# LinkedIn Hooks: First Lines That Stop the Scroll

Source: https://www.flypost.io/learn/linkedin-hooks
Author: Sam Cassman, Founder, Flypost
Last verified: 2026-10-12

What makes a good LinkedIn hook? Writing in the first person. Across 2,400+ LinkedIn creators, using "I", "my", "we" or "our" in the first line earned about 19% more reactions than the same person's other openers, and it worked for about 7 in 10 of them. Contrarian and question openers did nothing or made things worse, and a number in the first line helped only a little.

Are you trying to work out what to put in the first line of your LinkedIn post? You are not alone,
and the advice is unusually loud on this one. Open with a contrarian take. Lead with a number. Ask a
question. Keep it short and punchy. Almost none of it arrives with any evidence attached, and the
people repeating it rarely show their working, so we decided to test the formulas properly.

Most people believe a good hook is a trick you can learn, and that the right opening line will do
the heavy lifting. When we sorted the first lines written by 2,400+ LinkedIn creators into five
styles, and compared each person against their own other posts, we found something different. Four
of the five made no reliable difference or actively cost engagement. The only opener that earned more
was the one with no formula in it: writing about yourself.

In this article we'll cover what a hook actually is and how much of it people see, which of the five
opening styles earned more and which lost, why questions let you down at both ends of a post, and
what to write in your own first line instead. Read on to find out more.

_Change in reactions, compared with the same creator's other posts_

**Worth doing**

- Use "I", "my" or "we" in the first line: +19% (Worked for about 7 in 10 of the 1,734 people we tested)

**Don't bother**

- "Unpopular opinion…" openers: No change
- Making a number the first word: No reliable change
- Keeping the first line short: No pattern
- Writing at the reader ("you need to…"): −10%
- Opening with a question: −17%

## What is a LinkedIn hook?

Your hook is the first line of your post: everything before your first line break. It matters more
than the rest of what you write, because LinkedIn hides almost all of it. Only about **210
characters on desktop and 140 on mobile** show before the "…see more" link, and everything past
that point is something the reader has to choose to open.

That fold is the tightest of [LinkedIn's character limits](https://www.flypost.io/learn/linkedin-character-limits) that
actually changes how you write.

In our July 2026 data the typical first line ran to **75 characters**, with most between 49 and 129. So the
average hook uses around a third of the space it gets before the fold.

That much is uncontroversial. The interesting question is what you should put in it.

## Which LinkedIn hooks actually work?

We sorted first lines into five styles and compared each person's posts using a style against their
own posts that did not. One style earned more. Two made no reliable difference. Two did worse
than the writer's ordinary openers.

**How much each opening style helped or hurt a person's typical post** (Reactions, compared with the same creator's other posts)

| Tactic | Reactions vs their own posts | Change in reactions | Creators tested | Who benefited |
| --- | --- | --- | --- | --- |
| First person (I, my, we, our) | 1.19× | +19% | 1,734 | 69% |
| A number anywhere in the first line | 1.08× | +8% | 1,772 | 61% |
| Starting with a number | 1.05× | +5% | 288 | 57% |
| Contrarian ("unpopular…") | 0.98× | -2% | 398 | 47% |
| Direct address ("you") | 0.90× | -10% | 1,382 | 35% |
| Question hook | 0.83× | -17% | 685 | 31% |

_Source: Flypost S5 · 2,400+ B2B creators, Oct 2025–Sep 2026, each person compared with themselves_

That shape is the finding. If hook formulas worked the way they are taught, this chart would be a
staircase of clever techniques beating plain writing. Instead the only one that wins is the one
that asks you to have done something worth describing.

### Do first-person LinkedIn hooks perform better?

Yes, and it is the most reliable thing we know about first lines.

Posts whose first line used **I, my, we or our** earned about **19% more reactions** than the same
person's other openers. We measured that across **1,734 people**, and about **7 in 10** of them did
better. If your posts usually get 30 reactions, that is the difference between 30 and about 36.

