-give-a-model-instead","What to give a model instead",[12,109,110],{},"If you want output you would actually publish, give it these four things and stop tuning adjectives.",[12,112,113,116],{},[34,114,115],{},"Ten to twenty of your own posts, not three."," Three is enough to average, not enough to characterise. And use ordinary posts, not only your best: your best posts are your most edited, which means they are the least like how you write.",[12,118,119,122,123,127],{},[34,120,121],{},"A short list of banned words and constructions."," Ten to fifteen entries is plenty. Include the ones that are technically fine but not you. This is the single highest-return input and takes fifteen minutes; if you want a starting list, the ",[71,124,126],{"href":125},"\u002Fblog\u002Fai-writing-tells","common AI tells"," are most of it.",[12,129,130,133],{},[34,131,132],{},"A file of facts you are allowed to cite."," Numbers with the date they were true, customer results with permission status, opinions you hold and would defend. This is boring to assemble and it is the entire difference between a post about your company and a post that could be about any company.",[12,135,136,139],{},[34,137,138],{},"Two or three examples of what you rejected and why."," \"Too salesy\", \"I would never say hustle\", \"this claims something I cannot prove.\" Rejections carry more signal per word than approvals, because approval only says the output was acceptable while a rejection names a boundary.",[19,141,143],{"id":142},"the-test-that-tells-you-whether-it-worked","The test that tells you whether it worked",[12,145,146],{},"Take the draft and remove your name, your company name and the topic. Show it to someone who knows you.",[12,148,149],{},"If they can tell it is yours, the voice landed. If they say \"this could be anyone in your industry\", it did not — regardless of how well it matched your line breaks.",[12,151,152],{},"A cheaper version you can run alone: read it out loud. The places you stumble or feel faintly embarrassed are almost always sentences you would not have written. Voice mismatches are much easier to hear than to see, which is why they survive so many silent re-reads.",[12,154,155,156,160],{},"For the mechanical layer, the constructions that mark text as machine-written regardless of whose voice it was aiming at, you can run a draft through our ",[71,157,159],{"href":158},"\u002Fai-check","AI writing check"," and see the specific spans flagged rather than a score.",[19,162,164],{"id":163},"why-a-chat-window-keeps-losing-your-voice","Why a chat window keeps losing your voice",[12,166,167],{},"Even when you get a good result in ChatGPT, it does not persist. Next week you open a new conversation, the model has none of it, and you re-explain from scratch or paste a prompt block you now maintain by hand.",[12,169,170,171,175],{},"This is the actual daily cost, and it is not about output quality at all. It is that the voice lives in a conversation rather than in the account, so every session starts at zero and every improvement has to be manually carried forward. Anyone who has ",[71,172,174],{"href":173},"\u002Fblog\u002Fchatgpt-claude-for-social-media","used ChatGPT or Claude for social content"," for more than a month has a document somewhere that exists only to re-teach the model things it already learned.",[12,177,178,179,182],{},"Making the voice a stored profile rather than a prompt is most of what ",[71,180,7],{"href":181},"\u002Ffor\u002Ffounders"," does: the samples, the banned list, the facts and the rejections accumulate in one place, and every draft starts from them instead of from a blank context. The interesting part is not that drafts get better. It is that a correction you make once stops being a correction you make weekly.",[19,184,186],{"id":185},"what-will-still-not-work","What will still not work",[12,188,189],{},"Two honest limits, because the promise gets oversold.",[12,191,192,195,196,200],{},[34,193,194],{},"Nothing gives you material."," If you have not done anything worth writing about this month, no voice profile produces a post worth reading. The model can shape what you have; it cannot supply it. This is the same reason ",[71,197,199],{"href":198},"\u002Fblog\u002Flinkedin-ghostwriter-cost","hiring a ghostwriter"," does not remove the founder from the loop.",[12,202,203,206],{},[34,204,205],{},"Voice is not a substitute for judgment."," A well-tuned model will happily write a sentence in your voice that you should not publish. The last read is still yours, and the useful version of that read is checking claims rather than admiring phrasing.",[19,208,210],{"id":209},"faq","FAQ",[12,212,213,216],{},[34,214,215],{},"Why doesn't ChatGPT sound like me even when I give it examples?","\nBecause from a handful of examples it extracts format and register, the visible layer, not your vocabulary boundaries or your material. You get the right shape with generic substance, which reads as an impression. Fix it by supplying more samples, an explicit banned-word list and a file of facts you can cite, rather than more adjectives.",[12,218,219,222],{},[34,220,221],{},"How many writing samples does an AI need to learn a voice?","\nTen to twenty is the useful range for social posts. Three averages rather than characterises. Include ordinary posts, not only your best ones — heavily edited work is the least representative of how you actually write.",[12,224,225,228],{},[34,226,227],{},"Is it better to describe my voice or show examples?","\nBoth, for different layers. Examples teach rhythm and structure, which people describe badly. Explicit rules work better for boundaries: what you never say, what you refuse to claim. Descriptions like \"professional but approachable\" carry almost no information and are the most common wasted instruction.",[12,230,231,234,235,239],{},[34,232,233],{},"Can AI-written posts be detected?","\nDetectors measure statistical patterns and are unreliable on short text, so a confident verdict on a 200-word post is worth little either way. Readers are the real detector, and they respond to the absence of specifics rather than to phrasing. See ",[71,236,238],{"href":237},"\u002Fblog\u002Fwhat-ai-detectors-catch","what AI detectors actually catch",".",[12,241,242,245],{},[34,243,244],{},"What is the fastest thing I can do to make AI drafts sound more like me?","\nWrite a list of ten words and phrases you would never use, and put it in the prompt. It takes fifteen minutes, needs no tooling, and removes most of what makes a draft read as machine-written.",{"title":247,"searchDepth":248,"depth":248,"links":249},"",2,[250,251,252,253,254,255,256],{"id":21,"depth":248,"text":22},{"id":78,"depth":248,"text":79},{"id":106,"depth":248,"text":107},{"id":142,"depth":248,"text":143},{"id":163,"depth":248,"text":164},{"id":185,"depth":248,"text":186},{"id":209,"depth":248,"text":210},"2026-08-25","Prompting an AI to sound like you almost never works, and the reason is mechanical. What a voice is actually made of, what you have to give a model, and how to tell whether it landed.",false,"md",null,{},true,"\u002Fblog\u002Fai-that-sounds-like-you",{"title":6,"description":258},"blog\u002Fai-that-sounds-like-you",[268,269,270],"founders","ai","writing","o_PBouvxI0s03S8G-pPaBBw4lqbGchRi3EAnBrrt11c",{"id":273,"title":274,"author":7,"body":275,"date":257,"description":497,"draft":259,"extension":260,"image":261,"meta":498,"navigation":263,"path":499,"seo":500,"stem":501,"tags":502,"updated":261,"__hash__":506},"blog\u002Fblog\u002Fhootsuite-alternatives-for-agencies.md","Choosing a Hootsuite alternative for an agency",{"type":9,"value":276,"toc":488},[277,280,283,287,290,296,302,308,314,317,321,324,330,336,342,353,363,369,373,376,382,388,394,400,403,407,410,413,417,424,431,435,438,453,456,458,464,470,476,482],[12,278,279],{},"Agencies almost never leave a big social suite because a feature is missing. They leave because the seat count grew faster than the value, because onboarding a junior takes a week, or because the tool solved publishing and left the actual bottleneck untouched.",[12,281,282],{},"That distinction matters when shopping, because the market is full of tools that are a cheaper version of the same shape. Replacing a dashboard with a dashboard changes your invoice and nothing else.",[19,284,286],{"id":285},"work-out-which-problem-you-are-actually-solving","Work out which problem you are actually solving",[12,288,289],{},"Before comparing anything, name the reason. There are four common ones and they point at different products.",[12,291,292,295],{},[34,293,294],{},"Cost per seat has outgrown the value."," You are paying for listening, ads and analytics modules that two people open occasionally. The answer is usually a smaller tool, not a different big one — and the risk is that you cut something a client depends on without realising which.",[12,297,298,301],{},[34,299,300],{},"The bottleneck is production, not publishing."," Posts get scheduled fine. The problem is getting twelve accounts drafted in the right voices and approved before they go stale. A scheduler with a nicer calendar will not touch this, and it is the most commonly misdiagnosed of the four.",[12,303,304,307],{},[34,305,306],{},"Approvals live outside the tool."," The suite technically has an approval feature, but clients will not log in to use it, so approvals happen in email anyway and the tool holds a fiction. Fixing this needs client-facing links, not internal workflow states.",[12,309,310,313],{},[34,311,312],{},"Per-brand separation is weak."," Voices bleed, the wrong account gets posted to, a junior can publish for a client they should not touch. This is a permissions and isolation question.",[12,315,316],{},"If you cannot say which of these is yours, a trial will not tell you — you will evaluate on the demo's terms.",[19,318,320],{"id":319},"six-things-worth-comparing","Six things worth comparing",[12,322,323],{},"Feature grids are near-useless here because every tool ticks every box at some tier. These are the questions that separate them in practice.",[12,325,326,329],{},[34,327,328],{},"1. What does the client see, and do they need an account?"," The single most predictive question. If review requires a login, most stakeholders never review, and your approval process silently reverts to email. Ask to see the exact page an unlogged-in client sees.",[12,331,332,335],{},[34,333,334],{},"2. Is brand context stored, or retyped?"," Where do a client's voice rules, permitted claims and forbidden topics live? If the answer is \"a document the team keeps open\", the tool is not managing brands, it is scheduling posts for several accounts. That is the difference that shows up at client five, not client two.",[12,337,338,341],{},[34,339,340],{},"3. What is a seat, and what does a junior's seat cost?"," Agencies grow by adding people who do a slice of the work. If every contributor needs a full seat, the pricing model actively punishes delegating. Ask what a draft-only role costs.",[12,343,344,347,348,352],{},[34,345,346],{},"4. Can it stop the wrong person publishing?"," Not \"does it have roles\" — every tool has roles. Whether a junior can be given draft rights on one client and nothing else, and whether that is enforced at the account level. ",[71,349,351],{"href":350},"\u002Fblog\u002Fagency-roles-for-social-content","Role separation"," is what makes juniors employable on client accounts at all.",[12,354,355,358,359,239],{},[34,356,357],{},"5. Where does the decision trail live?"," When a claim turns out to be wrong, can you show who drafted it, what it was based on and who approved it — in one place, in under a minute? Most suites can show a publish log and not much more. ",[71,360,362],{"href":361},"\u002Fblog\u002Fwho-is-responsible-for-a-wrong-claim","That gap is expensive exactly once",[12,364,365,368],{},[34,366,367],{},"6. How long does a new hire take to be useful in it?"," Big suites are broad, and breadth costs training. If you turn over freelancers, a week of ramp per person is a real line item that never appears in the comparison.",[19,370,372],{"id":371},"the-categories-you-are-choosing-between","The categories you are choosing between",[12,374,375],{},"Rather than a table of brands, it is more useful to know the four shapes on the market. Most tools are clearly one of them.",[12,377,378,381],{},[34,379,380],{},"Lightweight schedulers."," Queue, calendar, a few accounts, simple pricing. Excellent if publishing genuinely is your only problem. They tend to thin out fast on approvals and client separation, which is exactly where agencies feel pressure.",[12,383,384,387],{},[34,385,386],{},"Full suites."," The category Hootsuite defined: publishing, listening, ads, analytics, inbox, enterprise permissions. Correct when you need listening and paid in one place and have someone whose job includes owning the tool. The usual complaints are cost and ramp time, not capability.",[12,389,390,393],{},[34,391,392],{},"Writing-first tools."," Built around drafting, usually strongest on X and threads. Great for the writing experience, generally light on multi-client governance.",[12,395,396,399],{},[34,397,398],{},"Workflow tools."," Built around what happens between an idea and an approved post: brand context, drafting in each voice, client approval, the trail. Narrower than a suite by design — the trade is that they do not replace listening or ads management.",[12,401,402],{},"Knowing which shape you are shopping for prevents the most common outcome, which is buying a lightweight scheduler to escape suite pricing and rebuilding approvals in email within a month.",[19,404,406],{"id":405},"the-trap-cheaper-same-shape","The trap: cheaper, same shape",[12,408,409],{},"The switch that disappoints is the one where an agency moves from a big suite to a mid-size suite. Six weeks later the calendar looks different, the invoice is smaller, and every real problem is intact: context still lives in documents, clients still approve by email, juniors still need supervision because permissions are coarse.",[12,411,412],{},"The test for whether a candidate actually changes anything: name the three things that cost you the most hours last month, and ask specifically how the tool changes each one. If two of the three answers are \"you would do that the same way\", it is a price negotiation, not a migration.",[19,414,416],{"id":415},"where-selfsm-sits","Where SelfSM sits",[12,418,419,420,423],{},"Worth being direct, since this is our blog. ",[71,421,7],{"href":422},"\u002Ffor\u002Fagencies"," is the fourth category — a workflow tool, not a suite. It holds per-brand voice and facts, drafts in each client's voice, sends approvals as links the client opens without an account, and keeps the drafting and approval trail on the post.",[12,425,426,427,239],{},"It does not do social listening or ads management. If those are why you are on a suite, a workflow tool is an addition, not a replacement, and anyone telling you otherwise is selling. The tool-by-tool version of this is in ",[71,428,430],{"href":429},"\u002Fblog\u002Fselfsm-vs-buffer-hootsuite-typefully","SelfSM vs Buffer vs Hootsuite vs Typefully",[19,432,434],{"id":433},"how-to-run-the-trial","How to run the trial",[12,436,437],{},"Two weeks, one real client, not a sandbox. Specifically:",[439,440,441,444,447,450],"ol",{},[30,442,443],{},"Onboard one live account end to end, including the client actually approving something.",[30,445,446],{},"Have someone who did not choose the tool draft in it, unsupervised, on day three.",[30,448,449],{},"Produce one client report from it.",[30,451,452],{},"Count the things you still did in email, Slack or a doc.",[12,454,455],{},"That last count is the result. A tool that removes two of five side-channels is worth switching for. One that removes none is a new calendar.",[19,457,210],{"id":209},[12,459,460,463],{},[34,461,462],{},"Why do agencies leave Hootsuite?","\nUsually cost per seat as the team grows, ramp time for new hires, and the discovery that the bottleneck was production and approvals rather than publishing. Feature gaps are rarely the trigger — full suites are broad by design.",[12,465,466,469],{},[34,467,468],{},"What should an agency compare when choosing a social media tool?","\nWhat the client sees and whether they need an account, whether brand context is stored or retyped, what a limited seat costs, whether permissions can restrict a junior to one client, where the decision trail lives, and how long a new hire takes to be productive. Feature grids do not separate tools; these questions do.",[12,471,472,475],{},[34,473,474],{},"Is a cheaper scheduler a good Hootsuite replacement for an agency?","\nOnly if publishing was genuinely the problem. Agencies that switch to a lightweight scheduler to cut cost commonly rebuild approvals in email within a month, because schedulers are thin exactly where multi-client work is heavy.",[12,477,478,481],{},[34,479,480],{},"Do we need social listening if we move off a full suite?","\nDepends who uses it. If a named person runs listening weekly and reports on it, keep it — either as a suite or a dedicated tool. If it is opened occasionally and nothing follows from it, it is a module you are funding out of habit.",[12,483,484,487],{},[34,485,486],{},"How long should a tool trial run?","\nTwo weeks with one real client, including a live client approval and one report. Sandbox trials test the interface; only a real account tests the workflow, and the workflow is what you are buying.",{"title":247,"searchDepth":248,"depth":248,"links":489},[490,491,492,493,494,495,496],{"id":285,"depth":248,"text":286},{"id":319,"depth":248,"text":320},{"id":371,"depth":248,"text":372},{"id":405,"depth":248,"text":406},{"id":415,"depth":248,"text":416},{"id":433,"depth":248,"text":434},{"id":209,"depth":248,"text":210},"Agencies rarely leave Hootsuite over features. Here is what actually drives the switch, the six criteria worth comparing, and the trap of replacing one dashboard with a slightly cheaper dashboard.",{},"\u002Fblog\u002Fhootsuite-alternatives-for-agencies",{"title":274,"description":497},"blog\u002Fhootsuite-alternatives-for-agencies",[503,504,505],"comparison","alternatives","agencies","71RvBUOy4VgVA84eLcwl1AtK4J2a2sycccGDhfk_jws",{"id":508,"title":509,"author":7,"body":510,"date":257,"description":799,"draft":259,"extension":260,"image":261,"meta":800,"navigation":263,"path":801,"seo":802,"stem":803,"tags":804,"updated":261,"__hash__":807},"blog\u002Fblog\u002Fhow-many-clients-can-one-social-media-manager-handle.md","How many clients can one social media manager actually handle?",{"type":9,"value":511,"toc":791},[512,515,523,527,530,598,601,604,608,611,617,623,629,632,636,639,650,661,667,673,681,685,688,699,705,711,722,728,735,739,742,756,759,761,767,773,779,785],[12,513,514],{},"Ask ten freelancers and you get answers between three and fifteen. Both ends are true, which is why the question is usually answered badly. The number depends almost entirely on things that have nothing to do with how fast you write.",[12,516,517,518,522],{},"The useful version of the question is not \"how many clients\" but \"how many ",[519,520,521],"em",{},"contexts"," can one person hold in a week without quality collapsing.\" That number is smaller, and it is the one that actually breaks.",[19,524,526],{"id":525},"where-the-hours-really-go","Where the hours really go",[12,528,529],{},"Take a typical mid-size client — twelve posts a month across two channels, one strategy call. The visible work is writing. Track it for a month and it comes out closer to this:",[531,532,533,546],"table",{},[534,535,536],"thead",{},[537,538,539,543],"tr",{},[540,541,542],"th",{},"Activity",[540,544,545],{},"Share of time",[547,548,549,558,566,574,582,590],"tbody",{},[537,550,551,555],{},[552,553,554],"td",{},"Writing and editing",[552,556,557],{},"25–35%",[537,559,560,563],{},[552,561,562],{},"Re-loading context (what did we agree, what's live, what's their tone)",[552,564,565],{},"15–25%",[537,567,568,571],{},[552,569,570],{},"Chasing and processing approvals",[552,572,573],{},"15–20%",[537,575,576,579],{},[552,577,578],{},"Reporting and account admin",[552,580,581],{},"10–15%",[537,583,584,587],{},[552,585,586],{},"Client comms outside scheduled calls",[552,588,589],{},"10–20%",[537,591,592,595],{},[552,593,594],{},"Scheduling and publishing",[552,596,597],{},"5–10%",[12,599,600],{},"Writing is a quarter of the job. Everything else is overhead, and overhead is what scales badly — because most of it repeats per client rather than per post.",[12,602,603],{},"This is why \"I can write a post in twenty minutes\" is a misleading input for capacity planning. Twelve posts is four hours. The client is not a four-hour client.",[19,605,607],{"id":606},"the-real-ceiling","The real ceiling",[12,609,610],{},"For most people running client work solo, the honest bands look like this:",[12,612,613,616],{},[34,614,615],{},"Three to four clients"," is comfortable, with room for the account that suddenly needs more. Each one gets real thought. This is the band where the work stays good and the money is often disappointing.",[12,618,619,622],{},[34,620,621],{},"Five to seven clients"," is the common working maximum for an experienced freelancer with a repeatable process. Achievable, but only if approvals are batched and context is written down rather than remembered. Without both, this band is where people burn out.",[12,624,625,628],{},[34,626,627],{},"Eight or more"," does not fail on writing capacity. It fails on context-switching and on being the single point of failure for eight relationships. Past this point you are running a small agency whether you call it that or not, and the answer is delegation, not efficiency.",[12,630,631],{},"Anyone selling you a workflow that gets you to fifteen is describing a volume business with templated output. That is a legitimate model. It is not the same job, and it does not command the same rate.",[19,633,635],{"id":634},"the-four-multipliers","The four multipliers",[12,637,638],{},"Two managers with the same number of clients can have wildly different loads. These are the variables that matter more than headcount.",[12,640,641,644,645,649],{},[34,642,643],{},"Approval latency."," A client who approves in a day costs a fraction of one who approves in a week. Slow approvals do not just delay work, they force you to hold the context open — you cannot close the account in your head until it ships. Two slow clients can cost more than four fast ones. This is the single biggest lever, and it is why ",[71,646,648],{"href":647},"\u002Fblog\u002Fsocial-media-approval-process","a designed approval process"," is a capacity intervention, not an admin one.",[12,651,652,655,656,660],{},[34,653,654],{},"Voice distance."," Three B2B SaaS clients are much cheaper to run together than a SaaS, a restaurant and a law firm. It is not the writing, it is the reset — ",[71,657,659],{"href":658},"\u002Fblog\u002Fmanaging-multiple-brand-voices","switching voices"," mid-day carries a real cost and the mistakes it causes are the embarrassing kind.",[12,662,663,666],{},[34,664,665],{},"Fact volatility."," A client whose numbers, pricing and product change monthly needs constant re-verification. A stable client's facts stay true. The volatile one can be twice the work at the same post count.",[12,668,669,672],{},[34,670,671],{},"Number of stakeholders."," One approver is a client. Three approvers who disagree is a committee, and a committee is two clients' worth of work regardless of how many posts it produces.",[12,674,675,676,680],{},"Price against these, not against post count. A five-post client with four stakeholders and weekly pricing changes is not cheaper than a twelve-post client with one decisive approver, and ",[71,677,679],{"href":678},"\u002Fblog\u002Fpricing-social-media-management","pricing by output"," is how people end up resenting their best-paying account.",[19,682,684],{"id":683},"raising-the-number-without-working-more","Raising the number without working more",[12,686,687],{},"Every real gain comes from removing repeat work, not from writing faster.",[12,689,690,693,694,698],{},[34,691,692],{},"Write the context down, once, per client."," Voice rules, permitted claims, forbidden topics, approver, turnaround, publishing days. The reason this raises capacity is not tidiness — it is that re-deriving context from memory is the largest hidden cost in the list above, and it recurs every single session. ",[71,695,697],{"href":696},"\u002Fblog\u002Fsocial-media-client-onboarding","Client onboarding"," is where this gets captured cheaply; after that it gets captured expensively.",[12,700,701,704],{},[34,702,703],{},"Batch by activity, not by client."," All drafting Monday, all approvals sent Tuesday, all scheduling Thursday. Switching costs are paid per switch, so five clients drafted in one block cost far less than five clients handled end-to-end one at a time.",[12,706,707,710],{},[34,708,709],{},"Make approval a fixed weekly slot per client."," Same day, same time, same batch size. It converts your most unpredictable dependency into a scheduled one.",[12,712,713,716,717,721],{},[34,714,715],{},"Standardise reporting."," Same format, same metrics, generated rather than assembled. Reporting is the most compressible line in the table because it is the least client-specific. ",[71,718,720],{"href":719},"\u002Fblog\u002Fclient-reporting-that-gets-read","What clients actually read"," is shorter than what most agencies send.",[12,723,724,727],{},[34,725,726],{},"Kill the client that eats a day."," Most people running five accounts have one that costs as much as the other four. Repricing it or ending it is usually worth more than any tooling change.",[12,729,730,731,734],{},"Tooling helps precisely where it removes repetition: brand context that persists per client, approval links that carry their own state, reports that build themselves. That is the part of the load ",[71,732,7],{"href":733},"\u002Ffor\u002Fsocial-media-managers"," is aimed at — not writing faster, but not re-loading the same five things every Monday.",[19,736,738],{"id":737},"a-number-you-can-actually-use","A number you can actually use",[12,740,741],{},"If you want a rough capacity estimate for yourself, budget per client per month:",[27,743,744,747,750,753],{},[30,745,746],{},"4–6 hours drafting and editing",[30,748,749],{},"2–3 hours approvals and chasing",[30,751,752],{},"1–2 hours reporting",[30,754,755],{},"2–4 hours comms and unplanned requests",[12,757,758],{},"That is 9–15 hours a month for a mid-size client, or roughly 2–4 hours a week. Against 25 genuinely productive client hours a week — not 40, since selling, admin and your own business exist — you land at five to seven accounts. Which is exactly where experienced freelancers report the ceiling to be, arrived at from the other direction.",[19,760,210],{"id":209},[12,762,763,766],{},[34,764,765],{},"How many clients can a freelance social media manager handle?","\nFive to seven is the realistic maximum for an experienced solo manager with batched approvals and written per-client context. Three to four is comfortable. Beyond eight the constraint stops being time and becomes context-switching plus being the only point of failure for eight relationships.",[12,768,769,772],{},[34,770,771],{},"How many hours does one social media client take per month?","\nNine to fifteen hours for a mid-size client at around twelve posts a month, of which writing is roughly a quarter. The rest is approvals, reporting, re-loading context and unscheduled communication.",[12,774,775,778],{},[34,776,777],{},"Why does adding one more client feel like adding two?","\nBecause most of the cost is per client, not per post: a new account adds a whole context, a new approver, a new set of facts and a new relationship. Post-count-based capacity planning consistently underestimates this, which is how people end up over-committed at four clients.",[12,780,781,784],{},[34,782,783],{},"Should I charge per post or per month?","\nMonthly retainers reflect the work better, because post count is a poor proxy for load. Two clients at the same post count can differ twofold based on approval speed, stakeholder count and how often their facts change. Price those, not volume.",[12,786,787,790],{},[34,788,789],{},"What is the fastest way to take on one more client without more hours?","\nBatch by activity rather than by client, and move each client's approvals to a fixed weekly slot. Together those remove most switching cost and most chasing, which are the two largest compressible items in the week.",{"title":247,"searchDepth":248,"depth":248,"links":792},[793,794,795,796,797,798],{"id":525,"depth":248,"text":526},{"id":606,"depth":248,"text":607},{"id":634,"depth":248,"text":635},{"id":683,"depth":248,"text":684},{"id":737,"depth":248,"text":738},{"id":209,"depth":248,"text":210},"The honest ceiling is lower than the job ads suggest, and writing hours are not what caps it. A breakdown of where the time really goes, the four multipliers, and how to raise the number without working more.",{},"\u002Fblog\u002Fhow-many-clients-can-one-social-media-manager-handle",{"title":509,"description":799},"blog\u002Fhow-many-clients-can-one-social-media-manager-handle",[805,806],"workflow","strategy","CPdLCbrPFfyT7JkmU389oHRg2tjfqyJ0RSHAajJIJSQ",{"id":809,"title":810,"author":7,"body":811,"date":257,"description":1061,"draft":259,"extension":260,"image":261,"meta":1062,"navigation":263,"path":647,"seo":1063,"stem":1064,"tags":1065,"updated":261,"__hash__":1066},"blog\u002Fblog\u002Fsocial-media-approval-process.md","A social media approval process clients actually follow",{"type":9,"value":812,"toc":1050},[813,816,819,823,826,832,838,844,850,853,857,860,868,871,875,878,881,887,890,894,897,900,920,927,931,934,937,940,944,947,953,959,962,966,969,972,979,985,989,992,1009,1012,1014,1020,1026,1032,1038,1044],[12,814,815],{},"Every agency has the same story. The posts were ready Tuesday. The client approved them the following Monday, by which point two were stale and one referenced a webinar that had already happened. Nobody was negligent. The process just had no mechanism for making Tuesday matter.",[12,817,818],{},"Approval gets treated as a courtesy step bolted onto the end of production. It is actually the most fragile part of the operation, because it is the only step that depends on someone who does not work for you and has no deadline.",[19,820,822],{"id":821},"the-four-ways-approval-breaks","The four ways approval breaks",[12,824,825],{},"Almost every stall is one of these. They have different fixes, and treating all of them as \"the client is slow\" fixes none.",[12,827,828,831],{},[34,829,830],{},"The client does not know they are the blocker."," They received a link, filed it mentally under \"read later\", and have no sense that anything is waiting. Nothing in the message said what happens if they do nothing.",[12,833,834,837],{},[34,835,836],{},"The client does not know what \"approved\" means."," Are they checking the claim, the tone, the timing, or all three? Is silence consent? Can they edit? An unclear ask gets deferred, because deferring is cheaper than guessing wrong.",[12,839,840,843],{},[34,841,842],{},"Approval is spread across three surfaces."," The draft is in a doc, the comments are in Slack, the sign-off is in email. Nobody can tell what state anything is in without reconstructing it from three histories, so everyone waits for someone else to do the reconstructing.",[12,845,846,849],{},[34,847,848],{},"The wrong person is approving."," A marketing coordinator with no authority over claims is asked to approve a post making a claim. They cannot, so they escalate, and the escalation has no deadline either.",[12,851,852],{},"Only the first of those looks like slowness. The other three look like slowness and are design faults.",[19,854,856],{"id":855},"rule-one-one-surface-one-state","Rule one: one surface, one state",[12,858,859],{},"The highest-leverage change is not a reminder cadence. It is putting the draft, the comments and the decision in the same place, so that \"what is the status of Thursday's post\" is something you look at rather than something you ask.",[12,861,862,863,867],{},"That is the real argument for approval links over email attachments. Not convenience — ",[71,864,866],{"href":865},"\u002Fblog\u002Fcontent-approval-workflow","state",". A post that carries its own status can be counted, filtered and chased. A post whose status lives in someone's inbox cannot.",[12,869,870],{},"If you take one thing from this: stop asking clients to reply in order to approve. A reply creates a second surface immediately.",[19,872,874],{"id":873},"rule-two-name-the-decision-narrow-the-question","Rule two: name the decision, narrow the question",[12,876,877],{},"\"Let me know if you have any thoughts\" is not a request for approval. It is an invitation to an open-ended review, and open-ended reviews are what produce the eleven-comment thread about a comma.",[12,879,880],{},"Ask for a specific decision on a specific scope:",[882,883,884],"blockquote",{},[12,885,886],{},"Three posts for next week. You are checking two things: are the facts right, and is anything commercially sensitive. Tone and structure are ours. Approve, or tell us which one to change.",[12,888,889],{},"That does three jobs. It bounds the review, it tells the client what is not their problem, and it gives them a verb. Clients given a narrow question answer it in minutes. Clients given \"thoughts?