Posts, threads and repurposing, without the LinkedIn-voice tell. Below are 8 copy-ready prompts. Fill in the [BRACKETS], copy, and paste into ChatGPT, Claude, Gemini or any capable assistant.
Social is where generated text is most instantly recognisable, because the platform conventions are narrow and models reproduce them exactly. The one-line-paragraph, the rhetorical question, the emoji bullet — readers pattern-match it in under a second.
The 8 prompts
Write a LinkedIn post that is not insufferable
Post professionally without the performative style the platform rewards.
Write a LinkedIn post. WHAT I WANT TO SAY: [THE SUBSTANCE] WHY IT MATTERS TO MY AUDIENCE: [RELEVANCE] WHO FOLLOWS ME: [AUDIENCE] MY ROLE AND CREDIBILITY ON THIS: [CONTEXT] GOAL: [start a conversation / share knowledge / build reputation / promote something] MY USUAL TONE: [PASTE A PREVIOUS POST, or describe] Write the post so that: - The first two lines contain the whole hook. LinkedIn truncates after roughly two lines, and that is what determines whether anyone expands it. - It says one thing and says it properly - Any story is real and told briefly, with the point made explicitly rather than left as a moral - It is specific. Numbers, examples, names of things. - It ends with either a real question or nothing. Not both. Hard rules - these are the patterns that make LinkedIn unreadable: - No one-line paragraphs stacked for effect throughout the whole post - No 'Here's what I learned:' followed by five platitudes - No fake vulnerability opening ('I was rejected 47 times...') - No 'Agree?' or 'Thoughts?' as a closing - No manufactured story that ends in a business lesson - No 'Unpopular opinion:' for a popular opinion - No hashtag wall. Three maximum, or none. - No emoji bullet points - Do not tag people who are not involved Then give me: - THE FIRST TWO LINES, isolated, so I can judge the hook - THE HONEST CHECK - is there actually something here worth posting? If my substance is thin, say so rather than inflating it into a post. Posting nothing is better than posting filler. - A SHORTER VERSION - roughly half the length, in case the long one is padded.
What you get: A specific, single-point post with the hook isolated, a shorter alternative, and an honest check on whether it is worth posting.
Tip: The honest check is the most useful output. LinkedIn rewards posting frequency, which is exactly why most of it is filler.
Build a content calendar you can sustain
Plan social content without needing a new idea every day.
Build a social media content plan. PLATFORMS: [WHERE YOU POST] WHAT MY BUSINESS DOES: [CONTEXT] AUDIENCE: [WHO AND WHAT THEY WANT] TIME AVAILABLE PER WEEK: [HOURS] WHAT I ALREADY PRODUCE: [BLOG POSTS, PROJECTS, CUSTOMER WORK, ANYTHING] GOAL: [awareness / leads / community / recruitment] WHY PREVIOUS ATTEMPTS FAILED: [IF APPLICABLE] Produce: 1. FREQUENCY REALITY CHECK - given my hours, what cadence can I sustain for a year? Be conservative. Most social plans fail in week five from over-commitment, not from bad content. Recommend fewer posts than I want. 2. PLATFORM CHOICE - should I be on all the platforms I listed? Concentrating on one platform properly beats being mediocre on four. Say which to drop. 3. CONTENT PILLARS - 3-4 recurring themes, derived from my business and audience. Each with: what it covers, why the audience cares, and roughly what proportion of posts. 4. REPEATABLE FORMATS - the actual point of a content plan. 5-6 post formats I can refill indefinitely without inventing a new idea each time. For example: a question customers ask, something that went wrong and what changed, a short breakdown of a piece of work, a contrarian take on a common practice, a useful thing found elsewhere with commentary. For each: the structure, roughly how long it takes to make, and where the raw material comes from. 5. THE SOURCE PIPELINE - where content comes from in my existing work. A plan requiring separate content creation is a plan that stops. Describe the capture habit. 6. THE CALENDAR - four weeks, mapped to formats and pillars, with specific topics for week one. 7. THE BATCHING PLAN - how to produce a week's posts in one session. 8. WHAT TO CUT from my ambitions - anything that costs more than it returns: daily posting, video if I have no capacity, being on every platform, or designed graphics. 9. THE MEASUREMENT - what actually indicates this is working for my stated goal. Followers usually is not.