![First-person openers earned about a fifth more reactions than the same writer's other opening lines](https://tblwcyfbemfosujjgqca.supabase.co/storage/v1/object/public/flypost-assets/content-studio/cf02f919-25d1-416b-8541-1c92410e174d/1790354323803-first-person-openers-drive-20-more-linkedin-reactions-v1.png)

_From our first run in July 2026 (20%, 1,790 creators). The October 2026 re-run found 19% across 1,734._

It is one of only three clear wins in our whole
[LinkedIn algorithm study](https://www.flypost.io/learn/linkedin-algorithm-study-data), alongside attaching an image and
[telling a personal story](https://www.flypost.io/learn/linkedin-algorithm-study-data#do-personal-stories-work-on-linkedin).
Posts where the writer described something they actually did, saw or went through earned about
**a third more** reactions than the same person's other posts, and that worked for 3 in 4 people.

> **About 19% more**
>
> What putting the first person in your opening line did, against the same writer's other hooks.
> About **7 in 10** people did better, and it held on text, image and video posts alike.
>
> _Flypost original research · S5, each person compared with themselves · n=1,734 creators_

We checked this one harder than anything else on the page when we first ran it in July 2026,
because our first assumption was that it had to be something else in disguise.

- **It is not really about images.** First-person posts carried an image 56% of the time, and so did
  everything else. No difference.
- **It is not really about length.** They ran to about 1,060 characters against about 1,040 for
  other posts, and [post length only moves engagement a little anyway](https://www.flypost.io/learn/linkedin-post-length).
- **It is not one format carrying it.** Split three ways it held in each: about 22% more on
  text posts, 18% on image posts and 13% on video posts, with roughly 6 to 7 people in every 10
  benefiting each time.

So what is left is the writing itself. **LinkedIn readers reward a post that sounds like a person
describing something that happened to them**, and they decide that from the first line.

**Instead of:** Most founders get their first sales hire completely wrong.

**Try:** I hired the wrong head of sales twice before I understood what I was getting wrong.

_Contrarian claim → something you did · Illustrative example, written by us. Not a real post from the study._

**Instead of:** Here are 5 lessons from 10 years in B2B marketing.

**Try:** We spent 10 years selling B2B the hard way. Five things I would tell myself at the start.

_Listicle framing → first person · Illustrative example, written by us. Not a real post from the study._

Same subject, same information, same length. The second version of each simply reports something
the writer did. So before you reach for a formula, check whether your first line could just say
what happened to you.

## Do question hooks work on LinkedIn?

No. Opening with a question was the worst style we tested: about **17% fewer reactions** than the
same person's other openers, across **685 people**, and only about **3 in 10** of them did better.
If a question opener is your habit, this is the single easiest thing on this page to change.

Three separate studies, using three different methods, land on the same answer.
[AuthoredUp's analysis of 309,614 posts](https://authoredup.com/blog/linkedin-hook-examples) ranked
question openers last of the five styles it tested, at 2.16% median engagement against 2.60% for
story openers. [MagicPost, looking at 1.2 million posts](https://magicpost.in/blog/linkedin-hooks),
found first lines that ask a question earn 19 likes against 29 for everything else.

It also fails at the other end of the post. We separately found that **ending** with a question
earned a typical gain of only **about half a comment**, while costing about **7% of reactions**,
across 924 people. Questions let you down at both ends, which is awkward for the most repeated tip
on LinkedIn.

The same goes for the other bait people put at the bottom of a post. "Comment GUIDE and I'll send
it to you" used to work, but since July 2026 those posts have earned about
[24% fewer reactions](https://www.flypost.io/learn/linkedin-algorithm-study-data#does-comment-keyword-to-get-it-still-work-on-linkedin)
than the same person's other posts. If you are offering something, share the link instead.

> Opening with a question cost about 17% of reactions. Only 3 in 10 of the people who tried it came out ahead.
>
> _Flypost study · 685 creators, each compared with their own other posts_

The reason is not mysterious. A question asks the reader to do work before you have given them
anything. A first-person opener does the opposite: it hands over a fact, and lets them decide
whether they want the rest.

## Do "unpopular opinion" hooks work on LinkedIn?

They made **no real difference**. Contrarian openers built on _stop_, _never_, _nobody_, _wrong_
or _unpopular_ came out within about 2% of the same person's other hooks, across **398 people**.
47% did better and 53% did worse, which is close to a coin flip. So if you are writing "Unpopular opinion:" out of habit,
you can stop without losing anything.

The claims made for it are not modest.
[Postiz](https://postiz.com/blog/linkedin-post-engagement-tips) says the "Unpopular opinion:"
opener can triple engagement. Across the people in our sample who actually use it, it moved their
typical post by nothing.