\" answer it never.",[19,891,893],{"id":892},"rule-three-a-default-for-silence","Rule three: a default for silence",[12,895,896],{},"Every approval step needs a documented answer to \"what happens if nobody responds by Thursday.\"",[12,898,899],{},"There are three honest defaults, and each is defensible:",[27,901,902,908,914],{},[30,903,904,907],{},[34,905,906],{},"Hold."," Nothing publishes without explicit approval. Correct for regulated clients, claims-heavy accounts and anything legal has touched. Accept that output will be lumpy.",[30,909,910,913],{},[34,911,912],{},"Publish on silence after N days."," Written into the contract, not assumed. Works when the client has said outright that they trust the editorial line and would rather have volume than veto.",[30,915,916,919],{},[34,917,918],{},"Publish a safe subset."," Evergreen posts go out on silence; anything with a number, a customer name or a comparison holds. This is the option most agencies want and few write down.",[12,921,922,923,926],{},"Which one you pick matters less than that the client agreed to it during ",[71,924,925],{"href":696},"onboarding",", in writing, before the first stall. A default invented during a stall is a negotiation, and negotiations take longer than the approval would have.",[19,928,930],{"id":929},"rule-four-approve-in-batches-on-a-rhythm","Rule four: approve in batches, on a rhythm",[12,932,933],{},"Ad-hoc requests arrive at random and get treated as interruptions. A standing slot gets treated as work.",[12,935,936],{},"One batch a week — same day, same time, same size — is far more reliable than seven posts arriving individually as they are finished. The client learns that Wednesday morning means twenty minutes of approving. That is a habit. Individual pings are noise, and noise gets muted.",[12,938,939],{},"Batching also makes capacity legible. Nine posts a week approved on Wednesday is a number both sides can plan around. Nine posts approved on scattered days is a permanent low-grade negotiation.",[19,941,943],{"id":942},"rule-five-separate-changes-requested-from-rejected","Rule five: separate \"changes requested\" from \"rejected\"",[12,945,946],{},"Most tools and most agencies collapse these, and they are entirely different events.",[12,948,949,952],{},[34,950,951],{},"Changes requested"," means the post is fine in principle and one thing needs fixing. It should come back with the specific fix, and it should not restart the review from zero.",[12,954,955,958],{},[34,956,957],{},"Rejected"," means the angle is wrong and the post should not exist. That is feedback about strategy, not about a draft, and it belongs in a different conversation than the one about Thursday.",[12,960,961],{},"When both are the same button, every piece of feedback reads as a rejection, writers get defensive, and clients start softening real objections into vague ones to avoid the friction. Two words on one screen prevent a surprising amount of relationship damage.",[19,963,965],{"id":964},"what-the-trail-is-for","What the trail is for",[12,967,968],{},"A recorded approval trail is usually sold as accountability, which makes it sound like a device for blaming clients. It is more useful than that.",[12,970,971],{},"It answers, without anyone reconstructing anything: which posts are waiting, how long each has waited, who was asked, whether they were the right person to ask, and what changed between the draft that was rejected and the one that shipped. That last one is how a writer learns a client's boundaries in six weeks instead of six months.",[12,973,974,975,978],{},"It also matters on the day something goes wrong. When a claim turns out to be false, the difference between an agency that survives it and one that does not is usually whether they can show ",[71,976,977],{"href":361},"who approved what"," without a forensic exercise across three tools.",[12,980,981,982,984],{},"That is what ",[71,983,7],{"href":422}," is built around: the draft, the approval link, the decision and the reason all sit on the post itself, and the client sees a page rather than a login. Not a novel idea — just rarely all in one place, which is exactly why approvals stall.",[19,986,988],{"id":987},"the-version-to-steal","The version to steal",[12,990,991],{},"If you change one thing this week:",[439,993,994,997,1000,1003,1006],{},[30,995,996],{},"Send one batch, one day a week, at the same time.",[30,998,999],{},"In the message, name what the client is checking and what they are not.",[30,1001,1002],{},"Include the deadline and what happens after it.",[30,1004,1005],{},"Give them a link that carries the decision, not a document that needs a reply.",[30,1007,1008],{},"Keep \"needs a change\" and \"kill this\" as separate answers.",[12,1010,1011],{},"None of that requires new software. It requires deciding that approval is a designed step rather than the bit that happens after the real work.",[19,1013,210],{"id":209},[12,1015,1016,1019],{},[34,1017,1018],{},"What is a social media approval process?","\nThe defined sequence a post moves through between being drafted and being published: who reviews it, what exactly they are checking, how they record the decision, what happens on silence, and where the record lives. Without those five things written down it is not a process, it is a habit that works until someone is on holiday.",[12,1021,1022,1025],{},[34,1023,1024],{},"How long should a client have to approve social media posts?","\nTwo to three working days for a normal batch is realistic for most clients, and worth agreeing in the contract rather than discovering. Shorter windows work only where a named person owns approvals as part of their job. Whatever the window, the default action when it expires matters more than its length.",[12,1027,1028,1031],{},[34,1029,1030],{},"Should social media posts be approved individually or in batches?","\nBatches, on a fixed weekly slot. Individual requests arrive as interruptions and get deferred; a standing slot becomes a habit. Batching also makes capacity and turnaround measurable instead of anecdotal.",[12,1033,1034,1037],{},[34,1035,1036],{},"Who should approve social media content at a client company?","\nWhoever has authority over the claims being made — usually a marketing lead, with legal involved only for regulated claims. The common failure is routing approval to a coordinator who cannot approve claims, which turns every post into an escalation with no deadline.",[12,1039,1040,1043],{},[34,1041,1042],{},"What should happen if a client never responds to an approval request?","\nWhatever you agreed in advance: hold everything, publish after a stated number of days, or publish only the posts carrying no claims, numbers or customer names. All three are defensible. Deciding during the stall is what turns a delay into a conflict.",[12,1045,1046,1049],{},[34,1047,1048],{},"Is email approval good enough for social media content?","\nIt works for one client and one approver, and stops working at three. Email splits the draft, the discussion and the decision across separate threads, so status becomes something you reconstruct rather than read. The failure is not email itself, it is that nothing carries state.",{"title":247,"searchDepth":248,"depth":248,"links":1051},[1052,1053,1054,1055,1056,1057,1058,1059,1060],{"id":821,"depth":248,"text":822},{"id":855,"depth":248,"text":856},{"id":873,"depth":248,"text":874},{"id":892,"depth":248,"text":893},{"id":929,"depth":248,"text":930},{"id":942,"depth":248,"text":943},{"id":964,"depth":248,"text":965},{"id":987,"depth":248,"text":988},{"id":209,"depth":248,"text":210},"Why approval stalls have almost nothing to do with lazy clients, the four failure modes that cause them, and how to design a process that survives contact with a busy person.",{},{"title":810,"description":1061},"blog\u002Fsocial-media-approval-process",[805,505],"lJfhO0TxD4AGVai-dFIUrqeXxyjRclGDz3967R7FjfE",{"id":1068,"title":1069,"author":7,"body":1070,"date":257,"description":1287,"draft":259,"extension":260,"image":261,"meta":1288,"navigation":263,"path":1289,"seo":1290,"stem":1291,"tags":1292,"updated":261,"__hash__":1293},"blog\u002Fblog\u002Fwhite-label-social-media-reporting.md","White-label social media reporting that survives a renewal conversation",{"type":9,"value":1071,"toc":1278},[1072,1075,1078,1082,1085,1091,1097,1103,1106,1110,1113,1119,1129,1135,1141,1147,1153,1156,1160,1166,1172,1178,1184,1188,1191,1197,1203,1209,1219,1223,1226,1229,1233,1236,1239,1246,1248,1254,1260,1266,1272],[12,1073,1074],{},"\"White label\" in most reporting tools means one thing: your logo instead of theirs, and a custom domain if you pay more. That is branding, not reporting, and it does not change the outcome that matters. The client still opens the PDF, scrolls to nothing in particular, and closes it.",[12,1076,1077],{},"A report earns its cost in one moment: the conversation about whether to keep paying you. Everything else it does is secondary. Most agency reports are not built for that moment at all.",[19,1079,1081],{"id":1080},"what-the-client-is-actually-checking","What the client is actually checking",[12,1083,1084],{},"When a client opens your report, they are answering one of three questions, and it is almost never the one the report is designed around.",[12,1086,1087,1090],{},[34,1088,1089],{},"\"Is this working?\""," Meaning: is anything happening that would not have happened otherwise. Not whether impressions went up.",[12,1092,1093,1096],{},[34,1094,1095],{},"\"Am I exposed?\""," Did anything go out that shouldn't have, is anything unanswered, is there a comment from an angry customer sitting there since Thursday.",[12,1098,1099,1102],{},[34,1100,1101],{},"\"What do I have to do?\""," Is there a decision waiting on me, and did my not-deciding cost anything.",[12,1104,1105],{},"A dashboard of platform metrics answers none of these. It answers \"what did the numbers do\", which the client did not ask, and which they cannot act on. That is why engagement charts get skimmed and the follower count gets a glance. It is the only number on the page that feels like a score.",[19,1107,1109],{"id":1108},"what-belongs-in-the-report","What belongs in the report",[12,1111,1112],{},"Order matters more than content here, because most clients read the first screen and stop.",[12,1114,1115,1118],{},[34,1116,1117],{},"One paragraph, in words, at the top."," What happened, what it means, what you are doing next. Written for someone who will read only this. If you cannot write it, the month did not have a story and the report will not invent one.",[12,1120,1121,1124,1125,1128],{},[34,1122,1123],{},"What shipped, and what didn't."," Posts published against posts planned, with the reason for any gap. Naming a gap you caused builds more trust than hiding it; naming a gap the client caused — four posts held eleven days waiting for approval — is the single most useful line in the whole document, and it belongs there rather than in a tense call. You only have that line if ",[71,1126,1127],{"href":647},"approvals leave a record","; otherwise it is your word against their memory.",[12,1130,1131,1134],{},[34,1132,1133],{},"Decisions made and their outcome."," \"We moved to Tuesday\u002FThursday, engagement per post rose, volume held.\" This is the only section that demonstrates judgment, and judgment is what a retainer buys.",[12,1136,1137,1140],{},[34,1138,1139],{},"Numbers, with a comparison and a denominator."," A number alone is decoration. Against last month, against the previous quarter, or against the account's own baseline, it becomes information. Three or four metrics that connect to the client's actual goal beat twenty that do not.",[12,1142,1143,1146],{},[34,1144,1145],{},"What needs the client."," Approvals waiting, facts to confirm, decisions pending. Explicit, with dates. This section does more for next month's turnaround than any reminder email.",[12,1148,1149,1152],{},[34,1150,1151],{},"Next month's plan, in one list."," Short. It is a commitment, so it should be small enough to keep.",[12,1154,1155],{},"That is a two-page document. Two pages that get read beat twelve that get filed.",[19,1157,1159],{"id":1158},"the-parts-to-cut","The parts to cut",[12,1161,1162,1165],{},[34,1163,1164],{},"Vanity screenshots."," A grid of your best-performing posts is a portfolio, not a report.",[12,1167,1168,1171],{},[34,1169,1170],{},"Platform metrics with no line to the goal."," Impressions matter if the goal is reach. If the goal is inbound leads, impressions are context at best and misdirection at worst.",[12,1173,1174,1177],{},[34,1175,1176],{},"Anything the client cannot act on."," \"Best time to post: 9am\" is either already implemented, in which case say so, or it is an admission you haven't.",[12,1179,1180,1183],{},[34,1181,1182],{},"Percentages without absolutes."," \"Engagement up 240%\" on a base of five interactions is a sentence that damages your credibility the moment someone checks.",[19,1185,1187],{"id":1186},"where-white-label-actually-matters","Where white-label actually matters",[12,1189,1190],{},"There are three places where the label genuinely changes the outcome, and none of them is the logo.",[12,1192,1193,1196],{},[34,1194,1195],{},"The client should not need an account."," The moment a report requires a login, half of the stakeholders never see it. A link that opens to a page is the difference between a report the CMO reads on their phone and a report only your day-to-day contact sees.",[12,1198,1199,1202],{},[34,1200,1201],{},"The data should be yours, not the platform's story."," Reporting tools present the metrics they can collect most easily, which shapes what your report argues. If a channel has no usable API, and several do not, an honest report says the number was read manually or is unavailable, rather than quietly omitting the channel.",[12,1204,1205,1208],{},[34,1206,1207],{},"Your process should be visible in it."," A white-labelled dashboard makes you interchangeable with any agency using the same dashboard. A report that shows what was decided and why makes the case that the thinking came from you. That is the actual anti-churn mechanism, and it is invisible in a logo swap.",[12,1210,1211,1212,1214,1215,1218],{},"This is roughly the shape ",[71,1213,7],{"href":422}," generates: a client-facing link with what shipped, what is waiting on them and what was decided, on the agency's own framing rather than a platform's default dashboard. The reason it looks unlike a typical reporting tool is that it is built from the approval trail rather than from platform analytics, which also means it knows ",[71,1216,1217],{"href":350},"who did what"," without anyone logging it separately.",[19,1220,1222],{"id":1221},"cadence","Cadence",[12,1224,1225],{},"Monthly is right for most retainers. Weekly reports are usually a symptom. The client does not trust that work is happening, and the fix for that is visibility into the pipeline, not more documents. If a client wants weekly, give them a live view of what is drafted, approved and scheduled and keep the report monthly.",[12,1227,1228],{},"Send it within three working days of month end, always on the same day. A report that arrives on a predictable date gets read. One that arrives when you get to it gets skimmed, because the client has already stopped expecting it.",[19,1230,1232],{"id":1231},"the-renewal-test","The renewal test",[12,1234,1235],{},"Before sending, read your report and ask: if this client were deciding today whether to renew, does this document help them say yes?",[12,1237,1238],{},"If the honest answer is \"it shows we were busy\", rewrite the top paragraph. Busy is not the case for renewal. A decision you made, its result, and the thing you are doing next — that is.",[12,1240,1241,1242,1245],{},"For the writing side of the same problem, ",[71,1243,1244],{"href":719},"what clients actually read"," covers the report itself. This piece is about the version with your name on it.",[19,1247,210],{"id":209},[12,1249,1250,1253],{},[34,1251,1252],{},"What does white-label social media reporting mean?","\nStrictly, a report delivered under the agency's brand rather than the tool's: logo, colours, sometimes a custom domain. In practice the label is the least important part: what determines whether the report works is whether the client can open it without an account and whether it answers the renewal question.",[12,1255,1256,1259],{},[34,1257,1258],{},"What should be in an agency social media report?","\nA one-paragraph summary, what shipped against what was planned with reasons for gaps, decisions made and their outcomes, three or four metrics with comparisons, an explicit list of what needs the client, and next month's plan. Two pages is usually enough.",[12,1261,1262,1265],{},[34,1263,1264],{},"How often should agencies send client reports?","\nMonthly, within three working days of month end, on a consistent date. Requests for weekly reports normally mean the client lacks visibility into work in progress; solve that with a live view of the pipeline rather than by sending four times as many documents.",[12,1267,1268,1271],{},[34,1269,1270],{},"Should client reports include failures?","\nYes, and specifically the gaps the client caused: posts held waiting on approval, facts never confirmed. Reported neutrally with dates, it is the section that most improves next month's turnaround, and hiding it means the same delay recurs while you absorb the blame for lower output.",[12,1273,1274,1277],{},[34,1275,1276],{},"Do clients need a login to view a white-label report?","\nThey should not. Any login requirement drops the number of stakeholders who ever see the report, usually to one. A tokenised link that opens directly to a page is what gets a report read by the person who decides on renewal.",{"title":247,"searchDepth":248,"depth":248,"links":1279},[1280,1281,1282,1283,1284,1285,1286],{"id":1080,"depth":248,"text":1081},{"id":1108,"depth":248,"text":1109},{"id":1158,"depth":248,"text":1159},{"id":1186,"depth":248,"text":1187},{"id":1221,"depth":248,"text":1222},{"id":1231,"depth":248,"text":1232},{"id":209,"depth":248,"text":210},"Most white-label reporting is a logo swap on a dashboard nobody reads. What clients actually check, what belongs in the report, and how to build one that answers the renewal question before it is asked.",{},"\u002Fblog\u002Fwhite-label-social-media-reporting",{"title":1069,"description":1287},"blog\u002Fwhite-label-social-media-reporting",[505,805],"EzYekyLUGzuj3J4pMiL3BJn_c0Us80YrPhH2-0oScyo",{"id":1295,"title":1296,"author":7,"body":1297,"date":1567,"description":1568,"draft":259,"extension":260,"image":261,"meta":1569,"navigation":263,"path":350,"seo":1570,"stem":1571,"tags":1572,"updated":261,"__hash__":1573},"blog\u002Fblog\u002Fagency-roles-for-social-content.md","Agency roles for social content: who drafts, who approves, who releases",{"type":9,"value":1298,"toc":1558},[1299,1302,1309,1313,1316,1322,1328,1334,1340,1347,1351,1354,1360,1366,1372,1378,1382,1385,1446,1453,1457,1460,1463,1466,1480,1484,1490,1496,1502,1508,1512,1515,1526,1528,1534,1540,1546,1552],[12,1300,1301],{},"Small agencies run on one person per account. That person writes, decides, sends to the client and publishes, and it works beautifully right up to the point where they are on holiday, or leave, or make one mistake at scale.",[12,1303,1304,1305,1308],{},"The fix is not more process. It is separating four things that are usually one thing: ",[34,1306,1307],{},"drafting, internal approval, client release, and publishing",". Once they are separate permissions rather than stages in one person's head, juniors can do real work, seniors stop being a bottleneck on every post, and no single mistake reaches a client feed unaccompanied.",[19,1310,1312],{"id":1311},"the-four-permissions","The four permissions",[12,1314,1315],{},"They are worth naming precisely because most tools and most agencies collapse them into two.",[12,1317,1318,1321],{},[34,1319,1320],{},"Draft."," Create and edit content. The permission a junior or a freelancer should have on day one, on real accounts. There is no faster way to train someone and no reason to withhold it, because drafting is reversible.",[12,1323,1324,1327],{},[34,1325,1326],{},"Approve internally."," Assert that this is good enough to show the client: on brand, factually sourced, within scope. This is a quality and risk judgment and it belongs to whoever is accountable for the account, not to whoever wrote it.",[12,1329,1330,1333],{},[34,1331,1332],{},"Release to client."," Send it out of the agency. Separate from internal approval because it is the moment work becomes visible to someone who is paying, and because the person who should own client communication is often not the person who owns editorial quality.",[12,1335,1336,1339],{},[34,1337,1338],{},"Publish."," Put it on the account. The narrowest permission, and the one most agencies hand out most casually.",[12,1341,1342,1343,1346],{},"The important claim: ",[34,1344,1345],{},"approve and publish must not be the same permission."," A post can be approved on Tuesday and correct, and publishing it on Thursday can still be wrong because the client announced a layoff on Wednesday. Someone has to own the moment, not just the artefact.",[19,1348,1350],{"id":1349},"the-three-roles-this-produces","The three roles this produces",[12,1352,1353],{},"Map the permissions onto people and you get a shape that works from three people to thirty.",[12,1355,1356,1359],{},[34,1357,1358],{},"Writer"," — drafts. Junior, freelancer, or a senior on a client they know well. Has access to that brand's voice definition, fact file and forbidden list, and nothing else. Cannot release, cannot publish.",[12,1361,1362,1365],{},[34,1363,1364],{},"Account lead"," — approves internally and releases to the client. Accountable for the account: the editorial standard, the factual sourcing, the client relationship. One named person per client, always, including holiday cover assigned in advance.",[12,1367,1368,1371],{},[34,1369,1370],{},"Publisher"," — publishes on schedule. Often the same person as the account lead in a small agency, and that is fine. What matters is that it is a distinct act with its own check, not a side effect of approval. On larger books this is genuinely a separate role with a daily pre-publish sweep.",[12,1373,1374,1375,239],{},"Plus the client, who approves what the account lead released. That flow is a different problem with its own failure modes — see ",[71,1376,1377],{"href":865},"the approval workflow that doesn't lose the client",[19,1379,1381],{"id":1380},"what-each-stage-actually-checks","What each stage actually checks",[12,1383,1384],{},"Roles without checks are just job titles. Each gate is looking for something specific, and the specificity is what makes it fast.",[531,1386,1387,1400],{},[534,1388,1389],{},[537,1390,1391,1394,1397],{},[540,1392,1393],{},"Stage",[540,1395,1396],{},"The question",[540,1398,1399],{},"Blocks on",[547,1401,1402,1413,1424,1435],{},[537,1403,1404,1407,1410],{},[552,1405,1406],{},"Draft",[552,1408,1409],{},"Does this say something, in this brand's voice?",[552,1411,1412],{},"Generic content, wrong register",[537,1414,1415,1418,1421],{},[552,1416,1417],{},"Internal approve",[552,1419,1420],{},"Is every factual claim sourced and in scope?",[552,1422,1423],{},"Unsourced numbers, over-generalised claims",[537,1425,1426,1429,1432],{},[552,1427,1428],{},"Release",[552,1430,1431],{},"Is this the right thing to send this client this week?",[552,1433,1434],{},"Bad timing, scope creep, tone against context",[537,1436,1437,1440,1443],{},[552,1438,1439],{},"Publish",[552,1441,1442],{},"Has anything changed since approval?",[552,1444,1445],{},"News events, client announcements, live incidents",[12,1447,1448,1449,1452],{},"The internal approval gate is where ",[71,1450,1451],{"href":361},"wrong claims"," get caught, and it is the gate most often skipped when the account lead is busy. If one gate has to be defended, defend that one.",[19,1454,1456],{"id":1455},"give-juniors-real-work-not-fake-work","Give juniors real work, not fake work",[12,1458,1459],{},"The common failure is over-correcting: juniors get research and scheduling, seniors write everything, and the agency has a capacity ceiling equal to its senior headcount plus a training programme that produces nobody.",[12,1461,1462],{},"A permission model makes the safe version possible. A junior drafting on a live account with no release permission cannot cause an incident, so there is no reason to hold them back from the work that actually teaches — writing in someone else's voice against real constraints and getting edited.",[12,1464,1465],{},"Two things make this work in practice:",[27,1467,1468,1474],{},[30,1469,1470,1473],{},[34,1471,1472],{},"Edits go back with reasons."," \"Cut this, it's puffery\" teaches; a silently rewritten draft teaches nothing and is the most common way agencies fail to develop people.",[30,1475,1476,1479],{},[34,1477,1478],{},"The voice definition and fact file are accessible to the writer."," If the standard lives only in the account lead's head, the junior cannot hit it and the feedback loop is arbitrary. Writing it down is what makes delegation possible at all.",[19,1481,1483],{"id":1482},"where-the-model-breaks","Where the model breaks",[12,1485,1486,1489],{},[34,1487,1488],{},"One person is every role on eight accounts."," Common, and usually described as being busy. It is a single point of failure with no holiday cover and no second read on any claim. The first fix is not hiring — it is assigning a named second approver per client, even if that person only reads.",[12,1491,1492,1495],{},[34,1493,1494],{},"The gates exist but everyone has every permission."," Then the model is documentation, not a control. Permissions that are not enforced are conventions, and conventions collapse exactly when the week is bad.",[12,1497,1498,1501],{},[34,1499,1500],{},"Approval lives in chat."," The gates ran, the record did not. Six weeks later nobody can say who approved what, which is the state you are in during precisely the conversation where it matters. Approvals belong attached to the post.",[12,1503,1504,1507],{},[34,1505,1506],{},"Nobody owns publish."," So it