AI LinkedIn Post Generator
A story, a numbered list or a contrarian take — written with a one-line hook, paragraphs short enough to survive a scroll, and a question a stranger can answer.
The AI LinkedIn Post Generator writes a post from a topic and an experience, with a hook that survives the three-line cut and at most ten hashtags. LinkedIn shows about three lines before “see more”, so the first sentence has to earn the click.
How it works
Describe the idea
Write the lesson or the story from your work.
Pick the format
Choose a story, a list or a contrarian take.
Copy the post
The ten hashtags come on one line under it.
The collapse happens after two lines
Whatever is above “see more” decides whether the post is read. Not a greeting, not a preamble, not a reflection on how busy the week was — the claim, the number, or the moment it went wrong.
“We lost the biggest client in March” earns the click. “Reflecting on resilience” does not.
- check_circleTwo lines that stop a scroller
- check_circleNo paragraph longer than two lines
- check_circleThe question is answerable in one sentence
- check_circleFive hashtags, on their own line
- check_circleNo link in the body
menu_book Storyexpand_more
A moment, a mistake, what changed. Works because it is specific and the reader can map it onto their own week.
format_list_numbered Numbered listexpand_more
Three lessons, one line each. Promises a finishable read, which is exactly what a half-attentive audience needs.
swap_horiz Contrarian takeexpand_more
Only if you can defend it in the replies. A take you will abandon at the first objection damages more than it earns.
LinkedIn questions
How long should a LinkedIn post be?expand_more
Do hashtags help?expand_more
Should I add a link?expand_more
Is a generated post obvious?expand_more
The one platform where everyone has to be there
LinkedIn is the only large social platform where the audience is professionally required to be there. People open it between meetings, half-reading, looking for something that sounds like a colleague's experience rather than an advertisement. That is why the writing that works here is the opposite of what works on a visual feed: shorter, plainer, and more specific about what actually happened.
The first two lines are the whole post
Feeds collapse long posts behind a “see more” link, and the collapse happens after roughly two lines. Everything above that line decides whether the rest is read — which means the opening cannot be a greeting, a context paragraph, or a sentence about how busy the week has been. It has to be the claim, the number, or the moment of failure.
“We lost the biggest client in March” earns the click on “see more”. “Reflecting on the importance of resilience in business” does not, because the reader already knows what the rest will say, and they are right.
Why paragraphs are one or two lines
A wall of text signals effort to a reader who is scrolling in a lift. Short paragraphs are read, long ones are scrolled past, and this is a formatting fact rather than a stylistic preference. Break the post every two sentences, leave a blank line, and let the white space do the work that punctuation cannot.
The same reasoning explains why lists perform: they promise a finishable read. A numbered list of three lessons can be consumed in fifteen seconds, and the reader knows that before starting.
The question at the end is not decoration
Replies are what push a post into other feeds, and the easiest reply to write is the answer to a specific question. “What would you have done?” gets answered. “Thoughts?” does not, because it asks the reader to do the thinking the writer skipped.
Keep the question narrow enough that a stranger can answer it in one sentence from their own experience. Broad questions attract opinions from the loudest few; narrow ones attract the people you actually want to hear from.
What to leave out
Hashtags beyond a handful dilute rather than help — five on one line at the end is the working convention. Engagement prompts, external links in the body, and long preambles about the importance of the topic all reduce reach and are, more importantly, tedious. A post that begins with “I don't usually post here” has told the reader it is about to waste their time.
Emojis are punctuation, not decoration. Two, at most, in a post; one is usually enough. Used as bullets they are useful; used as rhythm they read as a template.
Posting, and what to do afterwards
Comment early and often on your own post: each reply adds context the original could not fit, and the thread becomes more useful than the post. Answer questions with a detail rather than agreement, and ignore the reflexive contrarians, who are content rather than audience.
Then leave it alone for a week. A post that has stopped performing can be reshaped and published again with a different opening, and almost nobody will remember the first version. Recycling is normal here; the only thing that damages credibility is repeating it verbatim.
What to do when nobody replies
A post with no replies has usually asked a question that requires effort, or made a claim that requires agreement before anyone can respond. Rewrite the ending first: a question about a specific situation — “what did you do the week after losing a client?” — produces answers where “thoughts?” produces nothing. Then look at the first two lines again; if they describe the topic instead of a moment, the reader has no reason to open the post at all. Change one of those two things, publish again a week later, and compare.
The three-post test
If you are unsure whether an idea is worth posting, write it three ways: as a story, as a list, and as a question. Publish the one that takes the least explaining. If all three need a paragraph of context before the point arrives, the idea is probably an article rather than a post, and it will do better on a page where the reader has already chosen to read. Posting less, with more evidence behind each one, is the pattern that survives a year of doing this, and it is also the only one that still feels worth doing in month twelve.
More social tools social tools
fact_checkQuick facts
- LinkedIn shows about three lines before “see more”, so the first sentence has to earn the expand.
- One post makes one point: a post that carries three ideas is read as a newsletter and scrolled past.
- A language model reads text in tokens, and one token is roughly four characters of English, so a 300-word answer is about 400 tokens.menu_bookOpenAI Cookbook
- The output depends on the prompt: the same request with a different tone or length instruction produces a different text, not a different answer quality.menu_bookOpenAI Cookbook
- A single emoji can be several code points joined together, so “count the characters” gives a different number from “count what you see”.menu_bookUnicode Consortium
helpQuestions people ask
How do I write a LinkedIn post that gets engagement?expand_more
How long should a LinkedIn post be?expand_more
Should I post AI-written content on LinkedIn?expand_more
hubWhat this page is about
- Large language modelopen_in_new
A model that predicts the next token of a text; the site sends the prompt to a hosted model and returns its answer.
- Unicodeopen_in_new
The standard that gives every character of every script a code point, and defines what counts as one user-perceived character.
library_booksOfficial sources cited on this page
- How to count tokens with tiktokenOpenAI Cookbook
- The Unicode StandardUnicode Consortium
Keep going with the next tool
The tools in this group that do the next piece of the same work, all switched on right now.
Switched-off tools are not listed here: only what works right now.