LinkedIn Post Ideas, Each With an Opening Line

A LinkedIn post idea is worth nothing until you know the line that opens it and the page it sends people to.

That is the whole job, and it is smaller than the listicles make it look. What follows is twelve ideas for a founder with no marketing hire, each one with the opening line typed out, the format it should ship in, and the number that decides whether it earned a second go.

Lists of LinkedIn post ideas are lists of subjects, and a subject is the easy half. "Share a mistake you made" is a subject. What you type underneath it is the post, and by then you have lost the chance to decide what the post is for.

One figure sets the scale of everything else. The median LinkedIn post is seen about 496 times, and the median engagement rate on a post under 2,800 impressions is 2.86% (AuthoredUp, September 2026 refresh, 894,168 posts from July 2025 to June 2026).

496 is a school hall, not a stadium. So the ideas worth writing are the ones that use the hall for something you can count by Friday.

What a LinkedIn Post Idea Has to Survive Before You Write It

Three questions, one line each, answered in the doc before the draft starts.

1. Who is it for? One person, named. The owner of a 12-person agency, not "founders".
2. What does it make them do? Name the action and the page it lands on. If the honest answer is nothing, the idea is a mood, and moods do not compound.
3. Which number says it worked, and by when? Impressions against your own median, clicks on the destination, or replies inside 24 hours. Pick one before you write.

Ideas that fail the second question are why a founder posts for six weeks and books nothing. The post reads well. People nod. Nobody clicks, because there is nowhere to click.

Write your own baseline before you judge anything. Open your last ten posts, list the impressions, and take the middle number. The platform-wide median is 496 and the typical post's engagement rate is 2.86% (AuthoredUp), with profiles at 2.36% and pages at 2.59% in the same dataset. Yours is the number that matters, because yours is the one to beat.

Then hold the week to three jobs: one post that beats your median by half, one that sends clicks to a page, one that gets replies. None of the three is "stay visible".

That is the filter. Everything below is material that passes it.

Twelve LinkedIn Post Ideas, Each With the Opening Line

The table is the sheet to work from. Every opening line is written to be typed and not admired, so swap the specifics for yours and the line stays honest.

Pick three, not twelve. One document, one text post with a link in it, one that asks for nothing at all. That covers the reader who wants a framework and the reader who wants a verdict, and it gives the reader who has just met you something worth reading.

The post with no link is not the wasted one. It is what makes the other two credible. A feed where every post ends in a pitch is a feed people stop opening.

Two rows earn a slot first for a brand with a page to send people to, and the launch version of both sits in launch ideas priced from zero.

Swap one line and every entry above becomes yours. "A customer asked why her email landed in spam" is a sentence from a real week. The same idea written as "Email deliverability tips for small business" is an article title, and article titles do not get read on a feed.

Three of the Twelve That Earn a Slot First

The quote you got wrong

This line works because it costs you something to post. A price you got wrong, in dollars, buys attention that a tip cannot, and it does it without a single claim about how good you are at anything.

Write it in four sentences: the quote, the real cost, what you missed when you scoped the job, and the line you added to your quoting sheet since. Put the number in the first sentence, because the number is the whole reason anybody stops scrolling.

What it sends is replies, and the replies are the destination. Twenty comments under a post like this is a week of sales conversations you did not have to book, and two of them will be people who have the same problem you just described.

The comparison nobody publishes

Four options, four prices, one verdict. This is the most linkable post a small brand can write, because everyone in your category faces the same four options and almost nobody puts the table in public.

Use the prices on each vendor's own pricing page and date them in the post, the way a reader can check. Give the price, the minimum commitment, and the one thing that vendor is genuinely better at. Then say which one you chose and which one you stopped paying for.

What it sends is the comparison page on your site, which is where the same reader lands next week, when the decision is real. A dated source at every figure is what keeps the post out of trouble and what makes people save it as a reference.

The customer's words

A quote from a customer does two jobs at once. It is the only social proof a small brand has, and it is the one line a competitor cannot copy from you.

Get it by asking one specific question in your review email: what did we fix that someone else billed you for? Post the answer the way it was said, clumsy parts included, and get permission in writing before you do.

What it sends is the reviews page, though the number worth watching here is replies rather than clicks. This is a post people save, and a save outlives a like by months.

Match the Idea to the Format Before You Write a Word

Format is a distribution decision, and the three biggest studies of LinkedIn performance disagree about how much it is worth. All three agree on the order.

The first column is Socialinsider's 2026 LinkedIn benchmarks, built on 2,450,610 posts from pages between 1,000 and 1,000,000 followers. The second is Buffer's 2026 state of engagement, 52 million posts across ten platforms, reported as medians. Metricool's study of 577,180 posts puts carousels at 45.85% (Metricool).

Three numbers for one format: 45.85%, 21.77%, 7.00%. That spread is not a typo, it is a definition problem. An average taken over impressions, a median taken over a mixed sample, and a percentage published without a denominator will never agree with each other.