happens automatically at the scheduled time, including the morning of a client's outage. Automation is fine for the mechanics; someone still has to own the last look.",[19,1509,1511],{"id":1510},"scaling-it-across-a-book","Scaling it across a book",[12,1513,1514],{},"At three clients the model is a shared understanding. At fifteen it needs to be enforced by something, because the number of people-times-brands combinations exceeds what anyone tracks reliably, and the failure is silent — nobody notices a missing gate until an incident.",[12,1516,1517,1518,1520,1521,1525],{},"What that requires is per-brand scoping rather than agency-wide roles: a writer added to two accounts sees two accounts, and permissions travel with the brand rather than with a job title. That is how ",[71,1519,7],{"href":422}," models it — draft, approve and client release are separate per-brand permissions, and every transition is recorded against the post itself, so the trail exists without anyone maintaining it. When ",[71,1522,1524],{"href":1523},"\u002Fblog\u002Fhanding-over-a-social-account","an account changes hands",", that record is most of what makes the handover survivable.",[19,1527,210],{"id":209},[12,1529,1530,1533],{},[34,1531,1532],{},"What roles does an agency need for social media content?","\nAt minimum three: a writer who drafts, an account lead who approves internally and releases to the client, and someone who owns publishing. In small agencies the last two are the same person, but they should remain distinct acts with distinct checks.",[12,1535,1536,1539],{},[34,1537,1538],{},"Should approval and publishing be the same permission?","\nNo. A post can be approved and correct on Tuesday and wrong to publish on Thursday because circumstances changed. Someone has to own the moment of publishing, not just the approval of the artefact.",[12,1541,1542,1545],{},[34,1543,1544],{},"Can juniors draft on live client accounts?","\nYes, and they should. Drafting is reversible and it is the only thing that actually trains someone. What juniors should not have is release or publish permission — that separation is what makes real delegation safe.",[12,1547,1548,1551],{},[34,1549,1550],{},"How do you stop client approvals from living in Slack?","\nAttach approvals to the post rather than to a conversation. The gate running is not enough; if the record is in chat, nobody can reconstruct who approved what six weeks later, which is exactly when it is asked.",[12,1553,1554,1557],{},[34,1555,1556],{},"How many approvers should a client have?","\nOne named approver with a defined tiebreak, plus assigned cover for absence. Two peers who can both comment and disagree with no stated precedence is the most common cause of stalled approvals.",{"title":247,"searchDepth":248,"depth":248,"links":1559},[1560,1561,1562,1563,1564,1565,1566],{"id":1311,"depth":248,"text":1312},{"id":1349,"depth":248,"text":1350},{"id":1380,"depth":248,"text":1381},{"id":1455,"depth":248,"text":1456},{"id":1482,"depth":248,"text":1483},{"id":1510,"depth":248,"text":1511},{"id":209,"depth":248,"text":210},"2026-08-19","The three-role split that lets juniors do real work without release risk, why approval and release are different permissions, and how to scale it across a book of clients.",{},{"title":1296,"description":1568},"blog\u002Fagency-roles-for-social-content",[505,805],"eyiBKklu-EKdbjB8AgAXGqewqg7Cmv5JK7FEY6Z0SsY",{"id":1575,"title":1576,"author":7,"body":1577,"date":1567,"description":1849,"draft":259,"extension":260,"image":261,"meta":1850,"navigation":263,"path":719,"seo":1851,"stem":1852,"tags":1853,"updated":261,"__hash__":1854},"blog\u002Fblog\u002Fclient-reporting-that-gets-read.md","The client report that actually gets read",{"type":9,"value":1578,"toc":1840},[1579,1582,1585,1588,1592,1598,1601,1606,1609,1613,1616,1622,1628,1634,1640,1646,1650,1653,1658,1690,1695,1721,1725,1728,1731,1737,1743,1749,1756,1760,1766,1772,1778,1782,1785,1792,1798,1800,1806,1812,1818,1824,1830],[12,1580,1581],{},"The monthly report is the most-produced and least-read artefact in social media work. Twelve slides of charts, an impressions number nobody can act on, and a client who skims for the one line that says whether this is working.",[12,1583,1584],{},"That skim is the whole document. Everything else is you demonstrating effort, which is a bad thing to be paid for and a worse thing to be judged on. A report is not evidence of work; it is an instrument for making the next month's decisions.",[12,1586,1587],{},"Here is the version that gets read, and the parts to delete.",[19,1589,1591],{"id":1590},"start-with-the-answer","Start with the answer",[12,1593,1594,1595],{},"The first line of the report answers the question the client actually has: ",[34,1596,1597],{},"is this working, and what are we changing?",[12,1599,1600],{},"Not \"here's what we did in March.\" Something like:",[882,1602,1603],{},[12,1604,1605],{},"March: the customer-story posts outperformed everything else by a wide margin, so April shifts to two of those a month and drops the industry-commentary slot. One post drove 14 profile visits and 3 inbound messages — the highest of the year.",[12,1607,1608],{},"Everything after that first paragraph is support for a claim already made. If a client reads only the top of the page, they have the answer, and they will trust the rest more for not having to hunt for it.",[19,1610,1612],{"id":1611},"the-one-page-structure","The one-page structure",[12,1614,1615],{},"Five blocks, one page, same order every month. Consistency matters more than completeness — a report whose shape changes each month cannot be compared with the last one, which is most of what a report is for.",[12,1617,1618,1621],{},[34,1619,1620],{},"1. The verdict."," Two or three sentences. Working, not working, what changes.",[12,1623,1624,1627],{},[34,1625,1626],{},"2. What went out."," Volume and mix, stated plainly. \"11 posts: 4 customer stories, 3 product, 2 opinion, 2 hiring.\" This is the only place effort belongs, and one line is enough.",[12,1629,1630,1633],{},[34,1631,1632],{},"3. What worked, with the actual post."," Not a chart — the top two posts, embedded or quoted, with one sentence on why you think each landed. Clients engage with their own content far more than with any metric about it, and \"why it landed\" is the part they cannot get from a dashboard.",[12,1635,1636,1639],{},[34,1637,1638],{},"4. What did not, and the hypothesis."," The block most people delete, and the one that builds the most trust. A named underperformer with a specific reading — \"the two product posts led with the feature rather than the problem\" — proves you are reading results rather than collecting them.",[12,1641,1642,1645],{},[34,1643,1644],{},"5. Next month, and what you need."," Three to five bullets of what you will do, plus an explicit list of what you need from them and by when. This is where the report stops being a rearview mirror and starts causing something.",[19,1647,1649],{"id":1648},"which-metrics-to-keep","Which metrics to keep",[12,1651,1652],{},"Report metrics the client can act on and that are plausibly caused by the work.",[12,1654,1655],{},[34,1656,1657],{},"Keep:",[27,1659,1660,1666,1672,1678,1684],{},[30,1661,1662,1665],{},[34,1663,1664],{},"Reach or impressions",", as context only, one number, no chart.",[30,1667,1668,1671],{},[34,1669,1670],{},"Engagement rate",", because it is comparable across post sizes in a way raw likes are not.",[30,1673,1674,1677],{},[34,1675,1676],{},"Profile visits and follows from content"," — the nearest honest proxy for \"people wanted more.\"",[30,1679,1680,1683],{},[34,1681,1682],{},"Anything countable that touches pipeline:"," inbound messages, demo requests, link clicks to the page that matters, applications for an open role.",[30,1685,1686,1689],{},[34,1687,1688],{},"A qualitative line."," Who engaged, not how many. \"Two heads of engineering at target accounts commented\" outranks a thousand impressions and no report format captures it automatically.",[12,1691,1692],{},[34,1693,1694],{},"Drop:",[27,1696,1697,1703,1709,1715],{},[30,1698,1699,1702],{},[34,1700,1701],{},"Follower count as a headline."," It moves slowly, it is easy to game and it rarely causes anything.",[30,1704,1705,1708],{},[34,1706,1707],{},"Every chart with fewer than three months of data."," Two points is not a trend; presenting it as one trains the client to over-react to noise.",[30,1710,1711,1714],{},[34,1712,1713],{},"Platform-native \"engagement scores.\""," Different definitions per platform and not comparable.",[30,1716,1717,1720],{},[34,1718,1719],{},"Anything you cannot explain the cause of."," If you cannot say why it moved, it is decoration.",[19,1722,1724],{"id":1723},"reporting-a-bad-month","Reporting a bad month",[12,1726,1727],{},"Every account has them, and how you report the third bad month determines whether there is a fourth.",[12,1729,1730],{},"Three rules:",[12,1732,1733,1736],{},[34,1734,1735],{},"Say it in the first line."," Burying a bad month in slide nine reads as concealment when it is discovered, and it always is. Leading with it reads as control.",[12,1738,1739,1742],{},[34,1740,1741],{},"Distinguish the two kinds of bad."," \"The content underperformed\" and \"we published four posts instead of eleven because material never arrived\" are entirely different problems with different owners. Conflating them is either taking blame you do not deserve or hiding a dependency that needs fixing. Say which one it was, without editorialising.",[12,1744,1745,1748],{},[34,1746,1747],{},"Bring a change, not a reassurance."," \"We'll keep at it\" is what a client hears as nothing. One specific alteration, with what you expect it to do, is what keeps the next month funded.",[12,1750,1751,1752,1755],{},"The failure mode worth naming: smoothing a bad month by finding a metric that happened to go up. Everyone can tell, and it costs more credibility than the bad month did. If the data is thin, say the data is thin — ",[71,1753,1754],{"href":422},"honest measurement"," survives a quarter of noise far better than a curated one does.",[19,1757,1759],{"id":1758},"frequency-and-format","Frequency and format",[12,1761,1762,1765],{},[34,1763,1764],{},"Monthly, not weekly."," Weekly reporting on social is mostly noise and turns into a treadmill that consumes bucket-four time you are not being paid for.",[12,1767,1768,1771],{},[34,1769,1770],{},"One page, in the body of an email or a shared doc."," Not a PDF attachment, not a slide deck, not a dashboard link. The report should be readable on a phone in ninety seconds without a login. Dashboards fail as reports for one reason: a dashboard shows numbers, and a report makes an argument.",[12,1773,1774,1777],{},[34,1775,1776],{},"Quarterly, go longer."," Once every three months there is enough data for an actual trend, and that is the meeting where you discuss strategy, scope and rate. The monthly one-pager is not the place for that conversation.",[19,1779,1781],{"id":1780},"the-compounding-benefit","The compounding benefit",[12,1783,1784],{},"There is a second reason to write reports this way, and it shows up months later.",[12,1786,1787,1788,1791],{},"A year of one-page reports that each name a hypothesis and a result is an argument for your rate that no proposal can match. It is also the record that settles disagreements about what was agreed, which is the same reason ",[71,1789,1790],{"href":865},"approvals need a trail",". Reports written as decisions accumulate into evidence; reports written as dashboards accumulate into nothing.",[12,1793,1794,1795,1797],{},"Across a book of clients this is exactly the part that decays under load — the fifth client's report gets written at 11pm and quietly becomes a metrics dump. ",[71,1796,7],{"href":422}," keeps what was published, what was approved and what it produced attached to each brand, so the monthly report is assembled from a record rather than reconstructed from memory.",[19,1799,210],{"id":209},[12,1801,1802,1805],{},[34,1803,1804],{},"What should be in a social media report for a client?","\nA verdict in the first two sentences, what went out, the top posts with why they worked, what underperformed and your hypothesis, and next month's plan with what you need from the client. One page, same structure every month.",[12,1807,1808,1811],{},[34,1809,1810],{},"Which social media metrics actually matter to clients?","\nEngagement rate, profile visits and follows from content, and anything countable that touches pipeline — inbound messages, demo requests, relevant link clicks. Follower count and raw impressions are context at best and misleading as headlines.",[12,1813,1814,1817],{},[34,1815,1816],{},"How often should I report to a social media client?","\nMonthly for the one-pager, quarterly for the longer strategic review. Weekly reporting on social is mostly noise and consumes unbilled time without improving decisions.",[12,1819,1820,1823],{},[34,1821,1822],{},"How do I tell a client a month went badly?","\nIn the first line, distinguishing content that underperformed from work that could not happen because material never arrived, and paired with one specific change you are making. Burying it reads as concealment; leading with it reads as control.",[12,1825,1826,1829],{},[34,1827,1828],{},"Should I send a dashboard link instead of a report?","\nNo. A dashboard shows numbers; a report makes an argument about what to do next. Clients who receive dashboard links stop opening them within about two months.",[12,1831,1832,1835,1836,1839],{},[34,1833,1834],{},"What about reports going out under the agency's own brand?","\nSame structure, plus two things the logo swap does not give you: the client must be able to open it without an account, and your reasoning has to be visible in it. ",[71,1837,1838],{"href":1289},"White-label reporting"," covers the difference.",{"title":247,"searchDepth":248,"depth":248,"links":1841},[1842,1843,1844,1845,1846,1847,1848],{"id":1590,"depth":248,"text":1591},{"id":1611,"depth":248,"text":1612},{"id":1648,"depth":248,"text":1649},{"id":1723,"depth":248,"text":1724},{"id":1758,"depth":248,"text":1759},{"id":1780,"depth":248,"text":1781},{"id":209,"depth":248,"text":210},"Most social media reports are dashboards nobody opens. A one-page structure built around decisions, which metrics to drop, and how to report a bad month without losing the account.",{},{"title":1576,"description":1849},"blog\u002Fclient-reporting-that-gets-read",[805,505],"WZEBzJQKRUXqNCXuiO-elAtcvy4JqtHdwbCjo_36foA",{"id":1856,"title":1857,"author":7,"body":1858,"date":1567,"description":2070,"draft":259,"extension":260,"image":261,"meta":2071,"navigation":263,"path":2072,"seo":2073,"stem":2074,"tags":2075,"updated":261,"__hash__":2076},"blog\u002Fblog\u002Ffounder-content-30-minutes-a-week.md","Founder content in 30 minutes a week",{"type":9,"value":1859,"toc":2061},[1860,1863,1866,1869,1873,1876,1882,1888,1891,1895,1898,1904,1910,1920,1923,1927,1930,1956,1959,1963,1969,1980,1986,1997,2001,2004,2010,2016,2020,2023,2029,2031,2037,2043,2049,2055],[12,1861,1862],{},"Most founders who \"have no time for content\" have never tried the version that works. They have tried writing something every morning, which fails for everybody, then concluded the problem was time. The problem was the schedule.",[12,1864,1865],{},"Writing four posts in one sitting takes far less than four times writing one post, because the expensive part is not the typing. It is deciding what to say, getting into the register, and overcoming the blank page. Do that once a week instead of five times.",[12,1867,1868],{},"Here is the method, and the two places it breaks.",[19,1870,1872],{"id":1871},"the-two-habits-not-one","The two habits, not one",[12,1874,1875],{},"This runs on a capture habit and a writing habit, and people who try to skip the first one always fail.",[12,1877,1878,1881],{},[34,1879,1880],{},"Habit 1 — capture, 10 seconds, whenever."," One note, anywhere you will actually open. A customer said something. A number moved. You changed your mind about something you believed six months ago. You are not writing, you are dropping a line. \"Migration went 4h → 20m after the importer rewrite.\" That is enough.",[12,1883,1884,1887],{},[34,1885,1886],{},"Habit 2 — write, 30 minutes, once a week."," Open the capture list, pick four, write four posts, done.",[12,1889,1890],{},"Without habit 1, habit 2 is just a blank page with a calendar invite attached. The whole thing lives or dies on whether you can produce raw material without it feeling like work.",[19,1892,1894],{"id":1893},"what-actually-goes-in-the-30-minutes","What actually goes in the 30 minutes",[12,1896,1897],{},"Timeboxed, because the point is that it ends.",[12,1899,1900,1903],{},[34,1901,1902],{},"Minutes 0–5: pick four."," Not the four best. The four you can write fastest, because you already know what you think. Anything requiring research goes back on the list.",[12,1905,1906,1909],{},[34,1907,1908],{},"Minutes 5–25: write four drafts, five minutes each."," One pass, no editing, deliberately rough. Five minutes is short enough that you cannot polish, which is the point — polishing is where the hour goes.",[12,1911,1912,1915,1916,239],{},[34,1913,1914],{},"Minutes 25–30: one editing pass across all four."," Cut every first paragraph (it is almost always throat-clearing), replace the biggest adjective in each with a fact, and check each one has something only you could have written. The full pass is in ",[71,1917,1919],{"href":1918},"\u002Fblog\u002Fhumanize-chatgpt-draft","how to edit a draft so it sounds like you",[12,1921,1922],{},"Then queue them. Do not schedule all four for the same channel on consecutive days if they are all the same shape — mix a short observation with a longer piece.",[19,1924,1926],{"id":1925},"the-four-kinds-of-post-you-already-have-material-for","The four kinds of post you already have material for",[12,1928,1929],{},"The reason a founder's blank page is a lie: you generate publishable material every week and throw it away.",[439,1931,1932,1938,1944,1950],{},[30,1933,1934,1937],{},[34,1935,1936],{},"A number that moved, with the story behind it."," Not \"growth is up.\" What changed and why. Specific numbers are the single strongest thing a founder has and the thing they most often leave out.",[30,1939,1940,1943],{},[34,1941,1942],{},"A decision and the alternative you rejected."," \"We chose X over Y because Z\" is more interesting than any framework, and only you can write it.",[30,1945,1946,1949],{},[34,1947,1948],{},"Something a customer said, verbatim."," Half a sentence from a support ticket, with what you did about it.",[30,1951,1952,1955],{},[34,1953,1954],{},"Something you were wrong about."," The highest-performing founder content is almost always this, and the most under-supplied.",[12,1957,1958],{},"Notice that none of these require an opinion about the industry. Founders default to industry takes because they feel safe and require no disclosure — and they are the most replaceable thing in the feed, because a thousand people can write them and only you can write the other four.",[19,1960,1962],{"id":1961},"what-to-cut","What to cut",[12,1964,1965,1968],{},[34,1966,1967],{},"Cut the daily target."," Two good posts a week beats five filler ones, and filler is what a daily target produces on week three.",[12,1970,1971,1974,1975,1979],{},[34,1972,1973],{},"Cut the second channel until the first one works."," Most founders should run LinkedIn or X, not both. Cross-posting the same text to both is worse than picking one, because the formats reward different shapes: ",[71,1976,1978],{"href":1977},"\u002Fblog\u002Fx-thread-structures","threads"," are not long LinkedIn posts.",[12,1981,1982,1985],{},[34,1983,1984],{},"Cut comment-hour."," Engagement pods and reply-farming produce numbers that do not turn into anything. Reply when you have something to say.",[12,1987,1988,1991,1992,1996],{},[34,1989,1990],{},"Cut the content calendar spreadsheet."," For one brand at this volume it is overhead pretending to be strategy. A capture list and a queue is the entire system you need. (Running several brands is genuinely different — that is where ",[71,1993,1995],{"href":1994},"\u002Fblog\u002Fcontent-calendar-you-can-keep","a real calendar"," starts earning its keep.)",[19,1998,2000],{"id":1999},"where-it-breaks","Where it breaks",[12,2002,2003],{},"Two failure modes, and they are predictable.",[12,2005,2006,2009],{},[34,2007,2008],{},"The capture list runs dry."," Usually a sign you have stopped talking to customers, or that you are filtering too hard at capture time. Capture is not the place to judge whether something is interesting; write it down and decide later. If the list is genuinely empty for two weeks, that is information about your week, not your content.",[12,2011,2012,2015],{},[34,2013,2014],{},"A brutal month eats the block."," Fundraising, an outage, a launch. This is the one thing a founder cannot solve with discipline, and it is exactly the moment going dark hurts most. Two defences: keep two posts in the queue as a buffer at all times, and lower the bar rather than skipping — one rough post beats a silent month.",[19,2017,2019],{"id":2018},"when-30-minutes-stops-being-enough","When 30 minutes stops being enough",[12,2021,2022],{},"This method holds up well for one brand and one channel. It starts creaking when the constraint moves from writing to deciding: when you have three months of history to keep straight, a queue you cannot remember the contents of, and a growing list of things you already said. That is not a discipline problem, it is a memory problem, and it is the point where founders quietly stop.",[12,2024,2025,2026,2028],{},"That is the gap ",[71,2027,7],{"href":181}," fills for founders: the capture list becomes a queue Chloe reads, she proposes what the week should say based on what you already covered and what facts you confirmed, and drafts in your voice from your own material rather than inventing around the gaps. The 30 minutes becomes ten minutes of approving instead of thirty of writing.",[19,2030,210],{"id":209},[12,2032,2033,2036],{},[34,2034,2035],{},"How often should a founder post on LinkedIn?","\nTwo to three times a week, sustained, beats daily posting that stops after a month. Consistency over months is what compounds; frequency inside a given week matters far less than most advice implies.",[12,2038,2039,2042],{},[34,2040,2041],{},"What should a founder post about?","\nThings only you have access to: numbers that moved and why, decisions and the alternatives you rejected, verbatim customer feedback, and things you turned out to be wrong about. Industry commentary is the most replaceable content a founder can write.",[12,2044,2045,2048],{},[34,2046,2047],{},"How long should it take to write a LinkedIn post?","\nFive to eight minutes for a draft if you already know what you think, which is why batching works — the expensive part is deciding and getting into register, and batching pays that cost once instead of four times.",[12,2050,2051,2054],{},[34,2052,2053],{},"Is it better to batch content or write daily?","\nBatch. Daily writing means paying the blank-page cost five times a week and having no buffer when a bad week arrives. One 30-minute block plus a running capture list survives chaos in a way a daily habit does not.",[12,2056,2057,2060],{},[34,2058,2059],{},"What if I have nothing to write about?","\nIt is almost always a capture problem, not a supply problem. Founders generate publishable material constantly and discard it because it does not feel significant in the moment. Start writing one line a day for two weeks before concluding the well is dry.",{"title":247,"searchDepth":248,"depth":248,"links":2062},[2063,2064,2065,2066,2067,2068,2069],{"id":1871,"depth":248,"text":1872},{"id":1893,"depth":248,"text":1894},{"id":1925,"depth":248,"text":1926},{"id":1961,"depth":248,"text":1962},{"id":1999,"depth":248,"text":2000},{"id":2018,"depth":248,"text":2019},{"id":209,"depth":248,"text":210},"A batched weekly method for founders who keep going dark: one capture habit, one 30-minute block, four posts. What to cut, what to keep, and where it breaks.",{},"\u002Fblog\u002Ffounder-content-30-minutes-a-week",{"title":1857,"description":2070},"blog\u002Ffounder-content-30-minutes-a-week",[268,805],"c60V464WB1BI_vgU9XOGDOMBQj1HC-370I227cxxQkk",{"id":2078,"title":2079,"author":7,"body":2080,"date":1567,"description":2289,"draft":259,"extension":260,"image":261,"meta":2290,"navigation":263,"path":1523,"seo":2291,"stem":2292,"tags":2293,"updated":261,"__hash__":2294},"blog\u002Fblog\u002Fhanding-over-a-social-account.md","Handing over a social account without losing the voice",{"type":9,"value":2081,"toc":2281},[2082,2085,2088,2091,2095,2098,2104,2110,2119,2125,2128,2132,2135,2138,2167,2170,2174,2177,2183,2189,2195,2201,2207,2211,2214,2224,2230,2233,2237,2240,2246,2249,2251,2257,2263,2269,2275],[12,2083,2084],{},"An account changes hands. The new strategist reads the last three months of posts, the brand doc if there is one, and starts writing. Nothing obviously breaks. Six weeks later the client says something vague about the posts not feeling like them lately, and cannot point at a single one.",[12,2086,2087],{},"They are right, and the cause is structural. Reading published posts teaches you the output, not the reasoning: which angles were rejected, which words are banned, what the client reacted badly to in March, why the account stopped doing the Friday format. All of that lived with the person who left.",[12,2089,2090],{},"Voice does not survive a handover by default. It survives because someone captured the part that was never written down.",[19,2092,2094],{"id":2093},"what-is-actually-lost","What is actually lost",[12,2096,2097],{},"Four categories, in rough order of how much damage their loss causes.",[12,2099,2100,2103],{},[34,2101,2102],{},"1. Rejections."," The single most valuable and least recorded artefact. Every angle the client killed, every phrasing that came back marked up, every topic that turned out to be sensitive. The published archive shows what passed; it says nothing about what was tried. A new writer without the rejection history will re-propose the same rejected ideas within a month, which is what a client experiences as \"they don't know us.