LinkedIn works for a software founder when the account reads as one coherent person with a subject, and stops working when it reads as an assortment. Format, posting time and hook formulas are real but marginal. What decides reach is whether the platform, and the reader, can tell what you are about.
Most LinkedIn advice is written for brand accounts trying to manufacture engagement. This is written for a founder with a product, limited hours, and buyers who are already on the platform.
Coherence beats cadence
The instinct when reach drops is to post more. For most founders that makes it worse.
LinkedIn's ranking has moved toward reading the whole account rather than the individual post. Your headline, your About section, your history and your last twenty posts form a pattern, and a new post is assessed partly on whether it fits that pattern. An account that posts about hiring, then pricing, then a conference photo, then a product update has no pattern to fit.
We wrote up what that looks like in practice in why posting five things made you invisible. The short version: narrow enough that a buyer could describe your subject in one sentence, broad enough that you can still write next month.
That single decision does more than every formatting tactic combined.
What you are actually competing for
Not attention. Attention is cheap and converts badly.
You are competing to be the person a buyer remembers when the problem you solve becomes urgent, which may be months later. That reframes what a good post is. A post that gets 400 likes from other founders and no comments from buyers is a worse post than one that gets 11 comments from six people in your category.
The practical test: after each post, look at who engaged, not how many. If your audience is mostly peers, investors and people selling to founders, you have built a following in the wrong market. That is not a failure of the algorithm. It is a failure of subject.
The three things worth posting
Roughly half of what you publish should teach something your buyer can use whether or not they ever hire you. Not thought leadership in the abstract. The specific thing you know because you have done it: how a migration actually goes wrong, what a pricing conversation sounds like, why the obvious approach fails at a particular scale.
Roughly a third should be your position on something contested in your category. This is the part founders skip because it risks disagreement, and it is the part that makes an account memorable. Neutral accounts get read and forgotten.
The rest should be proof. Not testimonials formatted as graphics, but specifics: what a customer was doing before, what changed, what it cost. One honest number beats five adjectives.
The proportions matter less than the principle, which is that an account made entirely of any one of the three fails. All teaching and you are a newsletter. All position and you are a commentator. All proof and you are an advertisement.
What the format debates are actually worth
Text posts, carousels, video, polls. Every study on this contradicts the last one, because they measure different accounts in different categories at different times.
The honest position: format affects reach at the margin and subject affects it structurally. A carousel about nothing in particular will not outperform a plain text post that says something specific. Pick the format that suits the idea. A process is easier to show than to describe. An argument is usually better as text.
Two things do reliably matter. Say the useful part early, because most people decide within a line or two whether to keep reading. And write for the person who will read it without expanding it.
On timing, test your own account for a month rather than adopting anyone's published window, including ours. The published windows describe an average account, and yours is not one.
Links, and the thing everyone does wrong
LinkedIn prefers you keep people on LinkedIn. That is not a conspiracy, it is a business model, and it means a post whose main content is a link out tends to travel less far.
The workaround most people use is to put the link in the first comment. It works, and it is also a tell that the post was built to distribute a link rather than to say something.
The better version is to make the post complete on its own. Say the actual idea, in full, in the post. If someone wants the longer version, they will find it or ask. A post that stands alone gets distributed on its merits and the interested minority still arrives at your site.
Comments are where the work is
Reach follows conversation, and conversation is mostly something you start elsewhere.
Commenting substantively on other people's posts, in your category, several times a week, does more for your visibility than an extra post of your own. It puts you in front of exactly the audience you want, it builds the relationships that later engage with you, and it costs ten minutes.
The rule is that the comment must be able to stand alone as a thought. "Great post" is invisible. Two sentences that add something, disagree usefully, or supply a specific example are not.
Reply to every comment on your own posts, properly. A thread that continues signals the post was worth continuing.
Publishing in public is not a distribution strategy
Worth saying plainly, because it gets conflated. Posting consistently builds trust with people who already know you exist. It does not, by itself, put you in front of people who do not. Those are different jobs, and we covered the distinction in build in public is not a distribution strategy.
LinkedIn is a trust asset that compounds slowly. Treat it as a lead source and you will judge it on the wrong timescale and quit at month three.
What to do for the next month
Rewrite your headline so it names who you help and with what, not your job title.
Choose one subject narrow enough to describe in a sentence. Write it down. Judge every post against it.
Post twice a week, on that subject, with the useful part in the first two lines.
Comment substantively on five posts a week, in your category, from your own account.
At the end of the month, look only at who engaged. If the names are your buyers, keep going. If they are not, your subject is wrong, and no posting frequency will fix that.