Every few months someone announces that LinkedIn is "pushing video again", or that carousels are dead, or that plain text is the only honest format left. Then a thousand consultants change their whole publishing habit on the strength of a rumour.
That is the wrong question. Format is not a lottery ticket. It is a delivery mechanism. The useful question is: what do I want this specific reader to understand, feel or do by the end? Answer that, and the format picks itself.
Here is how to choose LinkedIn content formats by intent, with the trade-offs each one costs you in effort.
Format follows intent, not reach
Reach is a side effect. What reaches is what gets read, and what gets read is what matched what the reader wanted in that second.
So before you open a design tool, name the job of the post. In practice there are only about five jobs a B2B post ever does:
- Make someone think differently about a problem they already have.
- Teach a procedure they could run themselves tomorrow.
- Prove you can do the work — a before/after, a diagnosis, a teardown.
- Show who you are so the right people feel safe booking a call.
- Trigger a conversation — an opinion, a question, a line that provokes a reply.
Each of those jobs has a format that carries it best. Mismatch them and the post dies quietly: a procedure crammed into a text post becomes a wall nobody finishes, an opinion stretched into a ten-slide carousel becomes a lecture.
One mechanism underlies all of it. LinkedIn can only see behaviour — whether people stopped, stayed, expanded, commented. That is why how long people stay on your post matters more than the count of likes. Formats differ mainly in how much time they ask for and how reliably they get it.
Text posts: thinking, opinion, conversation
Plain text is the lowest-friction format on the platform. No download, no sound, no swipe. The reader is already reading before they decided to.
Use text when the value is a thought. An opinion that splits the room. A reframe of something your market takes for granted. A short story from a client engagement with one lesson at the end. A question you genuinely want answered in the comments.
Text is also the only format where the first two lines do all the work. Nothing else is visible before the expand. If you write one text post a week, spend a disproportionate share of your time on the opening — there are twenty opening lines that earn the second line if you need a starting point.
What text costs you: thinking time, not production time. Thirty minutes of honest thought, ten minutes of typing. That is why it is the right default for a solo consultant with client work to deliver.
What text is bad at: anything with steps, numbers, or a visual comparison. If you find yourself writing "first... second... third... fourth... fifth", you have written a carousel and formatted it as a paragraph.
Carousels: procedures, frameworks, before and after
Document posts — a PDF uploaded and swiped through in the feed — are the format for sequence. Anything where slide 3 only makes sense after slide 2.
They work for B2B because they ask for a specific, cheap action: swipe. Each swipe is another second spent, and the reader gets a small reward each time. That is the whole mechanism. No magic boost, just a format that is unusually good at holding attention if the content deserves it.
Good carousel subjects:
- A method with named steps (the audit you run, the five questions you ask in a first call).
- A comparison: what most people do / what actually works, side by side.
- A teardown: a weak profile headline on one slide, your rewrite on the next.
- A checklist a prospect can screenshot and use on Monday.
Rules that matter more than design. One idea per slide — if a slide has two ideas, split it. Readable on a phone held at arm's length, which means roughly ten words a slide, not forty. Eight to twelve slides is plenty. And write the caption as if the carousel did not exist: some people will only read that.
What carousels cost: real production time. Expect an hour minimum until you have a template you reuse. The payoff is that a good carousel keeps being useful — you can send it in a conversation six months later as an answer to a question.
What carousels are bad at: urgency and personality. They feel produced. Nobody argues with a carousel in the comments.
Video: trust, nuance, and the things text flattens
Video does one thing no other format does. It shows your face, your pace, your accent, the way you handle a hesitation. For anyone selling advice — consultants, coaches, fractional executives — that is not vanity. It is risk reduction for the buyer. They are about to put a person in a room with their team.
So use video when the intent is trust or nuance:
- Answering one real client question, straight to camera, in ninety seconds.
- Explaining a judgement call where the honest answer is "it depends, and here is on what".
- Showing a tool, a dashboard, a document on screen.
Three practical constraints. Most people watch without sound, so burn in captions or you lose them in the first second. Say the subject in the opening sentence — do not warm up. And keep it short enough that watching feels cheaper than scrolling past, which in the feed means under two minutes for most topics.
What video costs: nerve, mostly, and about three takes before you stop sounding like a hostage. The recording itself is fast. The editing is where people quietly give up, so choose a format you can ship unedited.
What video is bad at: being skimmed, and being referenced later. Nobody scrubs back through a video to find the third point.
A quick map of LinkedIn content formats
If you want it as one table in your head:
| Intent | Format | Why | |---|---|---| | Provoke thought, start a debate | Text | Lowest friction, invites reply | | Teach a repeatable method | Carousel | Sequence, one idea per slide | | Show a before/after | Carousel or image | Comparison is visual | | Build trust before a call | Video | Face, pace, nuance | | Share a client story | Text, or video if emotional | Story lives in voice | | Answer one recurring question | Video or short text | Specific, reusable in DMs | | Point to a long-form asset | Text, link handled carefully | See below |
That last line deserves a note. Sending people off-platform works against the way the feed is built, and the honest version of that trade-off is covered in what external links really do to your reach. Decide whether this post is for distribution or for conversion. It is rarely both.
Also worth saying: none of these formats are prospecting. A post reaches people who already follow you or who follow someone who engaged. Content warms the ground; outreach walks on it. If a post pulls three good comments, the next move is a conversation, not another post — and that conversation follows the same rules as any other, which is why outreach templates stopped working once everyone started using the same four.
Testing without changing everything at once
Most people never learn which format suits them because they change three variables a week. Do this instead.
Pick one format as your default for six weeks — the one you can sustain when a client deadline lands on a Thursday. For nearly everyone, that is text. Then add one of the other two, once a fortnight, for the same six weeks.
Track two things only: comments from people who look like buyers, and profile views in the following two days. Not impressions. Impressions tell you how widely something travelled, not whether it landed with anyone who could hire you. A carousel seen by eight hundred strangers is worth less than a text post read by forty people in your exact market.
After six weeks you will know something specific about your own audience that no general advice could have told you.
One honest caveat on tooling. If you use a prospecting workspace like BIZLINKED to run the outreach side, understand what it is: a Chrome extension that works in your own browser and LinkedIn session and automatically sends invitations and AI-written messages and follow-ups, inside limits the product sets itself. LinkedIn's User Agreement does not allow third-party software that automates activity, so using any such tool means accepting a risk of account restriction — at any volume, with any limits. That is a decision to make with your eyes open, and it has nothing to do with what you publish.
What to do Monday morning
Open your last ten posts. Beside each one write the intent in three words — "teach the audit", "argue against discounts", "show I'm human". If you cannot name the intent, that is why the post went nowhere.
Then take the two posts that got real replies from real buyers and ask whether the format was helping or fighting the intent. Rewrite one of them in the format that fits, and publish it this week. Same idea, right vehicle. You will usually find that the idea was never the problem.