\"",[12,2105,2106,2109],{},[34,2107,2108],{},"2. The forbidden list."," Claims they cannot legally make, competitors they will not name, words banned internally. Usually held in memory after the first time someone got told off.",[12,2111,2112,2115,2116,239],{},[34,2113,2114],{},"3. Fact provenance."," Which numbers are current, which were true in March, which need legal review. Undated facts in an old post are how a stale claim gets recycled into a new one — the second-order version of ",[71,2117,2118],{"href":361},"publishing a wrong claim",[12,2120,2121,2124],{},[34,2122,2123],{},"4. Relationship texture."," Who actually approves versus who comments, how blunt you can be, whether they want to be consulted or informed. Not voice exactly, but it determines how much friction the next three months carry.",[12,2126,2127],{},"The published posts capture none of these. That is why \"read the last quarter and you'll pick it up\" fails reliably.",[19,2129,2131],{"id":2130},"capture-it-before-you-need-it","Capture it before you need it",[12,2133,2134],{},"The only version of this that works is continuous. A handover document written during a handover is written by someone already mentally gone, under time pressure, from memory that is already degrading.",[12,2136,2137],{},"Four artefacts, maintained per account as a matter of course:",[27,2139,2140,2149,2155,2161],{},[30,2141,2142,2145,2146,239],{},[34,2143,2144],{},"The voice definition",", concrete enough that two writers produce similar posts from it. Three real sample posts beat any list of adjectives — the standard is in ",[71,2147,2148],{"href":658},"keeping several brand voices straight",[30,2150,2151,2154],{},[34,2152,2153],{},"The fact file",", with dates. Every citable number, when it was true, where it came from.",[30,2156,2157,2160],{},[34,2158,2159],{},"The forbidden list."," Claims, comparisons, words, customers who may not be named.",[30,2162,2163,2166],{},[34,2164,2165],{},"The rejection log."," One line per killed angle, with the reason. This is the one nobody keeps and the one that matters most. It costs ten seconds at the moment of rejection and is unreconstructable afterwards.",[12,2168,2169],{},"If these exist, a handover is a week. If they do not, it is a quarter, and the client pays for the difference in quality without being told.",[19,2171,2173],{"id":2172},"the-two-week-handover","The two-week handover",[12,2175,2176],{},"Assuming the outgoing person is still available. Compress it if they are not, and accept a worse result.",[12,2178,2179,2182],{},[34,2180,2181],{},"Days 1–2 — Written transfer."," The four artefacts above, updated to current. Plus one page the outgoing strategist writes from memory: what the client actually cares about, what they say they care about, and where those differ. That page is worth more than the rest combined.",[12,2184,2185,2188],{},[34,2186,2187],{},"Days 3–4 — The walkthrough."," Go through the last quarter's posts together, outgoing person narrating: why this angle, what got cut, what the client said. Record it. This is the reasoning layer that the archive does not contain, and an hour of it saves a month.",[12,2190,2191,2194],{},[34,2192,2193],{},"Days 5–8 — Shadow drafting."," The incoming writer drafts; the outgoing one edits with reasons attached. Nobody sends anything new to the client yet. Three or four rounds is usually enough for the register to transfer, and the edits themselves become additions to the voice definition.",[12,2196,2197,2200],{},[34,2198,2199],{},"Days 9–10 — Introduce, deliberately."," Tell the client, before they notice. Frame it as continuity with a name and a date, not as an apology. Clients handle a planned handover well and an unannounced one badly, and the difference is entirely in whether they found out from you.",[12,2202,2203,2206],{},[34,2204,2205],{},"Days 11–14 — Supervised solo."," Incoming writer runs the account; outgoing person reviews before release. Then stop.",[19,2208,2210],{"id":2209},"the-drift-check","The drift check",[12,2212,2213],{},"The failure is gradual, which is why it is caught late. Two cheap checks catch it inside a month.",[12,2215,2216,2219,2220,2223],{},[34,2217,2218],{},"The blind test."," Take three new posts, strip the logo and any product names, and ask someone who knows the brand which client they belong to. If the answer is \"could be any of them,\" the voice has already flattened toward the writer's default. This is the same check that catches ",[71,2221,2222],{"href":658},"voice bleed across clients",", and it works for the same reason.",[12,2225,2226,2229],{},[34,2227,2228],{},"The rejection replay."," Pull five angles from the rejection log and ask the new writer whether they would propose them. If they would propose three, the reasoning did not transfer, and the client is about to feel it.",[12,2231,2232],{},"Run both at week two and week six. After that, drift is usually the normal kind that any account has, not handover damage.",[19,2234,2236],{"id":2235},"why-this-keeps-happening","Why this keeps happening",[12,2238,2239],{},"Because the artefacts live in the wrong place. A brand doc in a shared drive gets updated during onboarding and never again; a rejection log in someone's notes app leaves with them; the fact file is a spreadsheet three versions behind whatever the client last confirmed.",[12,2241,2242,2243,2245],{},"The structural fix is that brand knowledge should live attached to the brand rather than to the person, and it should be updated as a side effect of the work rather than as a separate chore nobody does. That is what ",[71,2244,7],{"href":422}," holds per brand: voice profile, confirmed facts with their history, constraints, and the record of what was proposed, edited and rejected. The next strategist inherits the account rather than reverse-engineering it from published posts.",[12,2247,2248],{},"It also removes the incentive that quietly damages agencies — when knowledge lives with a person, the account is hostage to their retention, and everyone knows it.",[19,2250,210],{"id":209},[12,2252,2253,2256],{},[34,2254,2255],{},"How do you hand over a social media account to a new manager?","\nTransfer four written artefacts first — voice definition, dated fact file, forbidden list and rejection log — then walk through the last quarter's posts with the outgoing person narrating why each angle was chosen, then shadow-draft for a week before the new writer sends anything to the client.",[12,2258,2259,2262],{},[34,2260,2261],{},"Why does a brand's voice change when the writer changes?","\nBecause published posts show the output, not the reasoning. A new writer cannot see which angles were rejected, which words are banned or what the client reacted badly to, so they default toward their own register within a few weeks.",[12,2264,2265,2268],{},[34,2266,2267],{},"What is a rejection log and why does it matter?","\nA one-line record of every angle a client killed and why. It is the most valuable handover artefact and the least commonly kept — without it, a new strategist re-proposes rejected ideas within a month, which clients experience as not being understood.",[12,2270,2271,2274],{},[34,2272,2273],{},"Should you tell the client the account is changing hands?","\nYes, before they notice, framed as planned continuity with a name and a date. Clients handle announced handovers well and unannounced ones badly, and the difference is almost entirely in who told them first.",[12,2276,2277,2280],{},[34,2278,2279],{},"How long should a social media account handover take?","\nAbout two weeks with the outgoing person available: two days of written transfer, two of walkthrough, four of shadow drafting, then a supervised solo period. If the brand artefacts were maintained continuously it is closer to a week; if they were not, expect a quarter of degraded quality.",{"title":247,"searchDepth":248,"depth":248,"links":2282},[2283,2284,2285,2286,2287,2288],{"id":2093,"depth":248,"text":2094},{"id":2130,"depth":248,"text":2131},{"id":2172,"depth":248,"text":2173},{"id":2209,"depth":248,"text":2210},{"id":2235,"depth":248,"text":2236},{"id":209,"depth":248,"text":210},"When the strategist who ran an account leaves, the voice resets to whoever writes next. What to capture before it happens, a two-week handover plan, and the drift check that catches it early.",{},{"title":2079,"description":2289},"blog\u002Fhanding-over-a-social-account",[505,805],"4-PllwzsnCCa4EsTMXRWOcblfK2h5kKGj7ee3GcEeGQ",{"id":2296,"title":2297,"author":7,"body":2298,"date":1567,"description":2651,"draft":259,"extension":260,"image":261,"meta":2652,"navigation":263,"path":198,"seo":2653,"stem":2654,"tags":2655,"updated":261,"__hash__":2657},"blog\u002Fblog\u002Flinkedin-ghostwriter-cost.md","What a LinkedIn ghostwriter costs (and when one is worth it)",{"type":9,"value":2299,"toc":2642},[2300,2303,2306,2310,2313,2319,2325,2331,2334,2338,2344,2347,2367,2370,2373,2377,2380,2383,2414,2417,2421,2424,2433,2439,2449,2453,2574,2578,2581,2607,2609,2615,2621,2627,2636],[12,2301,2302],{},"Founders ask this question in the wrong order. They ask what a ghostwriter costs, get a number, decide it's a lot, and go back to posting nothing. The number is only half the decision. The other half is what your own time costs and how much of it a ghostwriter actually saves, which is less than most people assume.",[12,2304,2305],{},"Here are the real ranges, the cost that never appears on the invoice, and the point where hiring stops being a luxury.",[19,2307,2309],{"id":2308},"the-three-price-tiers","The three price tiers",[12,2311,2312],{},"Rates vary by market and by how much of the work is writing versus strategy, but ghostwriting offers cluster into three recognisable bands.",[12,2314,2315,2318],{},[34,2316,2317],{},"$500–1,200\u002Fmonth — the drafting tier."," Usually a freelancer, often junior or offshore, writing 8–12 posts a month from material you give them. You supply the ideas, the stories and the facts. They turn your voice notes into posts. At this tier you are buying typing and structure, not judgment.",[12,2320,2321,2324],{},[34,2322,2323],{},"$1,500–4,000\u002Fmonth — the standard tier."," This is where most experienced LinkedIn ghostwriters sit. You get a monthly interview call, 12–20 posts, some strategy on what to talk about, and edits. They will push back on weak angles. This is the band most founders mean when they say \"a ghostwriter.\"",[12,2326,2327,2330],{},[34,2328,2329],{},"$5,000–15,000\u002Fmonth — the operator tier."," An agency or a senior writer running the whole presence: content, comments, DMs, sometimes lead handling. Common for founders raising, hiring at scale, or selling something with a long sales cycle where LinkedIn is the pipeline.",[12,2332,2333],{},"Per-post pricing exists too, typically $100–400 a post at the standard tier, but it tends to be worse value: the writer's cost is mostly in learning your voice, and per-post deals never let them amortise that.",[19,2335,2337],{"id":2336},"the-cost-nobody-quotes-you-your-own-hours","The cost nobody quotes you: your own hours",[12,2339,2340,2341],{},"The invoice is the visible cost. The invisible one is that ",[34,2342,2343],{},"a ghostwriter cannot write about your business without you.",[12,2345,2346],{},"Every good ghostwriting relationship needs roughly:",[27,2348,2349,2355,2361],{},[30,2350,2351,2354],{},[34,2352,2353],{},"A 45–60 minute interview call",", monthly at minimum, often weekly at the start.",[30,2356,2357,2360],{},[34,2358,2359],{},"Voice notes or messages"," when something happens worth writing about — a customer result, a decision, a mistake.",[30,2362,2363,2366],{},[34,2364,2365],{},"A review pass on every draft."," Not a rubber stamp: you are the only person who knows whether the number is right.",[12,2368,2369],{},"Call it 3–5 hours a month once the relationship is warm, and considerably more for the first two months. If your hour is worth $200, that is another $600–1,000 of real cost on top of the fee.",[12,2371,2372],{},"This is not an argument against hiring. It is an argument against the fantasy that hiring makes content someone else's problem. It never does. The founder is the only source of the material, and no arrangement changes that.",[19,2374,2376],{"id":2375},"when-it-pays-for-itself","When it pays for itself",[12,2378,2379],{},"The honest test is not \"can I afford $2,000 a month.\" It is \"does one outcome from LinkedIn cover it.\"",[12,2381,2382],{},"Run the arithmetic on your own numbers:",[27,2384,2385,2391,2397,2403],{},[30,2386,2387,2390],{},[34,2388,2389],{},"If you sell something at $20k+ ACV",", one inbound conversation a quarter that closes pays for a year of ghostwriting. The bar is low and hiring is usually correct.",[30,2392,2393,2396],{},[34,2394,2395],{},"If you're hiring",", one senior engineer sourced inbound saves a recruiter fee of $25–40k. Same conclusion.",[30,2398,2399,2402],{},[34,2400,2401],{},"If you sell a $30\u002Fmonth product",", LinkedIn content almost never pays back as a direct acquisition channel at that price. It might still be worth it for hiring, fundraising or partnerships, but be honest that the payback is indirect.",[30,2404,2405,2408,2409,2413],{},[34,2406,2407],{},"If you're pre-revenue",", the ghostwriter is being paid out of runway to build an asset with an unknown payoff. Usually the wrong call. See ",[71,2410,2412],{"href":2411},"\u002Fblog\u002Fwhat-to-post-before-launch","what to post before you have a product"," for the cheaper version.",[12,2415,2416],{},"The second test is consistency. A ghostwriter's real value is that posts go out in a week when your product is on fire. If you would have posted anyway, you are buying polish. If you would have gone dark, you are buying the channel's existence.",[19,2418,2420],{"id":2419},"three-cheaper-setups-that-work","Three cheaper setups that work",[12,2422,2423],{},"Hiring is not the only way to solve \"I don't post.\"",[12,2425,2426,2429,2430,239],{},[34,2427,2428],{},"1. The 30-minute weekly block."," One recurring slot, four posts, batched. Most founders who \"don't have time\" have never actually tried the batched version — they have tried the write-something-every-morning version, which fails for everyone. The full method is in ",[71,2431,2432],{"href":2072},"founder content in 30 minutes a week",[12,2434,2435,2438],{},[34,2436,2437],{},"2. An editor instead of a writer."," You write badly and fast; someone else makes it publishable. Typically $300–800\u002Fmonth, and it keeps your voice because the words started as yours. Underrated, and much cheaper than full ghostwriting.",[12,2440,2441,2444,2445,2448],{},[34,2442,2443],{},"3. A tool that starts from your material instead of a prompt."," The reason generic AI does not replace a ghostwriter is that it does not know your facts, so it invents them, in a register that reads as machine-written to anyone who has seen a few. What closes that gap is not a better prompt but a system that stores your voice and your confirmed facts and drafts from those. That is ",[71,2446,2447],{"href":181},"what SelfSM does",", at roughly a tenth of the standard ghostwriting tier.",[19,2450,2452],{"id":2451},"the-comparison-plainly","The comparison, plainly",[531,2454,2455,2472],{},[534,2456,2457],{},[537,2458,2459,2461,2464,2467,2470],{},[540,2460],{},[540,2462,2463],{},"Ghostwriter",[540,2465,2466],{},"Editor",[540,2468,2469],{},"ChatGPT",[540,2471,7],{},[547,2473,2474,2491,2508,2525,2542,2559],{},[537,2475,2476,2479,2482,2485,2488],{},[552,2477,2478],{},"Monthly cost",[552,2480,2481],{},"$1,500–4,000",[552,2483,2484],{},"$300–800",[552,2486,2487],{},"~$20",[552,2489,2490],{},"$19",[537,2492,2493,2496,2499,2502,2505],{},[552,2494,2495],{},"Your hours per month",[552,2497,2498],{},"3–5",[552,2500,2501],{},"2–4",[552,2503,2504],{},"4–8",[552,2506,2507],{},"1–2",[537,2509,2510,2513,2516,2519,2522],{},[552,2511,2512],{},"Sounds like you",[552,2514,2515],{},"After 2–3 months",[552,2517,2518],{},"Yes, it's your text",[552,2520,2521],{},"Rarely",[552,2523,2524],{},"From your own posts",[537,2526,2527,2530,2533,2536,2539],{},[552,2528,2529],{},"Knows your facts",[552,2531,2532],{},"If you brief them",[552,2534,2535],{},"Yes",[552,2537,2538],{},"No — invents them",[552,2540,2541],{},"Stored per brand",[537,2543,2544,2547,2550,2553,2556],{},[552,2545,2546],{},"Decides what to post",[552,2548,2549],{},"They do",[552,2551,2552],{},"You do",[552,2554,2555],{},"You prompt it",[552,2557,2558],{},"Proposed weekly",[537,2560,2561,2564,2566,2569,2572],{},[552,2562,2563],{},"Survives a chaotic month",[552,2565,2535],{},[552,2567,2568],{},"Partly",[552,2570,2571],{},"No",[552,2573,2535],{},[19,2575,2577],{"id":2576},"what-to-ask-before-you-hire","What to ask before you hire",[12,2579,2580],{},"If you do hire, four questions separate the good ones from the expensive ones:",[439,2582,2583,2589,2595,2601],{},[30,2584,2585,2588],{},[34,2586,2587],{},"\"Show me a client's posts from before and after you started.\""," Voice match is the whole job. If the after looks like the writer's own feed, walk.",[30,2590,2591,2594],{},[34,2592,2593],{},"\"What happens when I go quiet for three weeks?\""," The honest answer is \"nothing gets written.\" Anyone promising otherwise is planning to invent your material.",[30,2596,2597,2600],{},[34,2598,2599],{},"\"Who owns the archive?\""," You want the voice guide, the angles list and the post history when the engagement ends. Otherwise you restart from zero with the next person.",[30,2602,2603,2606],{},[34,2604,2605],{},"\"What's the minimum I have to give you each week?\""," If they cannot answer in one sentence, the process is not real yet.",[19,2608,210],{"id":209},[12,2610,2611,2614],{},[34,2612,2613],{},"How much does a LinkedIn ghostwriter charge per post?","\nTypically $100–400 per post at the standard tier, though monthly retainers of $1,500–4,000 are far more common and usually better value, because most of a writer's cost is learning your voice once rather than writing each post.",[12,2616,2617,2620],{},[34,2618,2619],{},"Is a LinkedIn ghostwriter worth it for a founder?","\nIt depends almost entirely on deal size. At $20k+ ACV, one inbound deal a quarter covers a year of ghostwriting and the answer is usually yes. Below roughly $1k ACV, direct payback is rare, and the case has to come from hiring, fundraising or partnerships instead.",[12,2622,2623,2626],{},[34,2624,2625],{},"Can AI replace a LinkedIn ghostwriter?","\nIt replaces the drafting, not the sourcing. A model has no access to your customer numbers, your decisions or your mistakes, so it fills the gap with plausible invention. What works is a system that stores your real facts and voice and drafts only from those — the writing stops being the bottleneck, but you still have to supply the material.",[12,2628,2629,2632,2633,239],{},[34,2630,2631],{},"How do I stop ghostwritten posts sounding generic?","\nInsist on one specific thing in every post that only you could have said: a number, a name, a decision you regret. Generic posts are almost always posts written without enough raw material. The patterns to strip are in ",[71,2634,2635],{"href":125},"the tells that make writing sound like AI",[12,2637,2638,2641],{},[34,2639,2640],{},"How long before ghostwriting starts working?","\nTwo to three months for the voice to settle, and typically three to six before inbound is predictable enough to plan around. Anyone promising results in weeks is selling a different product.",{"title":247,"searchDepth":248,"depth":248,"links":2643},[2644,2645,2646,2647,2648,2649,2650],{"id":2308,"depth":248,"text":2309},{"id":2336,"depth":248,"text":2337},{"id":2375,"depth":248,"text":2376},{"id":2419,"depth":248,"text":2420},{"id":2451,"depth":248,"text":2452},{"id":2576,"depth":248,"text":2577},{"id":209,"depth":248,"text":210},"Typical ghostwriter pricing tiers, the hidden cost nobody quotes you, the maths on when it pays for itself, and the three cheaper setups that work almost as well.",{},{"title":2297,"description":2651},"blog\u002Flinkedin-ghostwriter-cost",[268,2656],"linkedin","yhl4R45Q4BJ1Ra_I6pjcs--YGASwQmhW5qEygeCN8Zc",{"id":2659,"title":2660,"author":7,"body":2661,"date":1567,"description":2898,"draft":259,"extension":260,"image":261,"meta":2899,"navigation":263,"path":678,"seo":2900,"stem":2901,"tags":2902,"updated":261,"__hash__":2903},"blog\u002Fblog\u002Fpricing-social-media-management.md","How to price social media management when half the work is invisible",{"type":9,"value":2662,"toc":2889},[2663,2666,2669,2672,2676,2679,2705,2712,2715,2719,2722,2748,2751,2755,2761,2767,2773,2779,2782,2786,2789,2798,2804,2810,2814,2817,2820,2837,2840,2844,2847,2853,2855,2865,2871,2877,2883],[12,2664,2665],{},"Ask a freelance social media manager what they charge and you get a monthly number. Ask what it works out to per hour and the conversation gets uncomfortable, because the number that gets quoted covers writing and posting, and writing and posting is roughly half of what actually happens.",[12,2667,2668],{},"The other half — chasing material, waiting on approvals, rewriting after a comment, reconstructing what the client agreed three weeks ago, the \"quick question\" on Slack — never appears in a proposal and never gets billed. It is not overhead. It is the job.",[12,2670,2671],{},"Price without counting it and you take on a fifth client to fix an income problem that was really a pricing problem, and end up worse off.",[19,2673,2675],{"id":2674},"count-the-invisible-half-first","Count the invisible half first",[12,2677,2678],{},"Before choosing a model, measure. For two weeks, log every minute spent on one client in four buckets:",[439,2680,2681,2687,2693,2699],{},[30,2682,2683,2686],{},[34,2684,2685],{},"Production."," Writing, editing, sourcing images, scheduling.",[30,2688,2689,2692],{},[34,2690,2691],{},"Input gathering."," Chasing the client for material, interview calls, digging through their site or old decks because nothing arrived.",[30,2694,2695,2698],{},[34,2696,2697],{},"Approval and revision."," Sending, waiting, chasing, rewriting after feedback, re-sending.",[30,2700,2701,2704],{},[34,2702,2703],{},"Account admin."," Reporting, invoicing, meetings, the Slack thread that took forty minutes.",[12,2706,2707,2708,2711],{},"Most people who do this exercise find production is 40–60% of the total. Buckets 2 and 3 together are usually larger than anyone expects, and they are the buckets that scale worst — they grow with the ",[519,2709,2710],{},"number"," of clients, not with output.",[12,2713,2714],{},"Now your real hourly rate is the monthly fee divided by all four buckets, not by bucket one. That number is the one to act on.",[19,2716,2718],{"id":2717},"why-per-post-pricing-loses","Why per-post pricing loses",[12,2720,2721],{},"Per-post looks fair and behaves badly.",[27,2723,2724,2730,2736,2742],{},[30,2725,2726,2729],{},[34,2727,2728],{},"It prices the cheapest part."," Writing is the commodity; knowing what to write and getting it approved is the expertise. Per-post pricing sells the commodity.",[30,2731,2732,2735],{},[34,2733,2734],{},"It punishes efficiency."," Getting faster at writing means earning less, which is a strange incentive to build a business on.",[30,2737,2738,2741],{},[34,2739,2740],{},"It invites scope creep with no meter."," Revisions, calls and \"can you also look at\" are outside the unit, so they arrive free.",[30,2743,2744,2747],{},[34,2745,2746],{},"It makes clients optimise the wrong thing."," A client paying per post asks for more posts. A client paying for an outcome asks whether the posts are working.",[12,2749,2750],{},"There is one situation where per-post is right: overflow work for another agency where you receive an approved brief and return text. That is genuinely piecework, and it should be priced as such.",[19,2752,2754],{"id":2753},"four-models-and-what-each-one-causes","Four models, and what each one causes",[12,2756,2757,2760],{},[34,2758,2759],{},"Monthly retainer, scoped by deliverable."," \"12 posts a month, 2 rounds of revision, one call.\" The default for good reason: predictable both ways, and the scope line makes overage a conversation rather than a resentment. Its weakness is that it still anchors on output volume. Mitigate with an explicit revision cap and an explicit response-time expectation from the client — most retainer losses come from bucket 3, not bucket 1.",[12,2762,2763,2766],{},[34,2764,2765],{},"Monthly retainer, scoped by outcome."," \"We run your LinkedIn presence.\" No post count. Higher trust, higher margin, only workable once you have a track record with that client, and dangerous early because an unscoped promise with an unresponsive client is unbounded work.",[12,2768,2769,2772],{},[34,2770,2771],{},"Day rate or hours."," Honest, easy to explain, caps your income at your calendar, and encourages clients to treat you as capacity rather than judgment. Useful as a strategy or audit offer alongside a retainer, less good as the main model.",[12,2774,2775,2778],{},[34,2776,2777],{},"Value or performance-linked."," Attractive in theory. In practice, attribution in social is weak enough that you will spend the relationship arguing about what caused what. Workable only where the outcome is countable and clearly yours — inbound demo requests from a single channel, for instance — and only with a floor retainer underneath.",[12,2780,2781],{},"For most freelancers the answer is model one, priced against all four buckets, with a hard scope line.",[19,2783,2785],{"id":2784},"price-the-things-nobody-prices","Price the things nobody prices",[12,2787,2788],{},"Three items belong in every proposal and are usually given away:",[12,2790,2791,2794,2795,239],{},[34,2792,2793],{},"Onboarding as a separate fee."," The first week is disproportionately expensive: interviews, voice definition, fact-gathering, access, calibration drafts. Charging it separately does two useful things — it recovers real cost, and it makes the client take onboarding seriously enough to actually send the material. The week itself is laid out in ",[71,2796,2797],{"href":696},"social media client onboarding",[12,2799,2800,2803],{},[34,2801,2802],{},"Revision limits."," Two rounds included, then charged. Not to make money on round three, but because unlimited revisions are how a profitable client becomes an unprofitable one without anyone noticing.",[12,2805,2806,2809],{},[34,2807,2808],{},"A client-responsiveness clause."," The most underused line in the industry: if material or approvals do not arrive by the agreed day, the work slips and the fee does not change. Without it, the client's delay becomes your unpaid overtime, every month.",[19,2811,2813],{"id":2812},"the-rate-raising-problem","The rate-raising problem",[12,2815,2816],{},"Rates go up when scope changes or when you have proof. The mistake is raising them quietly in month eleven with an apology attached.",[12,2818,2819],{},"Two things make it straightforward:",[27,2821,2822,2828],{},[30,2823,2824,2827],{},[34,2825,2826],{},"Announce increases at a fixed point"," — annually, or at renewal — so it is a calendar event rather than a negotiation you initiated.",[30,2829,2830,2833,2834,239],{},[34,2831,2832],{},"Attach it to evidence."," Not \"my costs went up\" but \"here is what the last six months produced.\" Which requires that you have been reporting on something more meaningful than impressions all year — see ",[71,2835,2836],{"href":719},"the client report that gets read",[12,2838,2839],{},"If a client will not move at renewal, that is information. The clients who resist a modest annual increase are almost always the ones consuming the most of buckets 2 and 3.",[19,2841,2843],{"id":2842},"the-structural-fix","The structural fix",[12,2845,2846],{},"Pricing better raises the ceiling. It does not change the fact that buckets 2 and 3 grow with client count, which is why the fifth client so often produces less profit than the third.",[12,2848,2849,2850,2852],{},"The structural fix is to shrink the unbillable half rather than bill for it: keep every brand's voice, facts and constraints somewhere durable so drafting does not start with reconstruction, and put approvals on the work itself with a recorded trail so nobody relitigates what was agreed. That is precisely what ",[71,2851,7],{"href":733}," is built to do — the drafting is the part that was already fine; the context and the chasing are what eat the week.",[19,2854,210],{"id":209},[12,2856,2857,2860,2861,2864],{},[34,2858,2859],{},"How much should I charge for social media management?","\nWork backwards from your real hourly rate: divide the monthly fee by ",[519,2862,2863],{},"all"," the time a client consumes, including chasing material, approvals and admin, not just production. Most people find production is only 40–60% of the total, which means an apparently healthy monthly number can be half the rate they think it is.",[12,2866,2867,2870],{},[34,2868,2869],{},"Should I charge per post or a monthly retainer?","\nA retainer, in almost every case. Per-post pricing sells the cheapest part of the work, punishes you for getting faster, and leaves revisions and calls unmetered. Per-post makes sense only for overflow work where you receive an approved brief and return text.",[12,2872,2873,2876],{},[34,2874,2875],{},"How many revisions should be included?","\nTwo rounds, stated explicitly, then billed. The cap matters less as a revenue line than as a signal — unlimited revisions turn a profitable client into an unprofitable one gradually enough that nobody notices until renewal.",[12,2878,2879,2882],{},[34,2880,2881],{},"Should I charge for onboarding separately?","\nYes. The first week costs disproportionately more than a typical month, and a separate onboarding fee also makes clients take the process seriously enough to actually send their material and grant access on time.",[12,2884,2885,2888],{},[34,2886,2887],{},"How do I raise rates with an existing client?","\nDo it at a fixed annual point rather than ad hoc, and attach it to evidence of what the work produced rather than to your costs. That requires having reported on something more meaningful than follower counts throughout the year.",{"title":247,"searchDepth":248,"depth":248,"links":2890},[2891,2892,2893,2894,2895,2896,2897],{"id":2674,"depth":248,"text":2675},{"id":2717,"depth":248,"text":2718},{"id":2753,"depth":248,"text":2754},{"id":2784,"depth":248,"text":2785},{"id":2812,"depth":248,"text":2813},{"id":2842,"depth":248,"text":2843},{"id":209,"depth":248,"text":210},"Why per-post pricing loses money, how to find your real hourly rate including the unbillable half, and four pricing models with the client behaviour each one causes.",{},{"title":2660,"description":2898},"blog\u002Fpricing-social-media-management",[805,505],"bbgenV_pQ3_4uJmsQCvUDcg5J6AxNr2sVcR2KATWVCE",{"id":2905,"title":2906,"author":7,"body":2907,"date":1567,"description":3300,"draft":259,"extension":260,"image":261,"meta":3301,"navigation":263,"path":696,"seo":3302,"stem":3303,"tags":3304,"updated":261,"__hash__":3305},"blog\u002Fblog\u002Fsocial-media-client-onboarding.md","Social media client onboarding: the first seven days",{"type":9,"value":2908,"toc":3288},[2909,2912,2915,2919,2922,2954,2957,2961,2964,3002,3007,3011,3014,3017,3043,3046,3050,3053,3059,3067,3074,3078,3081,3084,3088,3091,3117,3120,3124,3127,3147,3150,3154,3244,3248,3251,3256,3258,3264,3270,3276,3282],[12,2910,2911],{},"Almost every difficult client relationship was decided in the first week. Not by the work — by what was never asked, never written down, and never agreed. Three months later the argument is about a post, but the cause is that nobody defined who approves, what the brand refuses to say, or what happens when the client goes quiet for nine days.",[12,2913,2914],{},"Onboarding is the cheapest leverage in the whole engagement. Here is a seven-day version that front-loads the decisions.",[19,2916,2918],{"id":2917},"day-1-access-before-anything-else","Day 1 — Access, before anything else",[12,2920,2921],{},"Access requests take longer than anyone plans for, so they go first and run in parallel with everything else. Send one message with the complete list rather than discovering items one at a time over two weeks:",[27,2923,2924,2930,2936,2942,2948],{},[30,2925,2926,2929],{},[34,2927,2928],{},"Publishing access"," to each channel, granted the right way — as a role on the company page or a delegated account, never a shared password. Personal-profile logins break the moment the client changes their password, and they are an offboarding problem you do not want.",[30,2931,2932,2935],{},[34,2933,2934],{},"Analytics access",", read-only, for whatever they already have.",[30,2937,2938,2941],{},[34,2939,2940],{},"Brand assets:"," logo files, fonts, colours, image library, existing templates.",[30,2943,2944,2947],{},[34,2945,2946],{},"Anything already written:"," old posts, newsletters, website copy, sales decks, support macros. This is the single most valuable thing you will receive and clients rarely think to offer it.",[30,2949,2950,2953],{},[34,2951,2952],{},"One named person"," who can answer a question same-day. Not a group inbox.",[12,2955,2956],{},"Set the expectation now that missing access is a blocker with a date attached, not a background task.",[19,2958,2960],{"id":2959},"day-2-the-interview-that-determines-the-voice","Day 2 — The interview that determines the voice",[12,2962,2963],{},"An hour on a call, recorded. This is not a discovery call about goals; it is raw material extraction. Adjectives are useless here — \"professional but approachable\" tells you nothing two writers would agree on. Ask for specifics:",[27,2965,2966,2972,2978,2984,2990,2996],{},[30,2967,2968,2971],{},[34,2969,2970],{},"\"Show me three posts you were proud of, and three you disliked.