To be fair to the technique, the largest outside dataset disagrees with us here.
[AuthoredUp](https://authoredup.com/blog/linkedin-hook-examples) ranked contrarian openers second
best of the five styles it tested, at 2.31% median engagement. Theirs compares posts by different
people, so it partly measures who writes contrarian hooks in the first place; ours compares each
person against themselves. We would rather show you the disagreement than hide it.

> Worth being precise about what this does and does not say. We did not find that contrarian posts
> fail. We found that **putting a contrarian phrase in the first line** does not change your typical
> result. A genuinely surprising argument may well do well. Announcing one in the opener is not what
> makes it work.

## Should you start a LinkedIn post with a statistic?

It depends what you mean by "start with". **A number somewhere in your first line helps a little**:
about **8% more reactions** than the same person's other first lines, across **1,772 people**, and
6 in 10 of them did better. We saw the same small bump in both our July and October runs. But
**making the number the very first word** made no reliable difference, across 288 people.

So the practical version: if you have a specific figure, get it into the first line, where it
does its job of being concrete. Don't twist the sentence just to lead with the digit.

[MagicPost's analysis of 1.2 million posts](https://magicpost.in/blog/linkedin-hooks) found first
lines containing a number earn 35 likes against 26 for those without one. We agree on the
direction. The size is the difference: theirs compares posts by different people, so part of
that gap measures who writes number-led hooks. Compared within the same person, the bump is real
but small. (Our first version of this page, in July, only tested numbers as the opening word and
reported no effect. Testing a number anywhere in the line is the fairer match for their claim.)

## Should you write "you" in a LinkedIn hook?

Openers built on _you_ and _your_, the "You need to stop doing this" shape, cost about **10%**
across **1,382 people**. Only about **1 in 3** did better, the second-worst hit rate on this page.

It is the mirror image of the first-person finding, and together they are the clearest signal in
this data: **write about yourself, not at the reader.** The good news is that the same sentence
usually turns around at no cost.

**Instead of:** You're probably measuring your LinkedIn engagement wrong.

**Try:** I was measuring my LinkedIn engagement wrong for about two years.

_At the reader → about yourself · Illustrative example, written by us. Not a real post from the study._

## How long should a LinkedIn hook be?

There is no length that wins, so this is one rule you can stop worrying about. Compared with first
lines of 40 to 80 characters:

| First line length       | What happened           |
| ----------------------- | ----------------------- |
| Under 40 characters     | About 5% more reactions |
| 40 to 80 characters     | Our comparison point    |
| 80 to 140 characters    | No difference           |
| 140 characters and over | About 6% more reactions |

Short openers and long openers both come out marginally ahead of the middle, which is not a
pattern, it is noise. **"Keep your first line short" has no support in our data**, and neither does
its opposite.

That matches what we found in the body of the post, where
[chasing an ideal post length](https://www.flypost.io/learn/linkedin-post-length) turned out to be mostly measuring
images and account size rather than length. Write the first line the length your sentence needs.

## How do you write a better LinkedIn hook?

In order of how much our data supports it.

1. **Start with something you did.** "I", "my", "we". It is the only style that beat a person's own
   baseline, and it worked for about 7 in 10 of them.
2. **Delete the question mark.** Not at the start, not at the end. Ask when you genuinely want an
   answer, not as an engagement tactic.
3. **Stop turning the opener around on the reader.** "You need to" costs about 10%. The same point
   made in the first person does not.
4. **Drop the formula.** Contrarian phrasing came out at nothing, and stat-led openers at no
   reliable effect. That time is better spent on the thing you are actually reporting.
5. **Write the first line to the length the sentence needs.** No length band beat any other.
6. **Then attach an image.** It is worth about **a third more reactions**, more than any hook, and
   the two work independently of each other. The detail is in the
   [full algorithm study](https://www.flypost.io/learn/linkedin-algorithm-study-data#do-linkedin-posts-with-images-get-more-engagement).

> **Draft hooks from your own material**
>
> Flypost writes from what your business already has (posts, transcripts, case studies) rather than
> generating filler, which is the most reliable way to end up with a first line that is genuinely
> about something you did.
>
> [Try it free](https://app.flypost.io)

## How we tested LinkedIn hooks

The same method as the rest of our research, and the same limits.