What you get: A sustainable cadence, repeatable post formats with source pipelines, a four-week calendar and an explicit cut list.
Tip: Point 4 is what makes social sustainable. Formats you refill beat ideas you invent, because the blank page is what kills consistency.
Respond to a negative comment or a brewing pile-on
Handle criticism in public without making it worse.
Help me respond to this. WHAT WAS POSTED: """ [PASTE THE COMMENT OR POST] """ PLATFORM: [WHERE] WHO POSTED IT: [customer / stranger / competitor / employee / influential account] IS IT ACCURATE: [yes / partly / no / cannot verify] WHAT ACTUALLY HAPPENED: [YOUR SIDE] HOW MUCH TRACTION IT HAS: [ignored / some engagement / spreading] WHAT WE CAN DO ABOUT IT: [IF ANYTHING] Produce: 1. SHOULD YOU RESPOND AT ALL - the first question, answered honestly. Responding amplifies. Criteria: is it accurate, does it have traction, is the person actually a customer, is silence being read as admission? Sometimes the correct answer is to do nothing, and it is the hardest one to accept. 2. WHO YOU ARE WRITING FOR - not the commenter. The people reading the exchange. Every word should be chosen for them. 3. THE RESPONSE - if you should respond: - Acknowledge the specific point, not criticism in general - If we were wrong, say so plainly and first. Defensiveness costs more than the original complaint. - Correct a factual error once, neutrally, without arguing. Never twice. - Move it to a private channel with a specific named route - Under 80 words. Long responses look defensive regardless of content. 4. WHAT NOT TO SAY - the tempting replies that would escalate: the point-by-point rebuttal, the 'actually', the passive-aggressive thanks, the legal hint, anything that questions the person's motives. 5. IF IT IS SPREADING - when to move from a comment reply to a proper statement, and what that statement needs. 6. THE UNCOMFORTABLE QUESTION - is there something true in this criticism, even if the framing is unfair? If so, the response is different and the fix matters more than the reply. 7. WHAT NOT TO DO - do not delete it unless it violates a clear policy, do not block a genuine customer, do not respond from a personal account, do not get colleagues to pile in. Do not draft a defensive response. If my side of the story is weak, tell me.
What you get: A should-you-respond judgement, a short reply written for onlookers, escalating replies to avoid, and an honest check on whether the criticism has a point.
Tip: Point 1 is the discipline. Most social media crises are created by the response, not the original complaint.
Write a thread that people read to the end
Structure a multi-post thread with momentum.
Write a thread. PLATFORM: [X / Threads / LinkedIn / Bluesky] TOPIC: [WHAT IT IS ABOUT] MY POINT: [WHAT YOU ARE ARGUING OR TEACHING] MY EXPERIENCE WITH THIS: [WHY YOU CAN SPEAK TO IT] THREAD LENGTH: [NUMBER OF POSTS] AUDIENCE: [WHO FOLLOWS YOU] Produce: 1. THE FIRST POST - this determines whether anyone reads the rest. It must contain a complete, interesting claim on its own, not a promise of one. 'Here's what I learned about X 🧵' is a promise; 'X costs three times what everyone assumes, and here is the arithmetic' is a claim. Give three options. 2. THE THREAD - each post numbered, each: - Standing alone as a complete thought - Ending in a way that makes the next one worth reading - Under the character limit (state the count) - Containing something specific, not a transition 3. THE STRUCTURE - build it as: claim, why it matters, the evidence in steps, the complication or objection, the resolution, and what to do about it. Do not structure it as a list of tips; lists have no momentum and readers drop out at post four. 4. THE DROP-OFF POINTS - the two posts where readers are most likely to leave, and what to do about them. Usually: post two, if it is throat-clearing, and the middle, if it becomes a list. 5. THE LAST POST - the actual conclusion. Not a call to follow. If there is a link, it goes here. 6. THE STANDALONE TEST - would the first post work as a single post on its own? If yes, that is a good sign. If the thread is really one idea stretched to twelve posts, say so and give me the single post instead. Rules: - No 🧵 emoji, no 'a thread:', no 'let's dive in' - No 'bookmark this' or 'RT if you found this useful' - Do not number posts as '1/12' unless the order genuinely matters - Every post must contain content. None may exist purely to set up the next.