\""," Then ask what was wrong with the second set. The dislikes are more informative than the likes.",[30,2973,2974,2977],{},[34,2975,2976],{},"\"Who specifically are you talking to?\""," A job title and a situation, not a demographic.",[30,2979,2980,2983],{},[34,2981,2982],{},"\"What do you refuse to say?\""," Claims they cannot make legally, comparisons they will not draw, words that are banned internally. This list prevents more damage than any style guide.",[30,2985,2986,2989],{},[34,2987,2988],{},"\"What's true that your competitors can't say?\""," This is the positioning question in a form clients can actually answer.",[30,2991,2992,2995],{},[34,2993,2994],{},"\"I or we?\""," And whether that ever changes by channel.",[30,2997,2998,3001],{},[34,2999,3000],{},"\"What went wrong with your last agency?\""," The most useful question on the list. It tells you the failure mode you are being hired to avoid.",[12,3003,3004,3005,239],{},"Turn the recording into a written voice definition the same week. If you cannot write a post from it without guessing, it is not finished — the standard for a usable definition is in ",[71,3006,2148],{"href":658},[19,3008,3010],{"id":3009},"day-3-facts-and-boundaries-written-down","Day 3 — Facts and boundaries, written down",[12,3012,3013],{},"Voice tells you how it sounds. Facts tell you what it is allowed to say. Separate them, because they change at different rates and different people own them.",[12,3015,3016],{},"Collect and store, per client:",[27,3018,3019,3025,3031,3037],{},[30,3020,3021,3024],{},[34,3022,3023],{},"Numbers you may cite",", with the date they were true. Customer counts, funding, results, retention. Undated numbers become wrong silently.",[30,3026,3027,3030],{},[34,3028,3029],{},"Named customers you may mention",", and whether by name or anonymously.",[30,3032,3033,3036],{},[34,3034,3035],{},"Claims requiring legal or compliance review"," before publishing.",[30,3038,3039,3042],{},[34,3040,3041],{},"The forbidden list"," from day 2.",[12,3044,3045],{},"This is the file that stops the expensive mistake: a plausible, well-written post containing a number nobody verified. It is also the file that makes an AI-assisted workflow safe rather than dangerous, because generated drafts fill gaps with invention unless something holds the real values.",[19,3047,3049],{"id":3048},"day-4-agree-the-approval-mechanics","Day 4 — Agree the approval mechanics",[12,3051,3052],{},"Two agreements, both in writing, both boring, both worth more than the rest of the contract.",[12,3054,3055,3058],{},[34,3056,3057],{},"Who approves, and who is the tiebreak."," One name. If the founder and the head of marketing both comment and disagree, you need to know in advance whose answer wins, because discovering it during a disagreement is how relationships fray.",[12,3060,3061,3064,3065,239],{},[34,3062,3063],{},"What happens on silence."," \"No response by end of day Thursday\" has to mean something specific: either the post auto-proceeds because it is low-risk, or it slips to next week and the client owns that. Ambiguity here is what turns one-day approvals into one-week ones. The full argument is in ",[71,3066,1377],{"href":865},[12,3068,3069,3070,3073],{},"Agree also ",[519,3071,3072],{},"when"," they see work: approving angles for the week is fast and cheap, approving a finished scheduled post feels like being handed a bill. Move the approval earlier and the whole engagement speeds up.",[19,3075,3077],{"id":3076},"day-5-the-first-draft-batch-deliberately-early","Day 5 — The first draft batch, deliberately early",[12,3079,3080],{},"Write three posts and send them on day five, not day fifteen. They will be somewhat wrong, and that is the point: nothing extracts a real voice brief like a client reacting to concrete text. An hour of line-by-line feedback on three drafts is worth more than another discovery call.",[12,3082,3083],{},"Frame it explicitly as calibration so nobody thinks it is the finished standard. Ask for edits in the text rather than comments about the text — how they rewrite a sentence tells you more than what they say about it.",[19,3085,3087],{"id":3086},"day-6-the-plan-one-page","Day 6 — The plan, one page",[12,3089,3090],{},"Now that you have heard them react, write the first month:",[27,3092,3093,3099,3105,3111],{},[30,3094,3095,3098],{},[34,3096,3097],{},"Cadence per channel",", honestly scoped to what you were paid for.",[30,3100,3101,3104],{},[34,3102,3103],{},"Three to five recurring content types"," you will run. Named, so both sides can talk about them.",[30,3106,3107,3110],{},[34,3108,3109],{},"What you will measure",", and what you will not. Agree now that you are not reporting on follower count if follower count is not the goal.",[30,3112,3113,3116],{},[34,3114,3115],{},"The one thing to prove in month one."," A single objective beats a list.",[12,3118,3119],{},"Send it for approval as a plan, not a proposal. It is the thing you will point at in month three when the request arrives to do something entirely different.",[19,3121,3123],{"id":3122},"day-7-set-the-standing-rhythm","Day 7 — Set the standing rhythm",[12,3125,3126],{},"The engagement needs a metabolism, agreed now rather than negotiated weekly:",[27,3128,3129,3135,3141],{},[30,3130,3131,3134],{},[34,3132,3133],{},"A weekly material drop."," Five minutes from the client: what happened, what launched, what a customer said. Without this you will be writing from the website by month two, and it will show.",[30,3136,3137,3140],{},[34,3138,3139],{},"A fixed approval window."," Same day each week, both sides know it.",[30,3142,3143,3146],{},[34,3144,3145],{},"A monthly review",", thirty minutes, on what worked and what changes.",[12,3148,3149],{},"Then start. Do not extend onboarding into week three chasing polish; the remaining unknowns resolve faster through published work than through more meetings.",[19,3151,3153],{"id":3152},"the-one-page-checklist","The one-page checklist",[531,3155,3156,3169],{},[534,3157,3158],{},[537,3159,3160,3163,3166],{},[540,3161,3162],{},"Day",[540,3164,3165],{},"Deliverable",[540,3167,3168],{},"Owner",[547,3170,3171,3182,3193,3204,3214,3224,3234],{},[537,3172,3173,3176,3179],{},[552,3174,3175],{},"1",[552,3177,3178],{},"Access granted, assets and existing writing sent",[552,3180,3181],{},"Client",[537,3183,3184,3187,3190],{},[552,3185,3186],{},"2",[552,3188,3189],{},"Recorded voice interview",[552,3191,3192],{},"Both",[537,3194,3195,3198,3201],{},[552,3196,3197],{},"3",[552,3199,3200],{},"Voice definition + facts and forbidden list",[552,3202,3203],{},"You",[537,3205,3206,3209,3212],{},[552,3207,3208],{},"4",[552,3210,3211],{},"Approver named, silence rule agreed",[552,3213,3192],{},[537,3215,3216,3219,3222],{},[552,3217,3218],{},"5",[552,3220,3221],{},"Three calibration drafts sent and marked up",[552,3223,3192],{},[537,3225,3226,3229,3232],{},[552,3227,3228],{},"6",[552,3230,3231],{},"One-page month-one plan approved",[552,3233,3192],{},[537,3235,3236,3239,3242],{},[552,3237,3238],{},"7",[552,3240,3241],{},"Weekly rhythm scheduled",[552,3243,3192],{},[19,3245,3247],{"id":3246},"where-this-stops-scaling","Where this stops scaling",[12,3249,3250],{},"At one or two clients this runs on notes and attention. At five, the artefacts multiply — five voice definitions, five fact lists, five forbidden lists, five approvers, five silence rules — and they stop living anywhere reliable. That is the point where onboarding quality quietly degrades and every new client gets a slightly worse version of the process.",[12,3252,3253,3255],{},[71,3254,7],{"href":733}," is built around that: voice, facts, constraints and approval trail stored per brand, so onboarding a sixth client is the same process as the first rather than one more thing to hold in your head. The client approves through a link scoped to their brand and never sees the rest of your book.",[19,3257,210],{"id":209},[12,3259,3260,3263],{},[34,3261,3262],{},"What should I ask a new social media client?","\nAsk for three posts they liked and three they disliked, who specifically they are talking to, what they refuse to say, what is true that competitors cannot claim, and what went wrong with their last agency. Specifics beat adjectives — \"professional but approachable\" is not a brief.",[12,3265,3266,3269],{},[34,3267,3268],{},"How long should onboarding a social media client take?","\nAbout a week of elapsed time, with the first calibration drafts sent by day five. Longer onboarding rarely produces better output; the remaining unknowns resolve faster through real drafts than through additional discovery calls.",[12,3271,3272,3275],{},[34,3273,3274],{},"What access do I need from a social media client?","\nRole-based publishing access on each channel (never a shared password), read-only analytics, brand assets, everything they have already written, and one named person who can answer questions same-day.",[12,3277,3278,3281],{},[34,3279,3280],{},"How do I stop client approvals from dragging?","\nName one approver with a tiebreak, agree in writing what happens when nobody responds by the deadline, and get approval on the week's angles before drafting rather than on finished posts afterwards.",[12,3283,3284,3287],{},[34,3285,3286],{},"What should be in a brand voice document?","\nThree real sample posts, first person or brand, the words they use and the words they refuse, one line on stance, and a separate list of factual claims they may make with the dates those numbers were true.",{"title":247,"searchDepth":248,"depth":248,"links":3289},[3290,3291,3292,3293,3294,3295,3296,3297,3298,3299],{"id":2917,"depth":248,"text":2918},{"id":2959,"depth":248,"text":2960},{"id":3009,"depth":248,"text":3010},{"id":3048,"depth":248,"text":3049},{"id":3076,"depth":248,"text":3077},{"id":3086,"depth":248,"text":3087},{"id":3122,"depth":248,"text":3123},{"id":3152,"depth":248,"text":3153},{"id":3246,"depth":248,"text":3247},{"id":209,"depth":248,"text":210},"A day-by-day onboarding for a new SMM client: what to collect, the questions that actually determine the voice, the access checklist, and the two agreements that prevent every later argument.",{},{"title":2906,"description":3300},"blog\u002Fsocial-media-client-onboarding",[805,505],"HCdsz15KVQbx5wJsiHPCoEuvhyQLo_eN4OWApyZeRIg",{"id":3307,"title":3308,"author":7,"body":3309,"date":1567,"description":3509,"draft":259,"extension":260,"image":261,"meta":3510,"navigation":263,"path":2411,"seo":3511,"stem":3512,"tags":3513,"updated":261,"__hash__":3514},"blog\u002Fblog\u002Fwhat-to-post-before-launch.md","What to post before you have a product",{"type":9,"value":3310,"toc":3500},[3311,3314,3317,3320,3324,3327,3353,3360,3364,3370,3376,3382,3388,3394,3400,3406,3409,3413,3416,3419,3425,3431,3435,3438,3445,3449,3452,3455,3459,3462,3468,3470,3476,3482,3488,3494],[12,3312,3313],{},"The pre-launch founder has the hardest content problem and the best raw material, and almost always uses neither. No customers means no case studies, no numbers, no social proof — so most people fill the gap with industry takes that could have been written by anyone, or with traction language that is not quite true yet.",[12,3315,3316],{},"Both are worse than posting nothing. The takes are invisible; the inflated claims are remembered.",[12,3318,3319],{},"What you do have is the thing nobody with a mature product has any more: an unfinished decision, in public, in real time. That is the whole strategy.",[19,3321,3323],{"id":3322},"the-claims-that-cost-you-later","The claims that cost you later",[12,3325,3326],{},"Start with what not to write, because pre-launch is where founders damage trust fastest and least deliberately.",[27,3328,3329,3335,3341,3347],{},[30,3330,3331,3334],{},[34,3332,3333],{},"\"Our customers are seeing…\""," when there are three design partners on a free tier. The plural is the lie, and someone will ask.",[30,3336,3337,3340],{},[34,3338,3339],{},"\"We're growing fast\""," off a base of 40 signups. Not false, exactly, and everyone reads it as false.",[30,3342,3343,3346],{},[34,3344,3345],{},"Invented specificity."," A generated post will happily produce \"teams save an average of 6 hours a week\" because it is the shape a post takes. If you did not measure it, that number becomes something you have to defend or quietly delete.",[30,3348,3349,3352],{},[34,3350,3351],{},"Borrowed authority."," Writing about scaling when you have not scaled anything. Readers who have done it can tell in one paragraph.",[12,3354,3355,3356,3359],{},"The rule that survives all of these: ",[34,3357,3358],{},"never claim a scale you cannot name."," \"One beta customer cut migration from four hours to twenty minutes\" is stronger than \"customers save hours\" precisely because it is narrow. Specific and small beats vague and big, every time.",[19,3361,3363],{"id":3362},"seven-angles-that-work-with-zero-customers","Seven angles that work with zero customers",[12,3365,3366,3369],{},[34,3367,3368],{},"1. The problem, described better than your buyer can describe it."," If you can articulate the pain more precisely than the person living with it, you have earned attention without claiming anything. This is the single highest-value pre-launch post and most founders skip straight past it to the solution.",[12,3371,3372,3375],{},[34,3373,3374],{},"2. The decision you are in the middle of."," Not \"here's what we built\" but \"here's the fork we're at and what we're weighing.\" Live decisions get replies from people who have made them. Finished decisions get likes.",[12,3377,3378,3381],{},[34,3379,3380],{},"3. What you're building and why this way."," The architecture, the constraint you accepted, the thing you deliberately made worse. Technical honesty travels.",[12,3383,3384,3387],{},[34,3385,3386],{},"4. What you were wrong about."," You have been wrong about something in the last month. That post is more interesting than your roadmap and costs you nothing, because pre-launch nobody expected you to be right yet.",[12,3389,3390,3393],{},[34,3391,3392],{},"5. Research in public."," You are talking to potential users. Post what you are hearing, anonymised, without conclusions. \"Six of ten people we spoke to had already built a spreadsheet version of this\" is a fact you own.",[12,3395,3396,3399],{},[34,3397,3398],{},"6. The thing you refuse to build."," Positioning stated as a boundary is memorable and commits you publicly, which is useful discipline.",[12,3401,3402,3405],{},[34,3403,3404],{},"7. Work in progress with the seams showing."," A screenshot of something half-finished, a migration that broke, a metric that is bad. Build-in-public works when the \"in public\" part includes the parts that are not going well.",[12,3407,3408],{},"Notice that six of seven require no traction at all. They require you to be doing the work and willing to describe it accurately.",[19,3410,3412],{"id":3411},"build-the-audience-the-launch-lands-on","Build the audience the launch lands on",[12,3414,3415],{},"The reason to post pre-launch is not the posts. It is that a launch to zero people is a press release into a void, and a launch to 800 people who have watched you reason for six months is an event.",[12,3417,3418],{},"Two things matter more than volume:",[12,3420,3421,3424],{},[34,3422,3423],{},"Reply to people, individually, before you need anything."," Pre-launch is the only period where you have time for this, and it is the only distribution strategy that reliably works from a standing start. Twenty real conversations beat two hundred impressions.",[12,3426,3427,3430],{},[34,3428,3429],{},"Say what you are building, repeatedly, in one sentence."," Not a pitch. A sentence people can repeat to someone else. If your audience cannot describe what you do, they cannot refer you at launch, and referral is most of what a launch is.",[19,3432,3434],{"id":3433},"the-frequency-question","The frequency question",[12,3436,3437],{},"Lower than you think. Two posts a week for six months beats daily for five weeks and then silence, which is the standard pre-launch pattern: enthusiasm, exhaustion, disappearance a month before the thing you were building for.",[12,3439,3440,3441,3444],{},"Pre-launch is a stamina problem, not an intensity problem. The ",[71,3442,3443],{"href":2072},"30-minutes-a-week method"," exists precisely because the pre-launch period is long and the founder's attention is not.",[19,3446,3448],{"id":3447},"when-the-product-ships","When the product ships",[12,3450,3451],{},"Do not throw away the pre-launch archive. The decisions, the wrong turns and the research are the raw material for everything you write afterwards — a launch post that references six months of public reasoning is dramatically stronger than one that starts from \"today we're excited to announce.\"",[12,3453,3454],{},"This is also where most founders lose the plot: the voice that got attention pre-launch was specific, uncertain and technical, and the launch flips it to marketing register overnight. Readers notice the switch immediately. Whatever else changes, keep the register.",[19,3456,3458],{"id":3457},"where-a-tool-helps-and-where-it-does-not","Where a tool helps and where it does not",[12,3460,3461],{},"No tool solves the pre-launch content problem, because the problem is that only you know what happened this week. What a tool can do is stop the two failure modes: going dark when the build gets hard, and letting a generated draft quietly invent traction you do not have.",[12,3463,3464,3465,3467],{},"That second one is why ",[71,3466,7],{"href":181}," works the way it does — brand memory holds what is actually true, including \"customers: 0\", and a draft that would claim otherwise gets stopped rather than written. Pre-launch is exactly the situation where an AI writer is most dangerous and most tempting, because the gap between what you have and what a post wants is at its widest.",[19,3469,210],{"id":209},[12,3471,3472,3475],{},[34,3473,3474],{},"What should a startup post before launch?","\nThe problem stated more precisely than your buyer can state it, decisions you are currently weighing, what you are building and why this way, things you were wrong about, and anonymised findings from user research. All of these work with zero customers.",[12,3477,3478,3481],{},[34,3479,3480],{},"How do I build an audience with no product?","\nPost twice a week for months rather than daily for weeks, and spend more time replying to individuals than writing posts. Pre-launch is the only period where you have time for one-to-one conversations, and they convert into a launch audience far better than impressions do.",[12,3483,3484,3487],{},[34,3485,3486],{},"Should I build in public?","\nIt works when \"in public\" includes the parts going badly. Selective build-in-public that only shows wins reads as marketing and gets treated as such. If you are not willing to post the broken migration, the format is not for you.",[12,3489,3490,3493],{},[34,3491,3492],{},"Is it bad to post about traction I don't have yet?","\nYes, and the damage is durable. Never claim a scale you cannot name. One named result stated narrowly is more persuasive than a vague plural, and it does not become something you have to defend later.",[12,3495,3496,3499],{},[34,3497,3498],{},"How often should a pre-launch founder post?","\nTwo to three times a week, sustained. The pre-launch period is typically six to eighteen months, which is a stamina problem — the common failure is intense posting for a month followed by silence right before launch.",{"title":247,"searchDepth":248,"depth":248,"links":3501},[3502,3503,3504,3505,3506,3507,3508],{"id":3322,"depth":248,"text":3323},{"id":3362,"depth":248,"text":3363},{"id":3411,"depth":248,"text":3412},{"id":3433,"depth":248,"text":3434},{"id":3447,"depth":248,"text":3448},{"id":3457,"depth":248,"text":3458},{"id":209,"depth":248,"text":210},"Pre-launch content without fake traction: seven angles that work with zero customers, the claims that quietly destroy trust, and how to build an audience the launch can land on.",{},{"title":3308,"description":3509},"blog\u002Fwhat-to-post-before-launch",[268,806],"BSnR1nRUYwbsZ1GSrvhdK1ruP7msgCwyyzC96eDms3M",{"id":3516,"title":3517,"author":7,"body":3518,"date":1567,"description":3737,"draft":259,"extension":260,"image":261,"meta":3738,"navigation":263,"path":361,"seo":3739,"stem":3740,"tags":3741,"updated":261,"__hash__":3742},"blog\u002Fblog\u002Fwho-is-responsible-for-a-wrong-claim.md","Who is responsible when a published claim turns out to be wrong",{"type":9,"value":3519,"toc":3729},[3520,3523,3529,3532,3536,3539,3545,3551,3557,3563,3570,3574,3577,3583,3589,3595,3602,3606,3609,3612,3635,3641,3645,3648,3680,3683,3687,3690,3697,3699,3705,3711,3717,3723],[12,3521,3522],{},"A post goes out on a client's account. It says the product cuts onboarding time by 60%. Nobody at the agency invented that number on purpose — it came from a deck, or a draft, or a model filling a gap in a sentence that wanted a figure. A competitor's marketing lead screenshots it. The client's legal contact asks where it came from.",[12,3524,3525,3526],{},"Now the question arrives, and it arrives at the worst possible time: ",[34,3527,3528],{},"whose fault is this?",[12,3530,3531],{},"Almost every agency answers it for the first time while it is happening. That is the mistake. The answer is cheap to write down in advance and extremely expensive to negotiate under pressure.",[19,3533,3535],{"id":3534},"the-four-places-a-claim-breaks","The four places a claim breaks",[12,3537,3538],{},"Wrong claims do not come from one failure. They come from four, and they need different fixes.",[12,3540,3541,3544],{},[34,3542,3543],{},"1. The claim was never true."," Someone made it up, or a generated draft produced a plausible number because the sentence wanted one. This is the AI-era failure mode and it is genuinely new: a model has no way to know it invented a figure, and neither does the person pasting it, because it reads exactly like the ones that are real.",[12,3546,3547,3550],{},[34,3548,3549],{},"2. The claim was true once."," \"Trusted by 400 teams\" was accurate in March. It is now in a template. Nobody re-checked, because nothing marked it as perishable. Undated facts go wrong silently, which makes this the most common category by a wide margin.",[12,3552,3553,3556],{},[34,3554,3555],{},"3. The claim was true but not permitted."," The number is real; the client's legal or compliance team had not cleared it for public use, or it belonged to a customer who never agreed to be named.",[12,3558,3559,3562],{},[34,3560,3561],{},"4. The claim was fine and the scope was not."," One beta customer's result stated as \"our customers see.\" The data is real, the generalisation is not. This is the most defensible-feeling error and often the most damaging, because it is deliberate rather than accidental.",[12,3564,3565,3566,3569],{},"Notice that only the first is about invention. The other three are about ",[34,3567,3568],{},"process",": dating, permission and scope.",[19,3571,3573],{"id":3572},"assign-responsibility-before-it-happens","Assign responsibility before it happens",[12,3575,3576],{},"Three lines in the contract or the onboarding doc, agreed once, that decide the argument in advance.",[12,3578,3579,3582],{},[34,3580,3581],{},"The client owns the truth of facts about their business."," They are the only ones who can. The agency cannot verify a retention number, a customer count or whether a case study is cleared. What the agency owes is that no factual claim appears in a published post unless the client supplied it or approved it.",[12,3584,3585,3588],{},[34,3586,3587],{},"The agency owns that nothing unsourced ships."," This is the real obligation and it is a process commitment, not a knowledge one: every factual claim in a draft traces to something the client provided. If it cannot be traced, it does not go out. Missing a source is a blocker, not a style note.",[12,3590,3591,3594],{},[34,3592,3593],{},"Approval is a decision with a name on it."," Whoever approved the post approved the claims in it. Not \"marketing saw it\" — a person, a timestamp, a version. Without that, responsibility is a memory contest, and memory contests end relationships.",[12,3596,3597,3598,3601],{},"Add one operational rule that prevents most of category two: ",[34,3599,3600],{},"every number carries the date it was true."," Facts stored without dates rot invisibly. Facts stored with dates announce when they need re-checking.",[19,3603,3605],{"id":3604},"the-mechanism-that-actually-works","The mechanism that actually works",[12,3607,3608],{},"Policies do not stop wrong claims. The moment of failure is a writer at 4pm with a sentence that needs a number, and no policy is present in that moment.",[12,3610,3611],{},"What works is making the unsourced version harder to produce than the sourced one:",[27,3613,3614,3620,3629],{},[30,3615,3616,3619],{},[34,3617,3618],{},"Keep a per-client fact file"," — claims permitted, with dates and sources — that sits where drafting happens rather than in a folder nobody opens.",[30,3621,3622,3625,3626,3628],{},[34,3623,3624],{},"Keep a forbidden list"," alongside it: claims that may never be made, comparisons that may not be drawn, customers that may not be named. Collect it during ",[71,3627,925],{"href":696},", on day one.",[30,3630,3631,3634],{},[34,3632,3633],{},"Treat a missing fact as a stop, not a gap."," The instinct is to write around it with something softer, and softer usually means vaguer and larger — \"customers save hours\" is a worse claim than the specific one it replaced, because it silently generalises.",[12,3636,3637,3638,3640],{},"That third point is the one AI-assisted workflows get wrong by default. A general-purpose model will always produce something rather than refuse, because producing text is what it is for. That behaviour is fine for a first draft of an essay and unacceptable on a client's account, which is why ",[71,3639,7],{"href":422}," is built to stop instead: without a supporting fact it narrows the claim or refuses to draft and asks for evidence. The unsupported version never exists to be pasted by accident.",[19,3642,3644],{"id":3643},"when-it-happens-anyway","When it happens anyway",[12,3646,3647],{},"It will. The response matters more than the prevention, and speed is most of it.",[439,3649,3650,3656,3662,3668,3674],{},[30,3651,3652,3655],{},[34,3653,3654],{},"Correct or delete within the hour."," Deleting is not an admission of anything; leaving it up while you decide is.",[30,3657,3658,3661],{},[34,3659,3660],{},"Tell the client before they find it."," Every hour they discover it independently costs more than the error did.",[30,3663,3664,3667],{},[34,3665,3666],{},"Say which of the four categories it was."," Invented, stale, unpermitted or over-scoped. Each has a different fix and a different owner, and naming it stops the conversation from becoming about competence in general.",[30,3669,3670,3673],{},[34,3671,3672],{},"Show the trail."," Who drafted, what it was based on, who approved. Not to deflect — to move the discussion from \"how did this happen\" to \"which step do we change.\"",[30,3675,3676,3679],{},[34,3677,3678],{},"Change one step, visibly."," A named process change is what converts an incident into an argument for keeping you.",[12,3681,3682],{},"Agencies that survive these have a record. Agencies that do not have a Slack thread and three people's recollections.",[19,3684,3686],{"id":3685},"why-this-gets-worse-at-scale","Why this gets worse at scale",[12,3688,3689],{},"One client, one writer, one approver — the whole thing fits in someone's head. At eight clients with juniors drafting, the facts live in eight different places, half of them are stale, and the person writing has no fast way to check whether a number is current, permitted or scoped correctly.",[12,3691,3692,3693,3696],{},"That is the actual argument for putting brand facts, permissions and the approval trail on the work itself rather than in documents: not tidiness, but that the safe path has to be the fast path at 4pm. If checking is slower than guessing, people guess. ",[71,3694,3695],{"href":350},"Roles and release permissions"," are the other half of the same problem.",[19,3698,210],{"id":209},[12,3700,3701,3704],{},[34,3702,3703],{},"Who is liable if an agency publishes a false claim about a client?","\nContractually it depends on what was agreed, which is why it should be agreed in advance: the standard split is that the client owns the truth of facts about their business, the agency owns that nothing unsourced ships, and whoever approved a post approved the claims in it.",[12,3706,3707,3710],{},[34,3708,3709],{},"How do agencies prevent AI-generated posts from inventing facts?","\nBy keeping a per-client file of permitted claims with dates and sources, treating a missing fact as a blocker rather than something to write around, and using tooling that refuses to draft an unsupported claim instead of producing a plausible one. General-purpose models always produce something, because producing text is what they do.",[12,3712,3713,3716],{},[34,3714,3715],{},"What should be in a client fact file?","\nNumbers you may cite with the date each was true, customers you may name and how, claims requiring legal review before publishing, and a forbidden list of claims and comparisons that may never appear.",[12,3718,3719,3722],{},[34,3720,3721],{},"What do you do when a wrong post is already live?","\nCorrect or delete within the hour, tell the client before they discover it, name which category of failure it was, show the drafting and approval trail, and change one named process step. Speed and specificity matter more than the apology.",[12,3724,3725,3728],{},[34,3726,3727],{},"Is \"our customers see X\" safe if one customer saw X?","\nNo. Generalising one result to a plural is the most damaging category precisely because it feels defensible. State the claim as narrowly as the evidence supports — one named result is more persuasive than a vague plural anyway.",{"title":247,"searchDepth":248,"depth":248,"links":3730},[3731,3732,3733,3734,3735,3736],{"id":3534,"depth":248,"text":3535},{"id":3572,"depth":248,"text":3573},{"id":3604,"depth":248,"text":3605},{"id":3643,"depth":248,"text":3644},{"id":3685,"depth":248,"text":3686},{"id":209,"depth":248,"text":210},"A post goes out with a number nobody verified. How agencies should assign responsibility before it happens, the four places claims break, and the record that settles it afterwards.",{},{"title":3517,"description":3737},"blog\u002Fwho-is-responsible-for-a-wrong-claim",[505,805],"_lV0_77ds0BbJryIokQ7fX52CRs2snzpNwr0-cGVNVI",{"id":3744,"title":3745,"author":7,"body":3746,"date":3941,"description":3942,"draft":259,"extension":260,"image":261,"meta":3943,"navigation":263,"path":173,"seo":3944,"stem":3945,"tags":3946,"updated":261,"__hash__":3947},"blog\u002Fblog\u002Fchatgpt-claude-for-social-media.md","Why not just use ChatGPT or Claude for your social media?",{"type":9,"value":3747,"toc":3932},[3748,3751,3758,3762,3765,3768,3772,3775,3786,3789,3793,3796,3828,3831,3835,3838,3842,3845,3877,3881,3884,3898,3901,3906,3908,3914,3920,3926],[12,3749,3750],{},"It's a fair question. If you already pay for ChatGPT or Claude, why pay for anything else to write posts? They can obviously write a caption, a thread, a hook.",[12,3752,3753,3754,3757],{},"The honest answer: for one account and the occasional post, a chatbot might be all you need. The gap shows up the moment social becomes a ",[519,3755,3756],{},"job",", with several brands, a steady cadence, someone approving the work, and content that can't read as machine-made. That's the difference between \"a chatbot that can write a post\" and \"a system that ships on-brand content.