There is no universally correct LinkedIn format, and anyone selling one is selling a template that will be recognisable as a template within a month.
But there is a structural logic that substantive posts share, and understanding it lets you write in your own voice without guessing. It comes down to a single constraint: on LinkedIn you are competing with the reader’s thumb, and you have roughly one line before they decide.
Everything below serves that constraint.
The four-move structure
1. Hook: a claim, tension or specific that earns the next line
2. Argument: the actual thinking, delivered as reasoning not summary
3. Proof: the detail that could only come from someone who did this
4. Close: a question, a position, or a deliberate stop
Move 1: the hook
The hook’s only job is to make line two worth reading. It is not a headline, and it is definitely not a curiosity gap (“you won’t believe what happened next”). That style performs briefly and costs credibility permanently with exactly the audience you want.
Hooks that work are usually one of four things:
- A specific number with context. “We turned down four inbound deals last quarter” is more interesting than “how to qualify leads” because it implies a story and a willingness to say something.
- A disagreement. “Most personal brand advice is written for people who already have one.”
- A reversal. “I spent a year posting daily and it made my pipeline worse.”
- A precise observation. “Every SaaS pricing page I read this month answered a question nobody was asking.”
Note what is absent: a description of what the post will contain. “In this post I’ll share 5 tips about…” tells the reader the post is generic before they read a word of it.
Move 2: the argument
This is where most posts fail, usually by summarising instead of reasoning.
Compare:
Summary: “Consistency is key to building a personal brand. Post regularly and provide value to your audience.”
Reasoning: “Consistency works because of how feeds distribute. A post from someone who published four times this week is shown to people who engaged with them last week. A post from someone who published once in three months is shown to nobody. The mechanism isn’t discipline, it’s that the platform has nothing to work with.”
The first is a thing everyone already believes. The second explains why, and a reader can now use it. Explanation is what gets a post saved rather than scrolled.
Practical guidance:
- One idea per post. Two ideas produce two mediocre posts and one abandoned draft.
- Front-load the substance. Readers who reach the “see more” fold are a subset; make the first visible lines carry real content.
- Short paragraphs. One to three lines. Dense blocks are skipped on mobile, which is where most of your audience is.
- Write like you speak to a competent colleague. That means subordinate clauses, qualifications and the occasional blunt sentence, not motivational poster phrasing.
Move 3: the proof
The move that separates credible content from competent content. Something that could only have come from someone who actually did the work:
- The specific number, with its conditions
- What you tried first and why it failed
- The trade-off you accepted knowingly
- The thing a competitor or a popular framework gets wrong
- A detail small enough to be un-inventable: the objection a client raised, the wording that changed a response rate, the week you considered stopping
This is what makes a post attributable to you. It is also what ghostwriters most often cannot manufacture, which is why interview-led writing outperforms brief-led writing on this platform.
Move 4: the close
Three options, in descending order of usefulness:
- A stated position you are willing to defend. Ends the post on a claim, which invites substantive disagreement, the highest-value engagement on LinkedIn.
- A genuine question the author actually wants answered. Rare. If you already know the answer, do not ask.
- A deliberate stop. Sometimes the argument is complete. Ending cleanly beats appending “Thoughts?” to a finished point.
What to avoid: engagement-bait closers (“agree?”, “repost if you agree”, “what do you think, comment below”). They generate activity from the wrong people, and the right people (senior buyers evaluating whether you are serious) notice them.
Formats, honestly assessed
| Format | When it works | When it doesn’t |
|---|---|---|
| Text + no image | Strong reasoning; treats the feed as reading | When the claim genuinely needs a visual |
| Document/carousel | Frameworks, processes, before/after | When slide 1 is just a prettier headline |
| Text + single image | Personal, concrete, photographic evidence | Stock photography, which reads as filler |
| Native video | You are genuinely compelling in the first two seconds | When the same content would be better read |
| Poll | Genuine open questions with no obvious answer | When the answer is obvious and the poll is just reach bait |
| Repost with comment | Adding real analysis to someone else’s point | When it is a substitute for having your own |
The best-performing format for most B2B professionals writing substantive material is the plain text post. This annoys people who have bought design tools. It is nonetheless what the evidence shows repeatedly: on a professional network, the argument is the creative asset.
The failure modes
Five patterns that reliably underperform, all of which are easy to fix:
- The résumé post. “Excited to share that I’ve been promoted to…” Announcements are for your existing network; they teach nobody anything and reach almost nobody new. If you must announce, attach a lesson learned.
- The listicle with no spine. Seven tips, each of which could appear in any of ten thousand other posts. Lists work when each item contains a specific and slightly surprising claim.
- The borrowed take. Repackaging an idea from a podcast or a book without adding your own reasoning or experience. Readers can tell, and it positions you as a curator rather than a practitioner.
- The humblebrag. “We accidentally grew to 40k users” followed by three points that do not follow from the story. Just state the result and explain what actually caused it.
- The ghost post. No point of view anywhere in the text, every sentence agreeable, nothing arguable. Safe posts are invisible. That is not a rule of the algorithm; it is a rule of human attention.
A checklist before you publish
- Does line one create a reason to read line two?
- Am I explaining why, or just asserting what?
- Is there at least one detail that could only come from my actual experience?
- Would someone in my target audience disagree with anything here? If not, is there a real position in it?
- Could a competitor have written this verbatim?
- Would I be comfortable if my most sceptical reader (or my compliance reviewer) saw it?
- Does the close add something, or is it a reflex?
If the answer to the last two is no, the post is not ready. Publishing is not the deadline. Publishing something you would not defend is worse than publishing late.
Where to go next
- LinkedIn ghostwriting, how this structure is produced at cadence without losing your voice.
- LinkedIn personal branding, the strategy this writing sits inside.
- How to brief a LinkedIn ghostwriter, what to hand over so drafts come back right the first time.