What you get: A claim-led opening post with alternatives, a momentum-built thread, drop-off points identified and a check on whether it should be one post.
Tip: Point 6 is worth running every time. Most threads are one good post padded to twelve, and the single post performs better.
Repurpose one piece of work into a month of posts
Get maximum output from content you already made.
Turn this into multiple social posts. THE SOURCE: """ [PASTE ARTICLE, TALK, REPORT, PROJECT DESCRIPTION, OR CASE STUDY] """ PLATFORMS: [WHERE YOU POST] MY AUDIENCE: [WHO] MY TONE: [PASTE A SAMPLE POST, or describe] HOW MANY POSTS I WANT: [NUMBER] OVER WHAT PERIOD: [TIMEFRAME] Produce: 1. THE IDEA EXTRACTION - every distinct idea in the source that could stand alone as a post. There are usually more than expected: a claim, a counter-claim, a method, a mistake, a number, a story, an objection handled, a definition, something you changed your mind about. List them all before writing any. 2. RANK THEM - by how interesting each is on its own, to my audience. The most interesting idea is often buried in the middle of the source, not the headline. 3. THE POSTS - for the top [NUMBER], each: - Written natively for its platform - Delivering value without requiring the reader to click through - With the specific detail that makes it worth reading - Never opening with 'I wrote about this recently' 4. THE SCHEDULE - spread over my timeframe, ordered so that the strongest goes first and related ideas are not adjacent. 5. THE VARIETY CHECK - if all the posts are the same format, they will read as a campaign. Vary: a claim, a story, a practical tip, a question, a short observation. 6. WHICH ONES LINK BACK - not all of them. Posts that always drive to a link train the audience to scroll past. Recommend which two or three should carry a link. 7. WHAT IS LEFT OVER - ideas from the extraction that did not make this batch, banked for later. Rules: - Each post must be useful to someone who never sees the source - Do not summarise the source in each post. Extract one idea and develop it. - No 'new post on the blog' announcements
What you get: A full idea extraction ranked by interest, natively written posts, a spread schedule and leftover ideas banked.
Tip: Point 2 usually reorders things. The most interesting idea in an article is rarely the one in its title.
Write captions that do work
Pair a caption with an image or video so both earn their place.
Write captions. PLATFORM: [Instagram / LinkedIn / Facebook / other] WHAT THE VISUAL SHOWS: [DESCRIPTION] CONTEXT: [WHAT THIS IS, WHY YOU ARE POSTING IT] MY BUSINESS: [WHAT YOU DO] AUDIENCE: [WHO] GOAL: [engagement / traffic / awareness / sales] TONE: [DESCRIPTION OR SAMPLE] Produce four caption options, each a different approach: 1. THE CONTEXT - what the visual does not show. The story behind it, what happened before or after, what it cost. 2. THE USEFUL - a practical takeaway connected to the image. 3. THE SHORT - one or two lines. Sometimes the image carries it and a long caption is noise. Include this option even if it feels too short. 4. THE QUESTION - a genuine one, that a reader would actually have an answer to. For each: character count, and one line on when this approach is right. Rules: - Never describe what is visible. The caption must add something. - The first line must work alone; platforms truncate. - No 'link in bio' as the opening - No hashtag wall. Three to five relevant ones, or none. Say which for this platform. - No engagement bait: 'double tap if', 'comment below', 'tag a friend who' - No emoji bullet lists - Do not use a question as a closing unless you will reply to answers Then: - WHICH ONE for my stated goal, and why - THE FIRST LINE of each, isolated - THE HONEST CHECK - does this post need a caption at all, or is the visual enough?
What you get: Four caption approaches with character counts, isolated first lines, a goal-matched recommendation and a check on whether a caption is needed.
Tip: The short option is included deliberately because it is usually right and rarely chosen. A strong image with two lines outperforms the same image with a paragraph.
Analyse what is working in your social performance
Learn from your own data instead of guessing.