\" Here's where the line falls.",[19,3759,3761],{"id":3760},"youve-seen-this-pattern-before","You've seen this pattern before",[12,3763,3764],{},"Claude writes excellent code, and developers still pay for Cursor. ChatGPT generates images, and people still subscribe to Leonardo and Midjourney. The general model is the engine. The tool built around it is what you actually work in, because the engine alone hands you every other part of the job to do yourself.",[12,3766,3767],{},"Social content is the same shape. A chatbot can write a post. Turning that into on-brand content that sounds human and gets shipped, for several brands, week after week, is the tool's job. SelfSM is that tool, and the rest of this page is what \"the tool's job\" covers.",[19,3769,3771],{"id":3770},"what-chatgpt-and-claude-are-genuinely-great-at","What ChatGPT and Claude are genuinely great at",[12,3773,3774],{},"No sleight of hand: general LLMs are excellent tools.",[27,3776,3777,3780,3783],{},[30,3778,3779],{},"Brainstorming angles and hooks when you're stuck.",[30,3781,3782],{},"Turning a rough idea into a first draft in seconds.",[30,3784,3785],{},"Rewriting, shortening, changing tone on demand.",[12,3787,3788],{},"If drafting is your only bottleneck and you enjoy prompting, that alone is real value. Keep using them.",[19,3790,3792],{"id":3791},"where-a-raw-chatbot-leaves-the-work-to-you","Where a raw chatbot leaves the work to you",[12,3794,3795],{},"The writing is the easy part. The work is everything around it.",[27,3797,3798,3804,3810,3816,3822],{},[30,3799,3800,3803],{},[34,3801,3802],{},"You re-explain the voice every time."," A fresh chat doesn't know your brand. You paste the same context, the same rules, the same \"don't say this,\" every session, for every brand.",[30,3805,3806,3809],{},[34,3807,3808],{},"The output has tells."," Even good LLM writing carries the patterns people now recognize as AI: the tidy three-point structure, the puffed-up adjectives, the \"it's not X, it's Y\" reflex. Readers pattern-match those to \"generated\" and scroll past.",[30,3811,3812,3815],{},[34,3813,3814],{},"One brain, one context."," Juggling five clients in a chatbot is copy-paste chaos: five prompt setups, five threads, easy to mix up whose voice is whose.",[30,3817,3818,3821],{},[34,3819,3820],{},"It stops at text."," A chatbot hands you words. You still schedule, get approval, publish, track it, and go find the conversations worth replying to. None of that is in the box.",[30,3823,3824,3827],{},[34,3825,3826],{},"It forgets."," What you approved last week, what flopped, the phrase the client hates: gone by the next chat.",[12,3829,3830],{},"So the \"free\" draft isn't free. You pay for it in the manual work of turning a paragraph of text into published, on-brand, human-sounding content, over and over.",[19,3832,3834],{"id":3833},"but-i-set-up-a-custom-gpt-a-project","\"But I set up a Custom GPT \u002F a Project\"",[12,3836,3837],{},"Fair. You can load brand context into a Custom GPT or a Claude Project and get closer. It helps. But it's still you doing the systems work by hand: one setup per brand, maintained manually, with no scheduling, no approvals, no publishing, and no sense of what actually worked. It closes part of the drafting gap and none of the workflow gap. Past a brand or two, the upkeep becomes its own job.",[19,3839,3841],{"id":3840},"what-selfsm-adds-on-top","What SelfSM adds on top",[12,3843,3844],{},"SelfSM is AI-powered too. The difference is a chatbot versus a system built for this specific job:",[27,3846,3847,3853,3859,3865,3871],{},[30,3848,3849,3852],{},[34,3850,3851],{},"A generation pipeline, not a single prompt."," SelfSM drafts, then runs the output through checks that strip the tells that make content read as AI, and it learns from the material and the corrections you give it. Each brand's drafts get closer to right over time instead of resetting every session.",[30,3854,3855,3858],{},[34,3856,3857],{},"Voice that persists, per brand."," Each brand's tone, facts, and banned phrases are stored and applied automatically. No re-explaining, no mixing up clients.",[30,3860,3861,3864],{},[34,3862,3863],{},"Chloe, tuned as a brand manager and ghostwriter."," An assistant that guides the work end to end: she asks for the material she needs, drafts in the brand's voice, explains why a post is in review, and moves it from idea toward published. You get a guide who leads you through it, not a blank chat box.",[30,3866,3867,3870],{},[34,3868,3869],{},"A full workflow, not just a draft."," Idea → on-brand draft → approval → schedule → publish → replies. It carries the post the whole way, not just to the first paragraph of text.",[30,3872,3873,3876],{},[34,3874,3875],{},"It finds the replies."," The conversations worth joining get surfaced, not left to you to scroll for.",[19,3878,3880],{"id":3879},"do-you-need-it","Do you need it?",[12,3882,3883],{},"Be honest with yourself about the job.",[27,3885,3886,3892],{},[30,3887,3888,3891],{},[34,3889,3890],{},"One account, posting when you feel like it, and you like prompting?"," A chatbot is probably enough. Genuinely.",[30,3893,3894,3897],{},[34,3895,3896],{},"Running social for several brands or clients, on a schedule, that has to sound human and actually get shipped and approved?"," That's a workflow problem, and a chatbot doesn't solve it. That's the gap SelfSM fills.",[12,3899,3900],{},"You don't have to give up your Claude or ChatGPT subscription to use SelfSM. They're great for thinking out loud. SelfSM is what turns that thinking into on-brand content that ships, for every brand you manage.",[12,3902,3903,3904,239],{},"Comparing SelfSM to scheduling tools instead? See ",[71,3905,430],{"href":429},[19,3907,210],{"id":209},[12,3909,3910,3913],{},[34,3911,3912],{},"Can't I just use ChatGPT for social media?","\nFor a single account, often yes. The moment you're managing multiple brands, need approvals, or need content that reliably doesn't sound like AI, a chatbot leaves most of the job to you.",[12,3915,3916,3919],{},[34,3917,3918],{},"I already pay for Claude. Is SelfSM redundant?","\nNo, they solve different problems. Claude is a brilliant general assistant. SelfSM is the system around producing and shipping social content for brands: stored voice per brand, human-sounding output by design, scheduling, approvals, and reply discovery.",[12,3921,3922,3925],{},[34,3923,3924],{},"Does SelfSM just wrap ChatGPT?","\nSelfSM is AI-powered, but the value isn't which model writes a first line. It's the brand-tuning, the work to make output sound human, and the workflow that takes a post from idea to published. That's what a raw chatbot doesn't give you.",[12,3927,3928,3931],{},[34,3929,3930],{},"What do you mean by content that \"sounds like AI\"?","\nRecognizable patterns: formulaic hooks, filler adjectives, over-tidy structure, the \"not X, but Y\" construction. Once readers spot them, they tune out. SelfSM is built to remove them.",{"title":247,"searchDepth":248,"depth":248,"links":3933},[3934,3935,3936,3937,3938,3939,3940],{"id":3760,"depth":248,"text":3761},{"id":3770,"depth":248,"text":3771},{"id":3791,"depth":248,"text":3792},{"id":3833,"depth":248,"text":3834},{"id":3840,"depth":248,"text":3841},{"id":3879,"depth":248,"text":3880},{"id":209,"depth":248,"text":210},"2026-08-03","You already pay for a chatbot, so why a dedicated tool? An honest look at where ChatGPT and Claude stop, and what a content workflow adds on top.",{},{"title":3745,"description":3942},"blog\u002Fchatgpt-claude-for-social-media",[503,269],"LyfVOHU_P9l1Up0KIoQzLvH13Lwc25Qxbm2-TdgXjyw",{"id":3949,"title":3950,"author":7,"body":3951,"date":3941,"description":4273,"draft":259,"extension":260,"image":261,"meta":4274,"navigation":263,"path":429,"seo":4275,"stem":4276,"tags":4277,"updated":261,"__hash__":4278},"blog\u002Fblog\u002Fselfsm-vs-buffer-hootsuite-typefully.md","SelfSM vs Buffer vs Hootsuite vs Typefully: which fits you?",{"type":9,"value":3952,"toc":4267},[3953,3956,3982,3985,3989,4157,4161,4167,4173,4179,4185,4189,4192,4218,4229,4235,4241,4243,4249,4255,4261],[12,3954,3955],{},"Short answer, if you're in a hurry. The question is which job you're hiring the tool for:",[27,3957,3958,3964,3970,3976],{},[30,3959,3960,3963],{},[34,3961,3962],{},"Buffer:"," the simplest way to queue and schedule posts across a few accounts. Best for solo creators and small brands who mostly need a reliable queue.",[30,3965,3966,3969],{},[34,3967,3968],{},"Hootsuite:"," the broad, all-in-one dashboard built for large teams, with many accounts, seats, listening, and reporting. Powerful, and heavier to learn.",[30,3971,3972,3975],{},[34,3973,3974],{},"Typefully:"," writing-first, built around X (Twitter) and threads. Best when X is your main channel and drafting is the part you care about.",[30,3977,3978,3981],{},[34,3979,3980],{},"SelfSM:"," a content workflow for people who run social for others: ghostwriters, agencies, founders, and SMM teams. Drafts in each brand's voice, keeps content sounding human, surfaces replies worth making, clears approvals, and ships the calendar.",[12,3983,3984],{},"They all let you \"schedule a post.\" They're built for different jobs. Here's how to tell which one is yours.",[19,3986,3988],{"id":3987},"quick-comparison","Quick comparison",[531,3990,3991,4016],{},[534,3992,3993],{},[537,3994,3995,3997,4001,4006,4011],{},[540,3996],{},[540,3998,3999],{},[34,4000,7],{},[540,4002,4003],{},[34,4004,4005],{},"Buffer",[540,4007,4008],{},[34,4009,4010],{},"Hootsuite",[540,4012,4013],{},[34,4014,4015],{},"Typefully",[547,4017,4018,4037,4056,4074,4091,4106,4122,4140],{},[537,4019,4020,4025,4028,4031,4034],{},[552,4021,4022],{},[34,4023,4024],{},"Best for",[552,4026,4027],{},"Running social for multiple brands\u002Fclients",[552,4029,4030],{},"Solo creators, small brands",[552,4032,4033],{},"Large teams & enterprises",[552,4035,4036],{},"X\u002FTwitter-first writers",[537,4038,4039,4044,4047,4050,4053],{},[552,4040,4041],{},[34,4042,4043],{},"The core job",[552,4045,4046],{},"Produce + ship on-brand content",[552,4048,4049],{},"Schedule posts",[552,4051,4052],{},"All-in-one management at scale",[552,4054,4055],{},"Write X posts & threads",[537,4057,4058,4063,4066,4069,4072],{},[552,4059,4060],{},[34,4061,4062],{},"Multi-brand \u002F client management",[552,4064,4065],{},"Core focus",[552,4067,4068],{},"Limited",[552,4070,4071],{},"Yes, at team\u002Fenterprise scale",[552,4073,4068],{},[537,4075,4076,4081,4083,4086,4088],{},[552,4077,4078],{},[34,4079,4080],{},"AI drafting in each brand's voice",[552,4082,2535],{},[552,4084,4085],{},"Basic AI assist",[552,4087,4085],{},[552,4089,4090],{},"Writing-focused",[537,4092,4093,4098,4100,4102,4104],{},[552,4094,4095],{},[34,4096,4097],{},"Keeps content sounding human (anti-AI)",[552,4099,4065],{},[552,4101,2571],{},[552,4103,2571],{},[552,4105,2571],{},[537,4107,4108,4113,4115,4117,4120],{},[552,4109,4110],{},[34,4111,4112],{},"Finds conversations worth replying to",[552,4114,2535],{},[552,4116,2571],{},[552,4118,4119],{},"Via listening (larger plans)",[552,4121,2571],{},[537,4123,4124,4129,4131,4134,4137],{},[552,4125,4126],{},[34,4127,4128],{},"Approval workflow",[552,4130,2535],{},[552,4132,4133],{},"Higher tiers",[552,4135,4136],{},"Yes (team plans)",[552,4138,4139],{},"Lighter",[537,4141,4142,4147,4150,4152,4154],{},[552,4143,4144],{},[34,4145,4146],{},"Primary platforms",[552,4148,4149],{},"Multi-platform",[552,4151,4149],{},[552,4153,4149],{},[552,4155,4156],{},"X-centric",[19,4158,4160],{"id":4159},"how-to-choose-by-situation","How to choose (by situation)",[12,4162,4163,4166],{},[34,4164,4165],{},"You post for yourself, a few channels, and want zero fuss."," Buffer is hard to beat. It does one thing well: scheduling. You probably don't need the rest.",[12,4168,4169,4172],{},[34,4170,4171],{},"You're a large team or enterprise"," that needs many seats, social listening, and heavy reporting across dozens of accounts. Hootsuite is built for that scale. Expect a learning curve that comes with the breadth.",[12,4174,4175,4178],{},[34,4176,4177],{},"X (Twitter) is your whole game"," and you live in threads. Typefully's drafting experience is purpose-built for exactly that, and it's a pleasure to write in.",[12,4180,4181,4184],{},[34,4182,4183],{},"You run social for other people, as a ghostwriter, agency, or in-house SMM team."," This is where a scheduler stops being enough. The hard part is producing on-brand content at volume, keeping it from sounding like AI, getting it approved, and never missing the reply that matters. That's the job SelfSM is built for.",[19,4186,4188],{"id":4187},"where-selfsm-is-different","Where SelfSM is different",[12,4190,4191],{},"SelfSM is built for a different job than the schedulers: the post-for-others problem.",[27,4193,4194,4200,4206,4212],{},[30,4195,4196,4199],{},[34,4197,4198],{},"Every brand keeps its own voice."," Drafts are written in each brand's tone, not one generic house style. Running five clients doesn't mean five copies of the same voice.",[30,4201,4202,4205],{},[34,4203,4204],{},"It's built to sound human."," The whole product is oriented against the tells that make content read as AI-generated, because \"obviously generated\" posts get ignored, and that's the fastest way to lose a client's audience.",[30,4207,4208,4211],{},[34,4209,4210],{},"It surfaces replies, not just posts."," Finding the conversation worth joining is half of social; SelfSM points you at them instead of leaving you to scroll.",[30,4213,4214,4217],{},[34,4215,4216],{},"Approvals are part of the flow."," They're the review step that keeps a client relationship intact, built in rather than bolted on.",[12,4219,4220,4221,4224,4225,4228],{},"If your bottleneck is ",[519,4222,4223],{},"scheduling",", use a scheduler. If your bottleneck is ",[519,4226,4227],{},"producing good, on-brand, human-sounding content and getting it shipped across several brands",", that's the gap SelfSM fills.",[12,4230,4231,4232],{},"Wondering why not just use a chatbot you already pay for? See ",[71,4233,4234],{"href":173},"Why not just use ChatGPT or Claude for social media?",[12,4236,4237,4238,239],{},"If you are specifically replacing a full suite at agency scale, the criteria worth comparing are in ",[71,4239,4240],{"href":499},"choosing a Hootsuite alternative for an agency",[19,4242,210],{"id":209},[12,4244,4245,4248],{},[34,4246,4247],{},"Is SelfSM a Buffer alternative?","\nIt can replace scheduling, but that undersells it. Buffer is a scheduler; SelfSM is a content workflow for people managing social across multiple brands. If all you need is a queue, Buffer is simpler.",[12,4250,4251,4254],{},[34,4252,4253],{},"Do I still need a separate scheduler?","\nNo. SelfSM takes content from idea to scheduled and published. It replaces the scheduler, and does a lot more before the post reaches the queue.",[12,4256,4257,4260],{},[34,4258,4259],{},"I only post on X. Should I use Typefully instead?","\nIf X is your only channel and you mostly write threads, Typefully's focused experience is excellent. SelfSM makes more sense when you're managing multiple brands and\u002For multiple platforms.",[12,4262,4263,4266],{},[34,4264,4265],{},"What makes SelfSM's content \"sound human\"?","\nIt's designed to remove the patterns that make posts read as AI-generated: formulaic hooks, puffery, and mechanical symmetry. Content keeps the voice of the person or brand behind it.",{"title":247,"searchDepth":248,"depth":248,"links":4268},[4269,4270,4271,4272],{"id":3987,"depth":248,"text":3988},{"id":4159,"depth":248,"text":4160},{"id":4187,"depth":248,"text":4188},{"id":209,"depth":248,"text":210},"An honest, no-pricing comparison of four social media tools: who each is really for, why you'd pick it, and how it fits your workflow.",{},{"title":3950,"description":4273},"blog\u002Fselfsm-vs-buffer-hootsuite-typefully",[503,504],"ZeqaB0L3jTIyElvrQpEhAqM9UtzOeGGxHW0HWH1JYnQ",{"id":4280,"title":4281,"author":7,"body":4282,"date":3941,"description":4471,"draft":259,"extension":260,"image":261,"meta":4472,"navigation":263,"path":73,"seo":4473,"stem":4474,"tags":4475,"updated":261,"__hash__":4477},"blog\u002Fblog\u002Fwhy-ai-social-posts-get-ignored.md","Why AI social posts get ignored (and what to do instead)",{"type":9,"value":4283,"toc":4462},[4284,4287,4291,4294,4297,4301,4304,4336,4340,4343,4346,4350,4382,4386,4394,4397,4401,4404,4421,4425,4431,4437,4446,4452],[12,4285,4286],{},"Most feeds are now full of posts that technically say something and mean nothing. They open with a rhetorical question, stack three tidy bullet points, and close with \"What do you think?\" Readers scroll past them without deciding to. The brain files them as noise before the first line finishes. This is why that happens, the exact patterns behind it, and how to fix each one, with examples you can copy.",[19,4288,4290],{"id":4289},"recognizability-is-the-real-problem","Recognizability is the real problem",[12,4292,4293],{},"The issue isn't grammar, and it isn't effort. It's recognizability. Once a post pattern-matches to \"generated,\" a reader's trust drops and their attention leaves, often before they've consciously read a word. They're not judging the content on its merits; they're reacting to its shape. And the shapes that read as machine-made are now familiar enough that people filter them automatically, the way you tune out a banner ad without seeing it.",[12,4295,4296],{},"So the goal isn't to write \"better\" in some abstract sense. It's to not trip the pattern that makes a reader decide, in a second, that there's nothing here.",[19,4298,4300],{"id":4299},"the-tells-that-get-a-post-ignored","The tells that get a post ignored",[12,4302,4303],{},"Each of these is a shape people now associate with generated text. After each one is a quick fix.",[27,4305,4306,4312,4318,4324,4330],{},[30,4307,4308,4311],{},[34,4309,4310],{},"A hook that asks instead of says."," \"Ever wondered why some posts go viral?\" is a stall. Fix: make a statement they can't answer without reading on. \"The post that got us 500 followers took four minutes to write.\"",[30,4313,4314,4317],{},[34,4315,4316],{},"Suspicious symmetry."," Three points, each the same length, each starting the same way. Real thinking is lumpy. Fix: make it uneven. One long point, one short, cut the third if it's only there to make three.",[30,4319,4320,4323],{},[34,4321,4322],{},"Puffery."," \"Game-changing,\" \"powerful,\" \"seamless.\" Adjectives standing in for facts. Fix: replace the adjective with the specific. Not \"a powerful strategy\" but \"a strategy that cut our churn in half.\"",[30,4325,4326,4329],{},[34,4327,4328],{},"The contrast crutch."," \"It's not about working harder, it's about working smarter.\" Once you see it, you see it everywhere. Fix: cut it and state the point directly.",[30,4331,4332,4335],{},[34,4333,4334],{},"A closing that begs."," \"Thoughts?\" \"Agree?\" \"Drop a comment.\" A confident post lands its point and stops. Fix: end on a statement.",[19,4337,4339],{"id":4338},"why-generated-kills-engagement","Why \"generated\" kills engagement",[12,4341,4342],{},"Attention on social is a filter, not a fair reading. People scan the first line, and the shape of it tells them whether to invest. A generated shape signals low effort and low stakes, so the reader keeps moving. The content underneath might be fine. It never gets read, because the packaging already answered the only question that mattered: is a real person with something to say behind this?",[12,4344,4345],{},"That's also why chasing volume with generic posts backfires. Each one that reads as generated doesn't just underperform; it teaches your audience to skip your name.",[19,4347,4349],{"id":4348},"what-to-do-instead","What to do instead",[439,4351,4352,4358,4364,4370,4376],{},[30,4353,4354,4357],{},[34,4355,4356],{},"Lead with a specific."," A number, a name, a moment. \"Last Tuesday a client asked me to delete a post an hour after publishing\" beats any rhetorical question.",[30,4359,4360,4363],{},[34,4361,4362],{},"Let the structure be uneven."," One long point and two short ones reads more human than three matched bullets.",[30,4365,4366,4369],{},[34,4367,4368],{},"Cut the adjectives, keep the nouns."," Show the thing; don't label it impressive.",[30,4371,4372,4375],{},[34,4373,4374],{},"Add one detail only you could know."," A real specific from your week is the single fastest way to signal a person is here. It's also the one thing that can't be generated.",[30,4377,4378,4381],{},[34,4379,4380],{},"End on a statement."," Land the point. If a question earns its place, make it a real one, not a prompt.",[19,4383,4385],{"id":4384},"a-before-and-after","A before-and-after",[27,4387,4388,4391],{},[30,4389,4390],{},"Before: \"Ever struggled to stay consistent with content? 🤔 Here are 3 game-changing tips: 1) Plan ahead 2) Batch your work 3) Stay authentic. Consistency isn't about motivation, it's about systems! What's your biggest struggle? 👇\"",[30,4392,4393],{},"After: \"I missed posting for nine days straight last month, then figured out why. I was trying to decide what to post every morning instead of the night I had the idea. Now I write the idea down the moment it lands, and 'what do I post today' stopped being a question.\"",[12,4395,4396],{},"The rewrite keeps the same advice. It just replaces the rhetorical hook with a real moment, drops the puffery and the contrast crutch, and ends on a point instead of a plea.",[19,4398,4400],{"id":4399},"a-quick-checklist","A quick checklist",[12,4402,4403],{},"Before you post, read your first two lines and the last one, and check:",[439,4405,4406,4409,4412,4415,4418],{},[30,4407,4408],{},"Does it open with a specific, not a question or a setup?",[30,4410,4411],{},"Did you cut every adjective doing a fact's job?",[30,4413,4414],{},"Is there anything in it only you could have written?",[30,4416,4417],{},"Did you delete the \"not X, it's Y\" line?",[30,4419,4420],{},"Does it end on a statement instead of begging for a comment?",[19,4422,4424],{"id":4423},"frequently-asked-questions","Frequently asked questions",[12,4426,4427,4430],{},[34,4428,4429],{},"Does the algorithm penalize AI content?","\nThe bigger effect isn't a penalty from the platform, it's people skipping it. A post that reads as generated gets fewer of the early interactions that reach drives, so it quietly underperforms regardless of any rule.",[12,4432,4433,4436],{},[34,4434,4435],{},"Why do my AI-drafted posts get impressions but no engagement?","\nImpressions mean the feed showed your post and people scrolled past. That's a recognizability problem in the opening, not a reach problem. Rewrite the first line to a real specific before changing anything else.",[12,4438,4439,4442,4443,239],{},[34,4440,4441],{},"Can AI-written content sound human?","\nYes, once the tells are edited out and a real specific is added. The draft is a starting point; the human pass is what removes the pattern and puts a voice back in. See ",[71,4444,4445],{"href":1918},"how to edit a ChatGPT draft so it sounds like you",[12,4447,4448,4451],{},[34,4449,4450],{},"Do people really notice?","\nNot consciously, and that's the point. They don't think \"this is AI.\" They just feel a small drop in interest and keep scrolling. The reaction is faster than the judgment.",[12,4453,4454,4455,4459,4460,239],{},"None of this requires more time. It requires resisting the shape the tools push you toward. SelfSM does that resistance for you: it strips these patterns from every draft and writes in each brand's voice, so \"sounds generated\" stops being the reason your posts get skipped, and Chloe keeps each one on-brand from idea to published. ",[71,4456,4458],{"href":4457},"\u002F","See how it works",". For the full catalogue of what to strip yourself, see ",[71,4461,2635],{"href":125},{"title":247,"searchDepth":248,"depth":248,"links":4463},[4464,4465,4466,4467,4468,4469,4470],{"id":4289,"depth":248,"text":4290},{"id":4299,"depth":248,"text":4300},{"id":4338,"depth":248,"text":4339},{"id":4348,"depth":248,"text":4349},{"id":4384,"depth":248,"text":4385},{"id":4399,"depth":248,"text":4400},{"id":4423,"depth":248,"text":4424},"Why generated posts get scrolled past, the exact patterns readers recognize, and how to fix each one, with before-and-after examples and a checklist.",{},{"title":4281,"description":4471},"blog\u002Fwhy-ai-social-posts-get-ignored",[270,4476],"social media","kSpy0Pqw0CMb7-ktPvFwCStx7z37mA-TuKGUOqk7E_A",{"id":4479,"title":4480,"author":7,"body":4481,"date":4606,"description":4607,"draft":259,"extension":260,"image":261,"meta":4608,"navigation":263,"path":125,"seo":4609,"stem":4610,"tags":4611,"updated":261,"__hash__":4612},"blog\u002Fblog\u002Fai-writing-tells.md","The tells that make writing sound like AI (and how to fix each one)",{"type":9,"value":4482,"toc":4596},[4483,4486,4490,4493,4499,4503,4506,4511,4515,4518,4523,4527,4530,4535,4539,4542,4547,4551,4554,4559,4563,4566,4571,4575,4582],[12,4484,4485],{},"People can't always explain why a post feels off, but they sense it almost immediately and keep scrolling. What they're reacting to is a set of patterns: habits the models fall into that, once you've seen them a few times, are impossible to unsee. Here's the field guide, with the fix for each.",[19,4487,4489],{"id":4488},"the-em-dash-everywhere","The em-dash everywhere",[12,4491,4492],{},"Generated posts often overuse the em-dash. One per paragraph, sometimes two, usually in that \"clause — extra clause\" rhythm. Human writers tend to use it less predictably, and with more intention.",[12,4494,4495,4498],{},[34,4496,4497],{},"Fix:"," Read the post and keep at most one em-dash. Turn the rest into full stops. Short sentences carry more weight anyway.",[19,4500,4502],{"id":4501},"puffery-instead-of-specifics","Puffery instead of specifics",[12,4504,4505],{},"\"Powerful.\" \"Seamless.\" \"Game-changing.\" \"Robust.\" Adjectives that sound like value but carry none. They're what a sentence reaches for when it doesn't have a fact.",[12,4507,4508,4510],{},[34,4509,4497],{}," Delete the adjective and put a specific in its place. Not \"a powerful workflow\" but \"cut our drafting time in half.\" If you can't name the specific, the claim probably isn't true yet.",[19,4512,4514],{"id":4513},"the-its-not-x-its-y-reflex","The \"it's not X, it's Y\" reflex",[12,4516,4517],{},"\"It's not about working harder, it's about working smarter.\" Once you notice this construction, you'll see it in a lot of generated posts. It feels insightful and says almost nothing.",[12,4519,4520,4522],{},[34,4521,4497],{}," Cut it. State the point directly. If Y is the real idea, just say Y.",[19,4524,4526],{"id":4525},"suspicious-symmetry","Suspicious symmetry",[12,4528,4529],{},"Three bullet points, each the same length, each starting with the same kind of phrase. Real thinking is lumpy: one idea is big, another is a footnote. Perfectly balanced structure reads like a template.",[12,4531,4532,4534],{},[34,4533,4497],{}," Make the list uneven on purpose. One long point, one short. Cut the third if it's only there to make three.",[19,4536,4538],{"id":4537},"the-rhetorical-question-opener","The rhetorical-question opener",[12,4540,4541],{},"\"Ever wondered why some posts go viral?\" It's a stall dressed as a hook. The reader hasn't been given anything yet, just asked to do the work.",[12,4543,4544,4546],{},[34,4545,4497],{}," Open with a specific instead. A number, a moment, a claim. \"Last week a post with a typo outperformed everything I'd polished for a month.\"",[19,4548,4550],{"id":4549},"the-closing-that-begs","The closing that begs",[12,4552,4553],{},"\"What do you think?\" \"Agree?\" \"Drop your thoughts below.