Which one do you trust? The one you can reproduce. Take your own median, then rank your own formats against each other, because the ordering is the part every dataset gets right: documents and carousels on top, plain link posts at the bottom, in all three.

A good idea written as a link post therefore starts the race in last place, at 3.25% in one dataset and 3.81% in another. Here is how to carry a link without paying that price.

1. Write the whole argument as text, with no link inside the post.
2. Name the destination in the last line, and say where the reader finds it.
3. Put the link in the first comment you add yourself, then compare the post's impressions against your median.
4. If your own numbers say the version with the link in the post did better, keep the link in the post. Your data outranks theirs.

The carousel deserves its reputation, and it also deserves a build that takes twenty minutes instead of a day.

1. Slide one carries the claim in six words.
2. Slide two carries the number that proves it.
3. Slide three shows the screen, cropped to the one thing that changed.
4. Slide four names the mistake you made on the way.
5. Slide five is the fix, as three steps.
6. Slide six is the line you want repeated, with your name in the corner.

The failure mode is a document with thirty words on every slide. That is a PDF, and a PDF in a feed is a link post wearing a costume. Twelve words to a slide, and the argument belongs in the post's own text.

A Five-Day Cadence That Fits Around the Work

Thirty minutes a day, five days, three posts out. The days matter far less than the order.

Wednesday at 4 p.m. is the single strongest slot of the week in Buffer's analysis of 4.8 million posts, and the whole 3 p.m. to 8 p.m. window now beats the mornings (Buffer, July 2026). That same page notes LinkedIn's own marketing blog points to studies where Tuesday is among the most active days, so the two disagree on the day. Your last ten posts outrank both, and if Wednesday is quiet for your audience, move the good post to whenever your own report says otherwise.

Three posts a week is where cadence starts to pay for itself. Buffer's analysis of more than two million LinkedIn posts found that two to five posts a week was worth about 1,182 more impressions per post than posting once, and that the gain held at any audience size (Buffer, July 2026).

Put the block in the calendar tonight: thirty minutes at 3:30 p.m. on Wednesday, repeating, named "draft and post". Then write the opening line before you write anything else. If the first line does not make you want to read the second, the idea is not ready, and it will not be ready next week either.

The rest of the social media work a small business has to do happens on the other four days. This is the thirty minutes that feeds it.

The Seven-Day Scorecard, With Real Numbers

Here is a post from a real week, scored. It collected 1,180 impressions, 31 reactions, 11 comments, and 4 clicks on the destination page.

Impressions first. At 1,180 against a 496 median, that post reached 2.4 times the typical one, so the format earned a repeat.

Engagement rate next. Thirty-one plus eleven, divided by 1,180, is 3.6%, which clears the 2.86% median for a typical post (AuthoredUp). The subject earned a sequel.

Clicks last. Four clicks out of 1,180 impressions is 0.34%, and the comparison that matters is against your own last five posts rather than against a published benchmark. If this was the weakest click rate of the six, the close is the problem, not the idea.

There is one trap in that arithmetic. Engagement rate falls as reach rises, because the denominator grows faster than the reactions do. Posts above 15,600 impressions carry a median engagement rate of 0.89%, while posts under 2,800 impressions sit at 2.86% in the same dataset (AuthoredUp). A post with the lower rate and forty times the reach is the better post, and judging both by rate alone will have you repeating the smaller one forever.

The winner gets a sequel inside two weeks: same subject, new opening line, different format. That is how one good idea becomes four posts without four ideas. The losers go back in the doc with a note about the opening line rather than the subject, because the subject was rarely what failed.

The rest of the social posts on this blog collect under Social, and the cadence here feeds all of them.

Where the Post Sends People Decides the Idea

Write the destination first. Then choose the idea that walks somebody to it.

One page can serve every post in a week. Tag the link so those clicks arrive as their own source in your analytics instead of disappearing into a general social bucket, which is what a UTM code is for. Read the report on day seven, because what counts as social media traffic in GA4 is sessions from that tag, and it is the only number that shows the post did something a like cannot.

Then check what the visitor does after landing. A good bounce rate on that page means the promise in the post and the promise on the page agree with each other, which is the actual reason a founder bothers with any of this.

There is a second kind of destination, one that keeps working after the post stops being seen. Every row on this blog's own board is a paid placement, ranked strictly by what has been paid and nothing else, and 25 brands are on it right now. The link in a row is the product, and the rest is packaging.

Pick one destination page this week and point all three posts at it. Two will underperform your median. The third pays for the set, and the referral report on day seven is where you will see which one did it.

Three posts a week, one destination, one number written next to the impressions. That is the whole operation, and the setup takes an afternoon.

The ideas are the easy half. Any of the twelve above produces a post that reads well, and most of them produce a post that reads well and does nothing for the business, because the last line asked for a like instead of a click.

The order is the line, the format, the destination, then the number. Four decisions, made before you type a word of the post, and the writing is the part you already know how to do.