Analyse my social media performance. PLATFORM: [WHERE] POST DATA: """ [PASTE: DATE, POST TEXT OR TOPIC, FORMAT, REACH, ENGAGEMENT, CLICKS, FOLLOWERS GAINED] """ PERIOD: [TIMEFRAME] MY GOAL: [WHAT SUCCESS MEANS TO ME] AUDIENCE SIZE: [FOLLOWERS] Produce: 1. WHAT ACTUALLY WORKED - the top posts by the metric that matters for my stated goal, not by raw engagement. If my goal is leads, a post with 400 likes and no clicks did not work. 2. THE PATTERNS - across the top performers: format, topic, length, hook style, time posted, whether they had a link. State how confident you are; with a small sample, most apparent patterns are noise. Say so explicitly if my data is too thin to conclude anything. 3. THE UNDERPERFORMERS - what the bottom posts have in common. 4. THE CONFOUNDERS - reasons a post may have performed well or badly that have nothing to do with its quality: it was shared by a large account, it coincided with news, the platform tested it on a wider audience, it was posted at an unusual time. Flag any post where the result may not be about the content. 5. ENGAGEMENT VS OUTCOME - posts that got engagement but did nothing for my goal, and posts that quietly drove the outcome with modest engagement. The second group is the one to learn from and the easiest to overlook. 6. WHAT TO DO MORE OF - specific and testable, phrased as something I can act on this week. 7. WHAT TO STOP - formats or topics not earning their production time. 8. WHAT TO TEST - the two things my data suggests but does not confirm. 9. THE SAMPLE SIZE WARNING - how many posts I would need before these conclusions are reliable. Be honest; with under 20 posts, most of this is provisional. Do not manufacture insights from thin data. 'Not enough data to say' is a legitimate finding.
What you get: Performance judged against your actual goal, patterns with confidence levels, confounders flagged and an honest sample-size warning.
Tip: Point 5 is where the useful finding usually is. The post with modest engagement that drove a third of your enquiries is invisible on the engagement leaderboard.
Plan a month of local business social posts
Build a local-first social calendar for a real estate agent, shop, clinic or service business.
Plan a month of social media posts for my local business. BUSINESS: [TYPE AND AREA SERVED] PLATFORMS: [e.g. Instagram, Facebook, TikTok, Google Business Profile] POSTS PER WEEK: [NUMBER] WHAT I CAN SHOW: [e.g. properties, team, products, before/after, customers who agreed to feature] LOCAL EVENTS OR SEASONS THIS MONTH: [LIST] Create a calendar table: Date | Platform | Format | Post idea | Caption hook | Call to action Mix: - Local knowledge posts (neighbourhood guides, local tips) - these build trust and get shared - Proof posts (real results, reviews, sold/finished work) - only real ones - Personality posts (team, behind the scenes) - Offer or listing posts - no more than 1 in 4 Also give me 5 reusable caption templates and a note on posting to Google Business Profile, which helps local search visibility.
What you get: A month-long local social calendar with a balanced content mix, caption templates and a Google Business Profile note.
Tip: Neighbourhood guide posts get shared by people who are not buying yet - which is how you become the name they remember when they are.
Where AI actually helps here
- Repurposing one piece of work into several formats
- Ten openings for a post so you can pick one
- Adapting a message to each platform’s actual conventions
Where it falls down
- Anything requiring a take. Opinion is the currency and it has none
- Current context. It does not know what happened this week
- Sounding human on LinkedIn, where the default output is most obvious
The mistake almost everyone makes: Leaving the tells in
Ban them explicitly: no one-line paragraphs stacked for drama, no opening rhetorical question, no emoji as bullets, no ‘Here’s the thing:’, no ‘Let that sink in’. What is left is usually thin — and that tells you the post needed a real point, which is the actual finding.
Free tool: Output Grader
Runs in your browser. No sign-up, nothing uploaded.
Questions people ask
Why does AI social content get low engagement?
Because the format is recognisable and the content has no point of view. Engagement follows a person having an opinion, which is the one input a model cannot supply.
How do I repurpose a blog post into a thread?
Do not ask for a summary. Ask it to find the single most arguable claim in the piece and build the thread around defending that. A summarised thread reads like homework.