\" A confident post lands its point and stops. The plea for engagement reads as exactly that.",[12,4555,4556,4558],{},[34,4557,4497],{}," End on a statement. If a question genuinely belongs, make it a real one, specific enough that answering it takes thought.",[19,4560,4562],{"id":4561},"the-stacked-hedge","The stacked hedge",[12,4564,4565],{},"\"It's often the case that many teams sometimes struggle to…\" One qualifier can be honest. Three in a row drain the life out of a sentence and hide whether you're actually saying anything.",[12,4567,4568,4570],{},[34,4569,4497],{}," Keep the qualification that's true; cut the ones added only to sound safe. \"Teams often struggle with this\" says as much as the stacked version, and you can read it without wincing.",[19,4572,4574],{"id":4573},"how-to-actually-apply-this","How to actually apply this",[12,4576,4577,4578,4581],{},"You don't need to memorize the list. Do one pass with a single question in mind: ",[519,4579,4580],{},"would I say this sentence out loud to someone I respect?"," Most tells don't survive it. The puffery sounds fake, the symmetry sounds rehearsed, the begging sounds needy.",[12,4583,4584,4585,4587,4588,4591,4592,4595],{},"This is the same standard ",[71,4586,7],{"href":4457}," is built around, so the draft comes back with these patterns already stripped and the editing pass is short instead of a rescue mission. For the fuller argument on why this matters, start with ",[71,4589,4590],{"href":73},"why AI social posts get ignored",". When you're editing your own chatbot output, ",[71,4593,4594],{"href":1918},"this walkthrough"," turns the list above into a repeatable pass.",{"title":247,"searchDepth":248,"depth":248,"links":4597},[4598,4599,4600,4601,4602,4603,4604,4605],{"id":4488,"depth":248,"text":4489},{"id":4501,"depth":248,"text":4502},{"id":4513,"depth":248,"text":4514},{"id":4525,"depth":248,"text":4526},{"id":4537,"depth":248,"text":4538},{"id":4549,"depth":248,"text":4550},{"id":4561,"depth":248,"text":4562},{"id":4573,"depth":248,"text":4574},"2026-08-02","A field guide to the patterns readers now recognize as machine-written, with a concrete fix for every one, so your posts read like a person again.",{},{"title":4480,"description":4607},"blog\u002Fai-writing-tells",[270,269],"c_mykMSYGzeBanFnRrY_2aITCR3Za4SSqygS3tCg8kw",{"id":4614,"title":4615,"author":7,"body":4616,"date":4712,"description":4713,"draft":259,"extension":260,"image":261,"meta":4714,"navigation":263,"path":1918,"seo":4715,"stem":4716,"tags":4717,"updated":261,"__hash__":4718},"blog\u002Fblog\u002Fhumanize-chatgpt-draft.md","How to edit a ChatGPT draft so it sounds like you",{"type":9,"value":4617,"toc":4703},[4618,4621,4625,4628,4632,4635,4639,4642,4646,4649,4653,4656,4660,4663,4667,4686],[12,4619,4620],{},"The draft is done in ten seconds. Making it sound like a person takes longer, and it's the part most people skip, which is why so many feeds read like the same voice wearing different logos. You don't have to rewrite from scratch. You have to run a pass. Here's the one I use.",[19,4622,4624],{"id":4623},"_1-cut-the-first-paragraph","1. Cut the first paragraph",[12,4626,4627],{},"Chatbots warm up. The real opening is usually the second or third sentence, once the model has finished clearing its throat. Read the draft and find the first line that actually says something. Start there. Delete everything above it.",[19,4629,4631],{"id":4630},"_2-replace-the-biggest-adjective-with-a-fact","2. Replace the biggest adjective with a fact",[12,4633,4634],{},"Scan for the puffed-up words: \"powerful,\" \"seamless,\" \"essential,\" \"game-changing.\" Pick the one doing the most fake work and swap it for something concrete: a number, a name, a moment. The rest you can usually just delete. Nouns and verbs carry a sentence; adjectives are what you reach for when you don't have one.",[19,4636,4638],{"id":4637},"_3-break-the-symmetry","3. Break the symmetry",[12,4640,4641],{},"If there are three bullet points of equal length, the draft is wearing a uniform. Make it uneven. Merge two, cut one, let one point run long because it deserves to. Human emphasis is uneven; templates are tidy.",[19,4643,4645],{"id":4644},"_4-read-it-out-loud","4. Read it out loud",[12,4647,4648],{},"This is the step that does the most work and gets skipped the most. Read the whole thing aloud. Every sentence you stumble on, or wouldn't actually say to someone, gets rewritten or cut. \"It's not about X, it's about Y\" dies here. So does the hedging. So does the closing plea.",[19,4650,4652],{"id":4651},"_5-add-one-thing-only-you-know","5. Add one thing only you know",[12,4654,4655],{},"The fastest way to make writing human is a specific that couldn't have been generated: a detail from your week, a client's exact words, a number from your own work, an opinion you'd defend. One real specific does more than a hundred edits. It's also what search engines and readers both reward, because it's proof a person was actually here.",[19,4657,4659],{"id":4658},"_6-end-on-a-period-not-a-question","6. End on a period, not a question",[12,4661,4662],{},"Delete \"What do you think?\" Land the point and stop. If a question genuinely belongs, make it specific enough to be worth answering.",[19,4664,4666],{"id":4665},"the-whole-pass-in-order","The whole pass, in order",[439,4668,4669,4672,4675,4678,4681,4684],{},[30,4670,4671],{},"Cut the warm-up opening.",[30,4673,4674],{},"Swap the biggest adjective for a fact.",[30,4676,4677],{},"Make the structure uneven.",[30,4679,4680],{},"Read aloud; fix anything you wouldn't say.",[30,4682,4683],{},"Add one thing only you know.",[30,4685,4380],{},[12,4687,4688,4689,4692,4693,4696,4697,4699,4700,239],{},"It takes about five minutes once it's a habit. If you want to know exactly what you're editing ",[519,4690,4691],{},"out",", the ",[71,4694,4695],{"href":125},"full list of AI tells"," is the companion to this pass. And if doing this by hand for every post across several brands is the actual bottleneck, that's the part ",[71,4698,7],{"href":4457}," automates, so the draft arrives already past step four. The step before this one — getting the model to write in your voice at all, rather than editing it in afterwards — is ",[71,4701,4702],{"href":264},"here",{"title":247,"searchDepth":248,"depth":248,"links":4704},[4705,4706,4707,4708,4709,4710,4711],{"id":4623,"depth":248,"text":4624},{"id":4630,"depth":248,"text":4631},{"id":4637,"depth":248,"text":4638},{"id":4644,"depth":248,"text":4645},{"id":4651,"depth":248,"text":4652},{"id":4658,"depth":248,"text":4659},{"id":4665,"depth":248,"text":4666},"2026-07-31","A five-minute editing pass that turns generic chatbot output into something in your voice, no rewrite from scratch.",{},{"title":4615,"description":4713},"blog\u002Fhumanize-chatgpt-draft",[270,269],"SdXUzZvHI8yik8UOH3bCwmfsw55i-m_8TU8cDEYFrnw",{"id":4720,"title":4721,"author":7,"body":4722,"date":5114,"description":5115,"draft":259,"extension":260,"image":261,"meta":5116,"navigation":263,"path":5117,"seo":5118,"stem":5119,"tags":5120,"updated":261,"__hash__":5122},"blog\u002Fblog\u002Flinkedin-hooks.md","How to write LinkedIn hooks that earn the \"see more\" click",{"type":9,"value":4723,"toc":5091},[4724,4727,4731,4734,4737,4741,4744,4748,4753,4756,4764,4767,4771,4774,4782,4785,4789,4792,4800,4803,4807,4810,4818,4821,4825,4828,4836,4839,4843,4846,4850,4867,4871,4888,4892,4909,4913,4930,4934,4951,4955,5002,5006,5017,5020,5024,5027,5044,5047,5049,5055,5061,5067,5073,5079],[12,4725,4726],{},"LinkedIn hides most of a longer post behind the \"…more\" cutoff. Your opening lines decide whether anyone taps to read the rest, so the hook does more work than the other 200 words combined. This is a guide to writing that opening: what a hook is, the types that work, a stack of examples by role, the mistakes that kill a post, and a checklist you can run before you publish.",[19,4728,4730],{"id":4729},"what-a-linkedin-hook-actually-is","What a LinkedIn hook actually is",[12,4732,4733],{},"A hook is the first line or two of a post: the part visible in the feed before the reader taps \"…more.\" Its only job is to make stopping the scroll feel involuntary and make the tap feel worth it. It isn't a summary of the post and it isn't a headline. It's the smallest promise you can make that a reader can't comfortably ignore.",[12,4735,4736],{},"Good hooks share one property: they create a small gap the reader wants closed. A specific number they want explained. A story they need to see finish. A claim they want to argue with. Everything below is a different way to open that gap.",[19,4738,4740],{"id":4739},"how-the-more-cutoff-works","How the \"…more\" cutoff works",[12,4742,4743],{},"LinkedIn shows a preview of a post and truncates the rest behind \"…more.\" Exactly how much shows varies by device, screen size, and formatting, so don't optimize for a specific line count. Optimize for the truth underneath it: the first sentence, sometimes two, is all most people ever see. Front-load the interesting part and put nothing before it. A line spent introducing the post (\"I've been thinking about hiring lately…\") is a line the reader never gets past.",[19,4745,4747],{"id":4746},"five-hook-types-that-work","Five hook types that work",[4749,4750,4752],"h3",{"id":4751},"_1-start-mid-story","1. Start mid-story",[12,4754,4755],{},"Drop the reader into a moment instead of setting one up.",[27,4757,4758,4761],{},[30,4759,4760],{},"Weak: \"I want to share some thoughts on hiring.\"",[30,4762,4763],{},"Strong: \"I rejected the most qualified candidate I've ever interviewed. Here's why.\"",[12,4765,4766],{},"The second one puts a question in the reader's head before they've decided to care. You don't announce that a story is coming; you start it.",[4749,4768,4770],{"id":4769},"_2-use-a-specific-result","2. Use a specific result",[12,4772,4773],{},"A concrete number reads as proof that something real is behind the post.",[27,4775,4776,4779],{},[30,4777,4778],{},"\"We cut churn from 6% to 2% in one quarter.\"",[30,4780,4781],{},"\"I reviewed 200 cold emails last week. Four got a reply.\"",[12,4783,4784],{},"Vague beats nobody. \"We improved retention\" is a shrug; \"6% to 2%\" is a promise.",[4749,4786,4788],{"id":4787},"_3-take-a-defensible-position","3. Take a defensible position",[12,4790,4791],{},"If your first line could be posted by any account in your industry, it won't stop anyone. Say something a reasonable person could disagree with.",[27,4793,4794,4797],{},[30,4795,4796],{},"\"Most 'thought leadership' is fear of saying anything disagreeable.\"",[30,4798,4799],{},"\"Your best hire will probably fail your interview.\"",[12,4801,4802],{},"You don't have to be inflammatory. You have to be specific enough that someone could push back.",[4749,4804,4806],{"id":4805},"_4-reveal-a-surprising-contrast","4. Reveal a surprising contrast",[12,4808,4809],{},"Set an expectation in the first half of the sentence and break it in the second.",[27,4811,4812,4815],{},[30,4813,4814],{},"\"The best marketing hire we made had never worked in marketing.\"",[30,4816,4817],{},"\"Our worst-performing post last month took me four minutes to write.\"",[12,4819,4820],{},"The gap between the setup and the twist is the hook. The reader keeps going to resolve it.",[4749,4822,4824],{"id":4823},"_5-open-a-curiosity-loop","5. Open a curiosity loop",[12,4826,4827],{},"State that something happened, or that you learned something, without saying what.",[27,4829,4830,4833],{},[30,4831,4832],{},"\"One sentence in our onboarding email doubled activation. It's not the one you'd guess.\"",[30,4834,4835],{},"\"I fired our biggest client on purpose. Best decision we made all year.\"",[12,4837,4838],{},"A half-told thing is a loop the brain wants closed. Just make sure you actually close it in the post; an open loop with no payoff burns the next post's trust.",[19,4840,4842],{"id":4841},"_25-linkedin-hook-examples-by-role","25 LinkedIn hook examples by role",[12,4844,4845],{},"Use these as patterns, not scripts. Swap in your own specifics; a hook is only as strong as the real detail behind it.",[4749,4847,4849],{"id":4848},"for-founders","For founders",[27,4851,4852,4855,4858,4861,4864],{},[30,4853,4854],{},"\"We hit $1M ARR with no marketing team. Here's exactly how the pipeline actually worked.\"",[30,4856,4857],{},"\"I almost shut the company down in March. What changed in six weeks.\"",[30,4859,4860],{},"\"The feature everyone asked for is the one that nearly killed us.\"",[30,4862,4863],{},"\"We raised a round we didn't need. I regret it.\"",[30,4865,4866],{},"\"Three years in, here's the advice I'd give myself on day one.\"",[4749,4868,4870],{"id":4869},"for-recruiters-and-hiring","For recruiters and hiring",[27,4872,4873,4876,4879,4882,4885],{},[30,4874,4875],{},"\"I've read 10,000 resumes. The ones that get interviews do one thing.\"",[30,4877,4878],{},"\"The best candidate I ever hired had a two-line cover letter.\"",[30,4880,4881],{},"\"Stop asking 'what's your biggest weakness.' Here's what to ask instead.\"",[30,4883,4884],{},"\"We removed the take-home test. Our hiring got better.\"",[30,4886,4887],{},"\"A candidate ghosted us after the offer. Their reason was fair.\"",[4749,4889,4891],{"id":4890},"for-sales-and-b2b","For sales and B2B",[27,4893,4894,4897,4900,4903,4906],{},[30,4895,4896],{},"\"I lost a $40k deal on a single sentence. Here it is.\"",[30,4898,4899],{},"\"The best sales email I ever sent was four lines long.\"",[30,4901,4902],{},"\"Discovery calls don't fail at the demo. They fail in the first two minutes.\"",[30,4904,4905],{},"\"We stopped chasing 'decision makers.' Close rate went up.\"",[30,4907,4908],{},"\"A prospect told me exactly why we lost. I wrote it all down.\"",[4749,4910,4912],{"id":4911},"for-product-updates","For product updates",[27,4914,4915,4918,4921,4924,4927],{},[30,4916,4917],{},"\"We shipped the smallest feature of the quarter. It's the most used one.\"",[30,4919,4920],{},"\"You asked for dark mode for two years. Today it's live, and here's why it took so long.\"",[30,4922,4923],{},"\"We deleted half the settings page. Support tickets dropped.\"",[30,4925,4926],{},"\"The update nobody noticed made everything 40% faster.\"",[30,4928,4929],{},"\"We rebuilt the thing you complained about most. Try it and tell me if we got it right.\"",[4749,4931,4933],{"id":4932},"for-personal-stories","For personal stories",[27,4935,4936,4939,4942,4945,4948],{},[30,4937,4938],{},"\"I got laid off on a Friday. By Monday I'd learned something I wish I'd known years earlier.\"",[30,4940,4941],{},"\"The advice that changed my career came from someone I almost didn't call back.\"",[30,4943,4944],{},"\"I said no to a promotion. Here's what happened next.\"",[30,4946,4947],{},"\"For ten years I optimized the wrong number.\"",[30,4949,4950],{},"\"My first manager did one thing I still copy today.\"",[19,4952,4954],{"id":4953},"hooks-that-usually-fail-and-the-fix","Hooks that usually fail (and the fix)",[27,4956,4957,4966,4975,4984,4993],{},[30,4958,4959,4962,4963,4965],{},[34,4960,4961],{},"The rhetorical question."," \"Ever struggled with X?\" reads as a stall. ",[34,4964,4497],{}," make a statement the reader can't answer without reading on.",[30,4967,4968,4971,4972,4974],{},[34,4969,4970],{},"Announcing the format."," \"Thread 🧵\" or \"A quick post about…\" spends the hook on logistics. ",[34,4973,4497],{}," open with the content, not the container.",[30,4976,4977,4980,4981,4983],{},[34,4978,4979],{},"Warming up."," The good line is often your second or third sentence. ",[34,4982,4497],{}," cut everything above it.",[30,4985,4986,4989,4990,4992],{},[34,4987,4988],{},"The vague promise."," \"Here are some tips to improve your marketing.\" ",[34,4991,4497],{}," name the specific and the stakes: \"The one email change that doubled our reply rate.\"",[30,4994,4995,4998,4999,5001],{},[34,4996,4997],{},"Too clever to parse."," A pun or riddle the reader has to decode. ",[34,5000,4497],{}," clarity beats cleverness in the feed. Say the interesting thing plainly.",[19,5003,5005],{"id":5004},"before-and-after-rewrites","Before-and-after rewrites",[27,5007,5008,5011,5014],{},[30,5009,5010],{},"Before: \"Excited to share that we've launched a new feature!\" → After: \"We shipped the feature you've asked for since 2024. It took longer than it should have, and here's why.\"",[30,5012,5013],{},"Before: \"Some thoughts on remote work culture.\" → After: \"We went fully remote and our best people got worse. Here's what we changed.\"",[30,5015,5016],{},"Before: \"Here's how to write better cold emails.\" → After: \"I sent 200 cold emails last week. The four that got replies had one line in common.\"",[12,5018,5019],{},"Each rewrite does the same thing: replaces the setup with a specific, and replaces the announcement with a promise.",[19,5021,5023],{"id":5022},"a-5-point-hook-checklist","A 5-point hook checklist",[12,5025,5026],{},"Before you publish, read only your first two lines and check:",[439,5028,5029,5032,5035,5038,5041],{},[30,5030,5031],{},"Does it open with a specific (a number, a name, a moment), not a setup?",[30,5033,5034],{},"Is there a gap the reader wants closed?",[30,5036,5037],{},"Could someone disagree with it, or is it safe enough to ignore?",[30,5039,5040],{},"Did you cut the warm-up sentence above the good line?",[30,5042,5043],{},"If you saw only these two lines in the feed, would you tap \"…more\"?",[12,5045,5046],{},"If any answer is no, fix it before you touch the rest of the post. The hook is the only part most of your audience will read, so it's the part worth the most editing.",[19,5048,4424],{"id":4423},[12,5050,5051,5054],{},[34,5052,5053],{},"How long should a LinkedIn hook be?","\nShort enough to land before the cutoff. One strong sentence is usually plenty; two if the second one raises the stakes. If your hook needs three lines to work, the good line is probably buried in the middle.",[12,5056,5057,5060],{},[34,5058,5059],{},"Should I use emojis in the hook?","\nAn emoji won't save a weak line, and it can read as decoration. Lead with the words. If an emoji adds meaning, fine; if it's there to look lively, cut it.",[12,5062,5063,5066],{},[34,5064,5065],{},"What's the single best first line for a LinkedIn post?","\nThere isn't one. The best first line is the most specific true thing you have: the number, the moment, the opinion you'd defend. Specificity does the work, not a formula.",[12,5068,5069,5072],{},[34,5070,5071],{},"How many hooks should I write per post?","\nWrite three or four openers for the same post and pick the one you'd stop scrolling for. The first hook you think of is rarely the sharpest.",[12,5074,5075,5078],{},[34,5076,5077],{},"Why do my posts get impressions but no clicks to \"…more\"?","\nImpressions mean the feed showed your hook and people kept scrolling. That's a hook problem, not a reach problem. Rewrite the first two lines before you change anything else.",[12,5080,5081,5082,5084,5085,5088,5089,239],{},"If your openers keep coming out generic, the issue is usually voice, not format. SelfSM drafts in each brand's voice with the AI tells stripped, so the hook lands as a person's and not a prompt's, and Chloe helps sharpen the opener before it goes out. ",[71,5083,4458],{"href":4457},". Threads on X follow the same logic; if that's your channel, see ",[71,5086,5087],{"href":1977},"four X thread structures that get read","; for the craft underneath, start with ",[71,5090,2635],{"href":125},{"title":247,"searchDepth":248,"depth":248,"links":5092},[5093,5094,5095,5103,5110,5111,5112,5113],{"id":4729,"depth":248,"text":4730},{"id":4739,"depth":248,"text":4740},{"id":4746,"depth":248,"text":4747,"children":5096},[5097,5099,5100,5101,5102],{"id":4751,"depth":5098,"text":4752},3,{"id":4769,"depth":5098,"text":4770},{"id":4787,"depth":5098,"text":4788},{"id":4805,"depth":5098,"text":4806},{"id":4823,"depth":5098,"text":4824},{"id":4841,"depth":248,"text":4842,"children":5104},[5105,5106,5107,5108,5109],{"id":4848,"depth":5098,"text":4849},{"id":4869,"depth":5098,"text":4870},{"id":4890,"depth":5098,"text":4891},{"id":4911,"depth":5098,"text":4912},{"id":4932,"depth":5098,"text":4933},{"id":4953,"depth":248,"text":4954},{"id":5004,"depth":248,"text":5005},{"id":5022,"depth":248,"text":5023},{"id":4423,"depth":248,"text":4424},"2026-07-28","A practical guide to LinkedIn hooks: five hook types, 25+ examples by role, before-and-after rewrites, the mistakes that kill a post, and a short checklist.",{},"\u002Fblog\u002Flinkedin-hooks",{"title":4721,"description":5115},"blog\u002Flinkedin-hooks",[2656,5121],"hooks","-ZWLORCx8cXFCCLKAEq6jqLLT3Rv9ZrDxrmqLP3DteI",{"id":5124,"title":5125,"author":7,"body":5126,"date":5232,"description":5233,"draft":259,"extension":260,"image":261,"meta":5234,"navigation":263,"path":237,"seo":5235,"stem":5236,"tags":5237,"updated":261,"__hash__":5238},"blog\u002Fblog\u002Fwhat-ai-detectors-catch.md","What AI content detectors actually catch (and what they miss)",{"type":9,"value":5127,"toc":5226},[5128,5131,5135,5138,5152,5159,5163,5166,5180,5183,5187,5190,5193,5195,5198,5212],[12,5129,5130],{},"People keep asking: will this pass an AI detector? It's the wrong question, but it's worth understanding why, because the answer changes how you edit.",[19,5132,5134],{"id":5133},"how-detectors-actually-work","How detectors actually work",[12,5136,5137],{},"AI-text detectors don't \"know\" a machine wrote something. They measure statistical properties of the text, mainly two:",[27,5139,5140,5146],{},[30,5141,5142,5145],{},[34,5143,5144],{},"Perplexity:"," how predictable each next word is. Model output runs smooth and predictable; human writing is bumpier and more surprising.",[30,5147,5148,5151],{},[34,5149,5150],{},"Burstiness:"," how much sentence length and rhythm vary. People write a long sentence, then a short one. Models drift toward an even, uniform cadence.",[12,5153,5154,5155,5158],{},"A detector sees low perplexity and low burstiness and raises its hand. That's the whole trick. It's a probability guess about ",[519,5156,5157],{},"how the text is shaped",", not a fact about who made it.",[19,5160,5162],{"id":5161},"why-they-get-it-wrong-in-both-directions","Why they get it wrong in both directions",[12,5164,5165],{},"Because it's a shape-based guess, it fails predictably.",[27,5167,5168,5174],{},[30,5169,5170,5173],{},[34,5171,5172],{},"It flags real human writing."," Clear, plain, well-structured prose, the kind good writers aim for, can look \"too smooth\" and get marked as AI. Non-native English writers get flagged more often, unfairly.",[30,5175,5176,5179],{},[34,5177,5178],{},"It clears real AI."," Add a few odd word choices, vary the sentence lengths, and generated text sails through. The patterns detectors measure are the ones easiest to perturb.",[12,5181,5182],{},"So a passing score proves very little, and a failing score might just mean you wrote cleanly.",[19,5184,5186],{"id":5185},"the-right-goal-isnt-a-score-its-a-reader","The right goal isn't a score, it's a reader",[12,5188,5189],{},"Here's what the detector debate misses: your audience is the detector that matters. A person doesn't run a perplexity check. They feel the tells, the puffery, the tidy symmetry, the \"not X but Y\" reflex, and they lose interest. You can pass every automated detector and still get scrolled past, and you can fail one while writing something people actually read.",[12,5191,5192],{},"Optimizing for a detector score pulls you toward gaming statistics. Optimizing for a human pulls you toward specifics, real opinions, and uneven, natural rhythm. That also lowers your detector score, as a side effect rather than the goal.",[19,5194,4349],{"id":4348},[12,5196,5197],{},"Forget the score. Edit for the reader:",[27,5199,5200,5203,5206,5209],{},[30,5201,5202],{},"Lead with something specific and true.",[30,5204,5205],{},"Vary your sentence lengths on purpose.",[30,5207,5208],{},"Cut the adjectives that stand in for facts.",[30,5210,5211],{},"Add at least one detail only you could know.",[12,5213,5214,5215,5218,5219,5222,5223,5225],{},"That's the same standard behind the ",[71,5216,5217],{"href":125},"list of AI tells"," and the ",[71,5220,5221],{"href":1918},"editing pass",", and it's what ",[71,5224,7],{"href":4457}," optimizes for by default: sound like a person, because that's who's actually reading.",{"title":247,"searchDepth":248,"depth":248,"links":5227},[5228,5229,5230,5231],{"id":5133,"depth":248,"text":5134},{"id":5161,"depth":248,"text":5162},{"id":5185,"depth":248,"text":5186},{"id":4348,"depth":248,"text":4349},"2026-07-26","How AI-text detectors really work, why they flag human writing and clear real AI, and why chasing a detector score is the wrong goal.",{},{"title":5125,"description":5233},"blog\u002Fwhat-ai-detectors-catch",[269,270],"ivIVWYDJGhPwmmo9TB8N8WDqcYvZ4RYGdzQpOuu3fJE",{"id":5240,"title":5241,"author":7,"body":5242,"date":5363,"description":5364,"draft":259,"extension":260,"image":261,"meta":5365,"navigation":263,"path":865,"seo":5366,"stem":5367,"tags":5368,"updated":261,"__hash__":5369},"blog\u002Fblog\u002Fcontent-approval-workflow.md","The content approval workflow that doesn't lose the client",{"type":9,"value":5243,"toc":5355},[5244,5247,5251,5254,5274,5278,5289,5292,5312,5316,5323,5327,5330,5334,5337,5341],[12,5245,5246],{},"Most of the friction in running social for clients lives in the approval, not the writing: the back-and-forth in email threads, the \"can we hop on a call,\" the post that sat for a week because nobody clicked yes. Get the approval flow right and the whole operation speeds up. Get it wrong and every post becomes a small negotiation.",[19,5248,5250],{"id":5249},"why-approvals-go-wrong","Why approvals go wrong",[12,5252,5253],{},"Three things quietly break most approval processes:",[439,5255,5256,5262,5268],{},[30,5257,5258,5261],{},[34,5259,5260],{},"The client sees the work too late."," They review a finished, scheduled post. Their feedback then means redoing it, which feels expensive, so they hesitate.",[30,5263,5264,5267],{},[34,5265,5266],{},"Feedback lives in five places."," Some in email, some in a DM, some spoken on a call and never written down. Nobody's sure what was actually agreed.",[30,5269,5270,5273],{},[34,5271,5272],{},"Nothing has a deadline."," \"Let me know when you get a chance\" is where posts go to die.",[19,5275,5277],{"id":5276},"fix-the-sequence-not-the-effort","Fix the sequence, not the effort",[12,5279,5280,5281,5284,5285,5288],{},"The single biggest improvement is ",[519,5282,5283],{},"moving approval earlier",". Get a yes on the ",[34,5286,5287],{},"idea and the angle"," before the post is fully written. A client can react to \"here's the angle for Thursday: a short take on the pricing change\" in ten seconds. Reacting to a polished, scheduled post feels like being handed a bill.",[12,5290,5291],{},"So the sequence becomes:",[439,5293,5294,5300,5306],{},[30,5295,5296,5299],{},[34,5297,5298],{},"Approve the plan:"," the week's angles, in one glance. Cheap to change.",[30,5301,5302,5305],{},[34,5303,5304],{},"Draft against approved angles:"," now you're writing things already blessed.",[30,5307,5308,5311],{},[34,5309,5310],{},"Approve the draft:"," a formality most of the time, because the hard call happened at step 1.",[19,5313,5315],{"id":5314},"make-feedback-have-one-home","Make feedback have one home",[12,5317,5318,5319,5322],{},"Every comment about a post should live ",[519,5320,5321],{},"on that post",", not scattered across channels. When feedback and content sit in the same place, \"what did we agree?