**Who we looked at.** 2,400+ English-speaking B2B creators between 1,000 and 500,000 followers,
individuals only, spread across region and audience size. Their posts were collected in two
sweeps, in July and October 2026, and every test on this page was re-run on the latest twelve
months, October 2025 to September 2026. Reshares excluded. Nothing counted until a
post was at least 14 days old. We use typical figures (medians) rather than averages throughout, so
one viral post cannot drag a result upward.

**How we defined the hook.** The first line means the text before the first line break. Posts with
no line break at all were excluded, as were image-only posts with no text, so this analysis runs on
a subset of the 115,000+ posts with text in the study rather than all of them. We checked the
extraction before classifying anything: in the July 2026 data the typical hook ran to 75
characters, with most between 49 and 129.

**How we tested.** Every number on this page compares a person against themselves, with at least
four posts on each side of the comparison. People who tied exactly are left out of the "who did
better" percentages rather than counted as losses. Sample sizes differ by style, because not
everyone writes every kind of hook, and they are stated with each finding.

**What this page cannot tell you.** Three things, honestly.

> **We measured reactions, not scroll-stopping.** Nobody outside LinkedIn can see impressions, so we
> cannot tell you which first lines made people stop. We can only tell you which ones sat on posts
> that ended up with more reactions. The title of this page is how people search for the topic, not a
> claim about what we measured.
>
> **The styles are sorted by wording, not by quality.** "First person" means the first line contains
> I, my, we or our. It does not mean the writing is good. Styles we cannot spot from the wording
> alone, such as a curiosity gap or a story opener, are not in this hook test, so their absence here
> is not evidence against them. (Whether the whole post told a personal story we did test,
> separately, and it did well.)
>
> **Starting with a number rests on 288 people**, the smallest sample on this page. A number
> anywhere in the first line rests on 1,772 and gives a small bump, in the same direction as a much
> larger study that used a different method.

This page is one finding from our
[LinkedIn algorithm study of 116,000+ posts](https://www.flypost.io/learn/linkedin-algorithm-study-data), which tested
the most repeated LinkedIn rules the same way. See also
[how long a LinkedIn post should be](https://www.flypost.io/learn/linkedin-post-length) and
[every LinkedIn character limit](https://www.flypost.io/learn/linkedin-character-limits).

Spot a flaw? We would genuinely like to know. Tell us and we will correct it here.

## Frequently asked questions

### What is a LinkedIn hook?

The first line of your post, meaning everything before your first line break. It carries the whole post, because only about 210 characters on desktop and 140 on mobile show before the '…see more' fold. In our July 2026 data the typical first line ran to 75 characters.

### What makes a good LinkedIn hook?

Writing in the first person. Posts that used 'I', 'my', 'we' or 'our' in the first line earned about 19% more reactions than the same person's other openers, across 1,734 people, and about 7 in 10 of them did better. It worked on text, image and video posts alike.

### Do question hooks work on LinkedIn?

No. Opening with a question was the worst style we tested: about 17% fewer reactions than the same person's other openers, across 685 people, and only about 3 in 10 did better. Ending with a question does not help much either.

### Do 'unpopular opinion' hooks work on LinkedIn?

They made no real difference. Contrarian openers built on 'stop', 'never', 'nobody', 'wrong' or 'unpopular' came out within about 2% of the same person's other hooks, across 398 people, with 47% doing better and 53% doing worse. That is close to a coin flip.

### Should you start a LinkedIn post with a statistic?

Putting a number somewhere in your first line helps a little: about 8% more reactions than the same person's other first lines, across 1,772 people. Starting the post with the number itself made no reliable difference (288 people). So include a specific figure early, but don't contort the opening to lead with it.

### How long should a LinkedIn hook be?

There is no length that wins. Compared with first lines of 40 to 80 characters, shorter ones earned about 5% more and longer ones about 6% more, which is noise rather than a pattern. The 'keep your first line short' rule has no support in our data.

### Do LinkedIn hook formulas work?

Four of the five styles we tested did nothing reliable or cost reactions. The only one that worked was writing in the first person, which is a habit rather than a formula: it means the post is about something you actually did.

---

Published by Flypost (https://www.flypost.io), the LinkedIn content studio. More guides: https://www.flypost.io/learn