\" stops being a question. This is dull, and it matters more than any clever tactic.",[19,5324,5326],{"id":5325},"give-every-approval-a-default","Give every approval a default",[12,5328,5329],{},"\"No response by end of day tomorrow\" needs to mean something: either it auto-proceeds (for low-risk posts) or it clearly waits. Ambiguity is what turns a one-day approval into a one-week one. Agree the default once, up front, and the client stops being a bottleneck they didn't ask to be.",[19,5331,5333],{"id":5332},"keep-a-record","Keep a record",[12,5335,5336],{},"When a client says \"we'd never say that,\" you want to point to the approved version, not relitigate memory. A trail of what was approved, and when, is what keeps a disagreement from becoming a trust problem.",[19,5338,5340],{"id":5339},"where-this-lives-in-practice","Where this lives in practice",[12,5342,5343,5344,5346,5347,5351,5352,239],{},"You can run this in a spreadsheet and a shared doc, and plenty of agencies do. It works right up until you're managing several clients, at which point keeping every thread, deadline, and approved version straight becomes its own job. That's the part ",[71,5345,7],{"href":4457}," folds into the flow: approvals sit on the content itself, with a record of what was agreed, so the relationship stays intact while posts keep moving. If you're mapping the whole path from idea to published, see ",[71,5348,5350],{"href":5349},"\u002Fblog\u002Fidea-to-published-pipeline","the repeatable pipeline",". For the longer version of this page — the four ways approval breaks and how to design around each — see ",[71,5353,5354],{"href":647},"a social media approval process clients actually follow",{"title":247,"searchDepth":248,"depth":248,"links":5356},[5357,5358,5359,5360,5361,5362],{"id":5249,"depth":248,"text":5250},{"id":5276,"depth":248,"text":5277},{"id":5314,"depth":248,"text":5315},{"id":5325,"depth":248,"text":5326},{"id":5332,"depth":248,"text":5333},{"id":5339,"depth":248,"text":5340},"2026-07-24","Client approvals are where SMM work slows down and relationships fray. A simple, low-friction approval flow that keeps trust and keeps posts moving.",{},{"title":5241,"description":5364},"blog\u002Fcontent-approval-workflow",[805,505],"cPWzQY9eeCjZHoho8wSju9yAcERvToqpoxnwWcD-uX4",{"id":5371,"title":5372,"author":7,"body":5373,"date":5481,"description":5482,"draft":259,"extension":260,"image":261,"meta":5483,"navigation":263,"path":658,"seo":5484,"stem":5485,"tags":5486,"updated":261,"__hash__":5487},"blog\u002Fblog\u002Fmanaging-multiple-brand-voices.md","How to keep several brand voices straight (without them blurring)",{"type":9,"value":5374,"toc":5473},[5375,5378,5381,5385,5388,5414,5417,5421,5424,5428,5431,5435,5438,5442,5453,5457],[12,5376,5377],{},"Run one brand and its voice lives in your head. Run five and something creeps in: they all start sounding a little the same, a little like you, or a little like the same tool you draft in. Clients notice before you do. The founder who's blunt and funny starts sounding as polished as the enterprise account, and the thing they were paying for, sounding like themselves, quietly leaves.",[12,5379,5380],{},"Keeping voices distinct is a discipline, not a talent. Here's how to hold the line.",[19,5382,5384],{"id":5383},"write-the-voice-down-concretely","Write the voice down, concretely",[12,5386,5387],{},"\"Professional but friendly\" is not a voice. It's a shrug. A usable voice definition is specific enough that two different writers would produce similar posts from it:",[27,5389,5390,5396,5402,5408],{},[30,5391,5392,5395],{},[34,5393,5394],{},"Three real sample posts"," the client has actually published and liked. Examples beat adjectives.",[30,5397,5398,5401],{},[34,5399,5400],{},"First person or brand?"," \"I\" or \"we,\" and don't switch.",[30,5403,5404,5407],{},[34,5405,5406],{},"Words they use, words they'd refuse."," A short banned list does more than a long style guide.",[30,5409,5410,5413],{},[34,5411,5412],{},"One line on stance:"," are they contrarian, encouraging, deadpan, earnest?",[12,5415,5416],{},"If you can't write a post from the definition without guessing, it isn't finished.",[19,5418,5420],{"id":5419},"store-it-outside-your-head","Store it outside your head",[12,5422,5423],{},"The voice has to live somewhere durable and attached to the brand, not in a prompt you retype, not in your memory. When it's written down and stored with the client's account, a new team member can pick it up, and you can't accidentally hand Brand A's tone to Brand B.",[19,5425,5427],{"id":5426},"draft-one-brand-at-a-time","Draft one brand at a time",[12,5429,5430],{},"Voice bleed happens when you batch across clients in the same sitting, half in one head. If you're writing for three brands today, finish one brand's posts before you open the next. Context-switching mid-sentence is exactly when the founder starts sounding like the SaaS.",[19,5432,5434],{"id":5433},"do-a-whose-voice-is-this-check","Do a \"whose voice is this?\" check",[12,5436,5437],{},"Before anything ships, read it cold and ask: if I stripped the logo, would I know which client this is? If the answer is \"could be any of them,\" it's generic. Pull it back toward the specifics in the voice definition.",[19,5439,5441],{"id":5440},"protect-the-fingerprint-not-just-the-rules","Protect the fingerprint, not just the rules",[12,5443,5444,5445,5448,5449,5452],{},"Rules (\"no emojis,\" \"short sentences\") keep a voice ",[519,5446,5447],{},"safe",". What makes it ",[519,5450,5451],{},"theirs"," is the fingerprint: the signature phrases, the running references, the way they open. Rules are easy to store; fingerprint is easy to lose. Keep real sample posts on hand and re-read them; they hold what a rulebook can't.",[19,5454,5456],{"id":5455},"when-it-stops-fitting-in-your-head","When it stops fitting in your head",[12,5458,5459,5460,5462,5463,5465,5466,5468,5469,5472],{},"One or two brands, this runs on notes and attention. Past that, the number of voices, banned lists, and sample sets to keep straight outgrows memory, and that's precisely where they blur. ",[71,5461,7],{"href":4457}," keeps each brand's voice, facts, and banned phrases stored per client and drafts against them, so managing more brands doesn't mean more copies of the same voice. The neighboring problem, keeping the ",[519,5464,3568],{}," consistent across all of them, is ",[71,5467,5350],{"href":5349},". And if you're deciding how many brands is too many, ",[71,5470,5471],{"href":801},"the honest ceiling"," is lower than most job ads suggest.",{"title":247,"searchDepth":248,"depth":248,"links":5474},[5475,5476,5477,5478,5479,5480],{"id":5383,"depth":248,"text":5384},{"id":5419,"depth":248,"text":5420},{"id":5426,"depth":248,"text":5427},{"id":5433,"depth":248,"text":5434},{"id":5440,"depth":248,"text":5441},{"id":5455,"depth":248,"text":5456},"2026-07-22","Managing content for multiple clients is where voices start bleeding into each other. How to define, store, and protect each one so every brand still sounds like itself.",{},{"title":5372,"description":5482},"blog\u002Fmanaging-multiple-brand-voices",[805,505],"OQuGRU-_K7tpGYk9Jth_fcKs14h4jjp6s9hbf5YHwGU",{"id":5489,"title":5490,"author":7,"body":5491,"date":5835,"description":5836,"draft":259,"extension":260,"image":261,"meta":5837,"navigation":263,"path":1994,"seo":5838,"stem":5839,"tags":5840,"updated":261,"__hash__":5842},"blog\u002Fblog\u002Fcontent-calendar-you-can-keep.md","A 4-week content calendar you can actually keep",{"type":9,"value":5492,"toc":5824},[5493,5500,5504,5510,5513,5517,5520,5556,5559,5563,5566,5679,5682,5686,5689,5722,5725,5729,5732,5736,5739,5743,5746,5750,5753,5785,5787,5793,5803,5812,5821],[12,5494,5495,5496,5499],{},"Most content calendars collapse for one reason: \"post every day\" is a schedule, not a plan. Every morning becomes a decision, and decisions are where consistency goes to die. This is a calendar built to survive that. It decides the ",[519,5497,5498],{},"kind"," of post ahead of time so that filling any given day is a small task instead of a blank page. You'll get the five buckets it runs on, a filled four-week example you can copy, notes for adapting it per platform, and a template to reuse each month.",[19,5501,5503],{"id":5502},"what-a-content-calendar-actually-is","What a content calendar actually is",[12,5505,5506,5507,5509],{},"A content calendar is a plan that assigns each posting slot a ",[519,5508,3756],{}," before you write a word. Not the exact caption, and not a rigid script. Just the type of post that goes there: a teaching post, a proof post, a personal one. The value isn't the grid. It's that you never sit down to \"do social\" and instead sit down to \"write today's proof post,\" which is a task small enough to actually finish.",[12,5511,5512],{},"A good calendar does three things: it stops you repeating the same note two days running, it removes the daily \"what do I post\" decision, and it leaves room to react when something real happens.",[19,5514,5516],{"id":5515},"the-five-buckets","The five buckets",[12,5518,5519],{},"Rotate every post through five recurring types. Each does a different job for your audience.",[439,5521,5522,5528,5534,5540,5546],{},[30,5523,5524,5527],{},[34,5525,5526],{},"Expertise & frameworks."," Teach one thing you know well. Example: \"The three questions I ask before taking on any client.\" These build authority.",[30,5529,5530,5533],{},[34,5531,5532],{},"Industry take."," React to something happening in your field. Example: a short, opinionated post on a competitor's pricing change. Timely, and it shows you're paying attention.",[30,5535,5536,5539],{},[34,5537,5538],{},"Proof."," A result, a case, a before-and-after. Example: \"How we took a client from 200 to 2,000 followers in eight weeks.\" This is what convinces buyers.",[30,5541,5542,5545],{},[34,5543,5544],{},"Build in public."," What you're working on, what broke, what you learned this week. Example: \"We changed our onboarding and support tickets dropped by half.\" Honest and relatable.",[30,5547,5548,5551,5552,5555],{},[34,5549,5550],{},"Human."," A moment, a story, something that isn't strictly business. Example: \"What running taught me about shipping work.\" This is what makes people remember ",[519,5553,5554],{},"you",", not just your feed.",[12,5557,5558],{},"Across four weeks you touch all five repeatedly without ever running two of the same in a row.",[19,5560,5562],{"id":5561},"a-filled-example-four-weeks","A filled example: four weeks",[12,5564,5565],{},"Here's the rotation as a real grid, Monday to Friday. The bucket is the job; the note is a sample angle you'd swap for your own.",[531,5567,5568,5589],{},[534,5569,5570],{},[537,5571,5572,5574,5577,5580,5583,5586],{},[540,5573],{},[540,5575,5576],{},"Mon",[540,5578,5579],{},"Tue",[540,5581,5582],{},"Wed",[540,5584,5585],{},"Thu",[540,5587,5588],{},"Fri",[547,5590,5591,5613,5635,5657],{},[537,5592,5593,5598,5601,5604,5607,5610],{},[552,5594,5595],{},[34,5596,5597],{},"Week 1",[552,5599,5600],{},"Expertise: a framework you use",[552,5602,5603],{},"Industry: react to news",[552,5605,5606],{},"Proof: a recent result",[552,5608,5609],{},"Build: what you shipped",[552,5611,5612],{},"Human: a weekend story",[537,5614,5615,5620,5623,5626,5629,5632],{},[552,5616,5617],{},[34,5618,5619],{},"Week 2",[552,5621,5622],{},"Proof: a client before\u002Fafter",[552,5624,5625],{},"Expertise: a common mistake",[552,5627,5628],{},"Human: a lesson from outside work",[552,5630,5631],{},"Industry: a hot take",[552,5633,5634],{},"Build: a work-in-progress",[537,5636,5637,5642,5645,5648,5651,5654],{},[552,5638,5639],{},[34,5640,5641],{},"Week 3",[552,5643,5644],{},"Expertise: answer a FAQ",[552,5646,5647],{},"Build: a decision you made",[552,5649,5650],{},"Proof: a testimonial in context",[552,5652,5653],{},"Industry: a trend you disagree with",[552,5655,5656],{},"Human: something you changed your mind on",[537,5658,5659,5664,5667,5670,5673,5676],{},[552,5660,5661],{},[34,5662,5663],{},"Week 4",[552,5665,5666],{},"Industry: a prediction",[552,5668,5669],{},"Proof: a small win",[552,5671,5672],{},"Expertise: a step-by-step",[552,5674,5675],{},"Human: a personal update",[552,5677,5678],{},"Build: what's next",[12,5680,5681],{},"Twenty slots, five buckets, no two the same twice in a row. You can shift the order to fit your week; the point is that every day already has a job.",[19,5683,5685],{"id":5684},"adapt-it-per-platform","Adapt it per platform",[12,5687,5688],{},"The buckets stay the same across platforms; the format changes.",[27,5690,5691,5701,5710,5716],{},[30,5692,5693,5696,5697,5700],{},[34,5694,5695],{},"LinkedIn."," Longer posts work. Lean into Expertise, Proof, and Human, and open every one with a strong first line (",[71,5698,5699],{"href":5117},"how to write a hook that earns the click",").",[30,5702,5703,5706,5707,5700],{},[34,5704,5705],{},"X (Twitter)."," Shorter and faster. Industry takes and Build-in-public fit best, and longer points become threads (",[71,5708,5709],{"href":1977},"thread structures that get read",[30,5711,5712,5715],{},[34,5713,5714],{},"Instagram."," Visual-first. Proof (before\u002Fafter) and Human (behind-the-scenes) translate best to images and carousels.",[30,5717,5718,5721],{},[34,5719,5720],{},"Newsletter."," Depth. One Expertise or Proof piece a week, expanded, with the shorter social posts feeding it.",[12,5723,5724],{},"Run the same five-bucket rotation, and let each platform take the version of the bucket that suits it.",[19,5726,5728],{"id":5727},"map-the-month-in-one-sitting","Map the month in one sitting",[12,5730,5731],{},"Don't plan day by day. Sit down once, open the grid, and drop bucket labels across four weeks. You're not writing yet, just assigning jobs. That's twenty small decisions made together, when you have context, instead of twenty decisions made cold at 8am.",[19,5733,5735],{"id":5734},"batch-the-writing","Batch the writing",[12,5737,5738],{},"Once the slots have jobs, write in batches by bucket. Three \"proof\" posts back to back is far faster than one of each, because you stay in the same headspace and reuse the same raw material. Block an hour, pick a bucket, and draft everything of that type for the fortnight.",[19,5740,5742],{"id":5741},"leave-room-to-react","Leave room to react",[12,5744,5745],{},"Keep one or two slots a week loose. When something real happens (a launch, a news moment, a customer win), it goes there. A calendar that's 100% pre-planned can't respond to the present, and the present is usually your best material. The loose slots are where the calendar meets real life.",[19,5747,5749],{"id":5748},"a-reusable-template","A reusable template",[12,5751,5752],{},"Copy this and refill it each month:",[27,5754,5755,5761,5767,5773,5779],{},[30,5756,5757,5760],{},[34,5758,5759],{},"Buckets:"," Expertise, Industry, Proof, Build, Human.",[30,5762,5763,5766],{},[34,5764,5765],{},"Grid:"," 4 weeks × your posting days, one bucket per slot, no repeats back to back.",[30,5768,5769,5772],{},[34,5770,5771],{},"Loose slots:"," 1–2 per week, left open for reactive posts.",[30,5774,5775,5778],{},[34,5776,5777],{},"Batch blocks:"," one writing session per bucket, per fortnight.",[30,5780,5781,5784],{},[34,5782,5783],{},"Feed loop:"," at month's end, note which posts landed and make more of that bucket next month.",[19,5786,4424],{"id":4423},[12,5788,5789,5792],{},[34,5790,5791],{},"How many times a week should I post?","\nWhatever you can keep up without dropping quality. Five weekdays is a full rotation; three is fine and beats five you can't sustain. Consistency matters more than frequency.",[12,5794,5795,5798,5799,5802],{},[34,5796,5797],{},"How far ahead should I plan?","\nPlan the buckets for four weeks; write one to two weeks ahead. Planning the ",[519,5800,5801],{},"jobs"," far out is cheap and stabilizing; drafting the exact words too far ahead means rewriting when the moment shifts.",[12,5804,5805,5808,5809,239],{},[34,5806,5807],{},"What if I run out of ideas for a bucket?","\nThat's usually a capture problem, not a creativity one. Keep a running note of raw ideas as they hit you, and shaping a week's posts becomes selection instead of invention. See ",[71,5810,5811],{"href":5349},"the idea-to-published pipeline",[12,5813,5814,5817,5818,5820],{},[34,5815,5816],{},"Do I need a tool for this?","\nA spreadsheet and a notes app run it fine for one brand. Across several clients, each needing its own grid, buckets, and drafts, keeping it all straight becomes its own job. That's where ",[71,5819,7],{"href":4457}," takes over: it plans the month, drafts each slot in the brand's voice with the AI tells stripped, and ships the posts, with Chloe running the calendar alongside you like a brand manager.",[12,5822,5823],{},"The goal is simple: never face a blank page cold. Give each day a job, batch the work, and leave a little space for real life to interrupt.",{"title":247,"searchDepth":248,"depth":248,"links":5825},[5826,5827,5828,5829,5830,5831,5832,5833,5834],{"id":5502,"depth":248,"text":5503},{"id":5515,"depth":248,"text":5516},{"id":5561,"depth":248,"text":5562},{"id":5684,"depth":248,"text":5685},{"id":5727,"depth":248,"text":5728},{"id":5734,"depth":248,"text":5735},{"id":5741,"depth":248,"text":5742},{"id":5748,"depth":248,"text":5749},{"id":4423,"depth":248,"text":4424},"2026-07-20","A five-bucket content calendar that keeps you posting without a blank page every morning: what to post, a filled four-week example, per-platform notes, and a reusable template.",{},{"title":5490,"description":5836},"blog\u002Fcontent-calendar-you-can-keep",[806,5841],"content calendar","eN-UW7Z51MqdV3_GT3y9Hp5saR1TaIgL7BNCblfsuFs",{"id":5844,"title":5845,"author":7,"body":5846,"date":5931,"description":5932,"draft":259,"extension":260,"image":261,"meta":5933,"navigation":263,"path":5349,"seo":5934,"stem":5935,"tags":5936,"updated":261,"__hash__":5937},"blog\u002Fblog\u002Fidea-to-published-pipeline.md","From idea to published: a repeatable social media pipeline",{"type":9,"value":5847,"toc":5921},[5848,5851,5855,5858,5862,5869,5873,5876,5880,5883,5887,5890,5894,5897,5901,5904,5908],[12,5849,5850],{},"When posting feels like a scramble, effort is rarely the problem. Every post starts from zero: a blank page, a fresh decision, a new negotiation. A pipeline fixes that by turning one big vague task (\"do social\") into a line of small clear ones, each with a defined start and finish. Here are the six stages.",[19,5852,5854],{"id":5853},"_1-capture","1. Capture",[12,5856,5857],{},"Ideas don't arrive when you sit down to write; they arrive in the shower, mid-meeting, reading something. If there's no place to put them, they're gone by evening. Stage one is a single inbox for raw ideas: a line, a link, a screenshot. No formatting, no judgment. Just catch them.",[19,5859,5861],{"id":5860},"_2-shape","2. Shape",[12,5863,5864,5865,5868],{},"Once a week, turn raw captures into ",[519,5866,5867],{},"briefs",": a title, an angle, the format. \"Thoughts on pricing\" becomes \"short post: why we raised prices and what we told customers.\" A brief is small enough to approve in seconds and clear enough to write from without re-thinking the whole thing.",[19,5870,5872],{"id":5871},"_3-draft","3. Draft",[12,5874,5875],{},"Now, and only now, you write, against a brief that already exists. This is the stage everyone starts with, and it's the one that goes fastest when the four stages around it are in place. You're not deciding what to say; you're saying a thing already decided.",[19,5877,5879],{"id":5878},"_4-review","4. Review",[12,5881,5882],{},"Every draft gets one pass for voice and one for approval. Voice: does this sound like the brand, or like a tool? Approval: is anyone else's yes required before it ships? Keeping these two checks explicit is what stops \"I'll just post it\" from becoming \"why did we post that.\"",[19,5884,5886],{"id":5885},"_5-schedule","5. Schedule",[12,5888,5889],{},"Approved content goes onto a calendar, not into a vague intention to post later. Seeing the week laid out is also how you catch problems: three posts on the same theme, a Friday with nothing, a gap before a launch.",[19,5891,5893],{"id":5892},"_6-publish-and-watch","6. Publish and watch",[12,5895,5896],{},"The post goes out, and then the loop closes: what happened? Not to obsess over every number, but to feed the next round of Capture. The post that worked tells you what to make more of.",[19,5898,5900],{"id":5899},"why-the-pipeline-beats-willpower","Why the pipeline beats willpower",[12,5902,5903],{},"Each stage is small and has an obvious next action, so nothing depends on motivation. You don't face \"do social,\" only \"shape this week's captures\" or \"draft these three briefs.\" Small, defined tasks get done; vague big ones get postponed. The pipeline is the trick of not letting a post become a vague big task.",[19,5905,5907],{"id":5906},"running-it-for-real","Running it for real",[12,5909,5910,5911,5913,5914,5917,5918,239],{},"For yourself, this can be a note app and a calendar. Across several clients, the six stages multiply: every brand needs its own capture, briefs, voice review, approvals, and calendar, and holding all of that in your head is where it breaks. That's the shape ",[71,5912,7],{"href":4457}," is built around: idea → draft in the brand's voice → approval → schedule → publish, with the replies worth making surfaced along the way. If clients and approvals are your specific pain, see ",[71,5915,5916],{"href":865},"the approval flow that keeps them","; if it's voices blurring, see ",[71,5919,5920],{"href":658},"keeping several voices straight",{"title":247,"searchDepth":248,"depth":248,"links":5922},[5923,5924,5925,5926,5927,5928,5929,5930],{"id":5853,"depth":248,"text":5854},{"id":5860,"depth":248,"text":5861},{"id":5871,"depth":248,"text":5872},{"id":5878,"depth":248,"text":5879},{"id":5885,"depth":248,"text":5886},{"id":5892,"depth":248,"text":5893},{"id":5899,"depth":248,"text":5900},{"id":5906,"depth":248,"text":5907},"2026-07-18","When posting feels chaotic, the fix is a pipeline. Six stages that turn \"what do I post today\" into a system you can run on repeat.",{},{"title":5845,"description":5932},"blog\u002Fidea-to-published-pipeline",[805],"1is-_4sv7cb2YQYuhRWgixVRTMdbvtyM8P0FnGp9Hd0",{"id":5939,"title":5940,"author":7,"body":5941,"date":6192,"description":6193,"draft":259,"extension":260,"image":261,"meta":6194,"navigation":263,"path":1977,"seo":6195,"stem":6196,"tags":6197,"updated":261,"__hash__":6199},"blog\u002Fblog\u002Fx-thread-structures.md","Four X (Twitter) thread structures that actually get read",{"type":9,"value":5942,"toc":6177},[5943,5946,5950,5953,5957,5960,5974,5977,5981,5985,5988,5991,5999,6002,6006,6009,6011,6019,6022,6026,6029,6031,6039,6042,6046,6049,6051,6059,6062,6066,6069,6086,6089,6093,6096,6100,6103,6107,6139,6141,6147,6153,6159,6165],[12,5944,5945],{},"A thread lives or dies on two things: whether the first tweet earns the tap, and whether each tweet pulls you into the next. Most threads nail neither. They open with a table of contents and then list points that don't need to be read in order. This is a guide to the structures that hold a reader from tweet one to the end, with example openers for each, a full worked example, the mistakes that lose people, and answers to the common questions.",[19,5947,5949],{"id":5948},"what-makes-a-thread-get-read","What makes a thread get read",[12,5951,5952],{},"A thread isn't a blog post chopped into tweets. It's a chain where every link has to earn the next tap. Two forces decide whether it works: the first tweet (does anyone start?) and the transitions (does anyone continue?). Structure is how you engineer both. The four below are the ones that reliably keep people moving.",[19,5954,5956],{"id":5955},"tweet-one-is-everything","Tweet one is everything",[12,5958,5959],{},"No structure survives a weak first tweet. It has to do one job, make stopping impossible, and it can't spend itself announcing the thread.",[27,5961,5962,5965,5968,5971],{},[30,5963,5964],{},"Weak: \"A thread on productivity 🧵\"",[30,5966,5967],{},"Weak: \"Here are some lessons I've learned about hiring.\"",[30,5969,5970],{},"Strong: \"I lost our biggest client on a Tuesday. Here's the email that did it.\"",[30,5972,5973],{},"Strong: \"We rewrote one onboarding email and activation doubled.\"",[12,5975,5976],{},"Start with the sharpest specific you've got: the result, the mistake, the surprising claim. Earn the tap, then worry about the rest.",[19,5978,5980],{"id":5979},"four-structures-that-hold-a-reader","Four structures that hold a reader",[4749,5982,5984],{"id":5983},"_1-the-story","1. The story",[12,5986,5987],{},"Start in the middle of a moment, then walk it forward. Each tweet is the next beat. A half-told story is a loop the brain wants closed, so the reader can't comfortably leave until they know how it ends.",[12,5989,5990],{},"Example openers:",[27,5992,5993,5996],{},[30,5994,5995],{},"\"I got fired the week we hit our best numbers ever.\"",[30,5997,5998],{},"\"A customer emailed to cancel. Twenty minutes later they upgraded.\"",[12,6000,6001],{},"Best for: lessons, failures, behind-the-scenes.",[4749,6003,6005],{"id":6004},"_2-the-list-with-teeth","2. The list with teeth",[12,6007,6008],{},"A numbered list, but only if each item earns its place and the items build. Five items that each land beat ten where item 7 is filler. Order them so momentum rises: a strong one first to prove the thread's worth, your best one last so the ending pays off.",[12,6010,5990],{},[27,6012,6013,6016],{},[30,6014,6015],{},"\"Five hiring mistakes that cost us a year. The last one still stings.\"",[30,6017,6018],{},"\"I've written 500 cold emails. Four rules got me every reply that mattered.\"",[12,6020,6021],{},"Best for: tactics, frameworks, lessons learned.",[4749,6023,6025],{"id":6024},"_3-the-teardown","3. The teardown",[12,6027,6028],{},"Take one specific thing (a post, a launch, a campaign) and break down why it worked or failed, piece by piece. Specificity is the whole appeal: \"here's exactly what this launch page did, line by line\" promises something a generic tips thread can't.",[12,6030,5990],{},[27,6032,6033,6036],{},[30,6034,6035],{},"\"This landing page converts at 8%. Let me show you every decision behind it.\"",[30,6037,6038],{},"\"A tweet of mine did 2M views. Here's the anatomy of why.\"",[12,6040,6041],{},"Best for: analysis, case studies, showing expertise.",[4749,6043,6045],{"id":6044},"_4-the-contrarian-case","4. The contrarian case",[12,6047,6048],{},"Open with a claim most people in your field would push back on, then spend the thread defending it. Disagreement is magnetic. People read to argue, and reading to argue is still reading.",[12,6050,5990],{},[27,6052,6053,6056],{},[30,6054,6055],{},"\"Most content calendars are why you post less, not more.\"",[30,6057,6058],{},"\"Your best hire will probably fail your interview process.\"",[12,6060,6061],{},"Best for: opinion, positioning, standing out in a crowded niche.",[19,6063,6065],{"id":6064},"a-full-example-the-story-thread","A full example: the story thread",[12,6067,6068],{},"Here's the story structure as a five-tweet chain, so you can see the beats and the transitions.",[439,6070,6071,6074,6077,6080,6083],{},[30,6072,6073],{},"\"I lost our biggest client on a Tuesday. It came down to one sentence in an email.\"",[30,6075,6076],{},"\"The account was healthy. Renewal was a formality. So I got comfortable, and I stopped over-communicating.\"",[30,6078,6079],{},"\"When a small delay came up, I sent a quick 'sorry, running behind, will update soon.' No plan, no date. Just an apology.\"",[30,6081,6082],{},"\"To them, that one line read as 'this vendor isn't in control.' They'd been quietly nervous for weeks, and I'd just confirmed it.\"",[30,6084,6085],{},"\"The fix wasn't working harder. It was never sending an update without a next step and a date attached. We haven't lost an account to silence since.\"",[12,6087,6088],{},"Notice each tweet ends slightly unresolved, and the last one lands the single lesson and stops. That's the whole engine.",[19,6090,6092],{"id":6091},"transitions-matter-more-than-the-tweets","Transitions matter more than the tweets",[12,6094,6095],{},"Whatever structure you pick, the seam between tweets is where readers leave. End tweets on a small forward pull: an unfinished thought, a \"but here's the part nobody mentions,\" a number you haven't explained yet. A tweet that resolves completely gives the reader permission to stop. Leave one thread hanging into the next.",[19,6097,6099],{"id":6098},"the-last-tweet","The last tweet",[12,6101,6102],{},"Land the point. If you want engagement, earn it with the thread. Begging for it in the final tweet (\"what did you think?\") undoes the momentum you just built. A strong close restates the one idea worth remembering and stops.",[19,6104,6106],{"id":6105},"mistakes-that-kill-a-thread","Mistakes that kill a thread",[27,6108,6109,6115,6121,6127,6133],{},[30,6110,6111,6114],{},[34,6112,6113],{},"A table-of-contents opener."," Spending tweet one on \"here's what this thread covers.\" Open with the content instead.",[30,6116,6117,6120],{},[34,6118,6119],{},"Padding to hit a number."," Ten tweets when you have five. Cut to the ones that land.",[30,6122,6123,6126],{},[34,6124,6125],{},"Points in any order."," If the tweets don't build, it's a list of tweets, not a thread. Sequence for rising momentum.",[30,6128,6129,6132],{},[34,6130,6131],{},"Resolving every tweet."," Each fully-closed tweet is an exit. Leave a pull into the next.",[30,6134,6135,6138],{},[34,6136,6137],{},"A begging finale."," \"Retweet if you agree\" undercuts a strong ending. Let the content earn the share.",[19,6140,4424],{"id":4423},[12,6142,6143,6146],{},[34,6144,6145],{},"How long should a thread be?","\nAs long as it stays interesting and no longer. Five to eight tweets is a common sweet spot, but a tight four beats a padded twelve. Cut anything that doesn't move the story or the argument forward.",[12,6148,6149,6152],{},[34,6150,6151],{},"Does the hook matter more than the content?","\nThe hook decides whether anyone reads; the content decides whether they finish and share. You need both, and the hook comes first because nothing else runs if tweet one fails.",[12,6154,6155,6158],{},[34,6156,6157],{},"Should every tweet have a line break or emoji?","\nFormatting helps readability, but it won't rescue a weak structure. Get the sequence and transitions right first; treat formatting as polish.",[12,6160,6161,6164],{},[34,6162,6163],{},"How do I end a thread without begging?","\nRestate the one idea worth remembering, in a single strong line, and stop. If you want people to follow or share, a natural close earns it better than a request.",[12,6166,6167,6168,6170,6171,6173,6174,239],{},"Threads are a format; a voice is what makes anyone care about yours. If your drafts read like every other thread in the niche, the structure won't save them. SelfSM drafts threads in each brand's voice with the AI tells already stripped, so the structure and the sound are handled in one pass, and Chloe helps shape the thread before it goes out. ",[71,6169,4458],{"href":4457},". For the craft itself, start with ",[71,6172,2635],{"href":125},"; to nail the opener, the same rules drive ",[71,6175,6176],{"href":5117},"a LinkedIn hook that earns the click",{"title":247,"searchDepth":248,"depth":248,"links":6178},[6179,6180,6181,6187,6188,6189,6190,6191],{"id":5948,"depth":248,"text":5949},{"id":5955,"depth":248,"text":5956},{"id":5979,"depth":248,"text":5980,"children":6182},[6183,6184,6185,6186],{"id":5983,"depth":5098,"text":5984},{"id":6004,"depth":5098,"text":6005},{"id":6024,"depth":5098,"text":6025},{"id":6044,"depth":5098,"text":6045},{"id":6064,"depth":248,"text":6065},{"id":6091,"depth":248,"text":6092},{"id":6098,"depth":248,"text":6099},{"id":6105,"depth":248,"text":6106},{"id":4423,"depth":248,"text":4424},"2026-07-16","Most threads lose people by tweet two. Four structures that hold a reader, with example openers, a full example thread, common mistakes, and an FAQ.",{},{"title":5940,"description":6193},"blog\u002Fx-thread-structures",[6198,270],"x","HKBfxZMFLEhYPVxqZ2GpOEts8Ta4rbzKxLFNFizJdrs",1788343316279]