LinkedIn Hooks: What the Data Shows and How to Write Your Own

The first line of a LinkedIn post carries more weight than any other. In the feed, readers see only the opening before the "see more" cut, which is roughly 140 characters on a phone (details in our guide to writing a post). If those characters don't give them a reason to tap, the rest of the post doesn't get read.
That's why there are hundreds of hook swipe files online. This guide takes a different route: first what large studies of real posts say about opening lines, then why copied hooks work against you, then five ways to find a hook in your own material.
What the data says about opening lines
Two recent studies looked at first lines at scale. Both come from companies that sell LinkedIn tools and analyze their own data, so treat them as useful signals rather than settled science.
AuthoredUp classified 309,614 posts from personal profiles, published between December 2025 and May 2026, by the style of the first line (AuthoredUp):
| Opening style | Posts | Median engagement rate |
|---|---|---|
| Story | 22,745 | 2.60% |
| Contrarian | 19,603 | 2.31% |
| Statement | 171,391 | 2.27% |
| Results | 64,311 | 2.19% |
| Question | 31,564 | 2.16% |
MagicPost looked at about 1.18 million posts from the previous 12 months (MagicPost). Posts whose first line contained a question mark had a median of 19 likes, against 29 for the rest. First lines with a digit in them had a median of 35 likes, against 26 without. First-line length mattered less than you might expect: 1 to 5 words had a median of 30 likes, 16 words or more had 28.
Three things are worth taking from this, with caveats:
- Question openers do worse in both datasets, even though "start with a question" is one of the most common pieces of hook advice.
- Stories and numbers do better. The gaps are modest, and neither study reports significance tests.
- Length matters less than content. A short line isn't automatically a better one.
These are correlations, and the digit itself is unlikely to be the cause: posts that open with a number tend to be posts where the author has something concrete to report. The same goes for stories. The openers that perform well are the ones that show the reader there's real material underneath.
Why swipe-file hooks backfire
A template hook looks like a shortcut: take "Here's what nobody tells you about [topic]", fill in the topic, publish. Three problems come with it.
Readers have seen it. Popular templates spread fast, and a feed full of "Unpopular opinion:" and "I was today years old when..." trains people to skip them. Several of the most common templates also overlap with the patterns readers now associate with AI text, such as the teaser line and the fragment run ("Two years. No sleep. One lesson."). We list those patterns in How to make AI writing sound human.
AI tools copy them exactly. While tuning Elynd's prompts, we gave a model a sample opening sentence as an example. It reproduced that construction in 18 of 20 generated posts and only swapped the topic. If you paste a swipe file into ChatGPT, expect the same: your post opens the way thousands of others do.
The template makes a promise the post has to keep. "This one change tripled our revenue" works only if the post shows the change and the number. If the body is general advice, the reader feels tricked, and the next post from you gets less benefit of the doubt.
Five places to find a hook in your own material
The reliable way to write a first line is to look for it in what you already know about the episode, not in a list of templates. If you followed the steps for writing a post, you already wrote down what happened, what you did and what came of it. The hook is usually the most specific line in those answers, moved to the top.
Each example below pairs a generic opener with a specific one about the same topic. The examples are illustrations, not real posts.
1. The number
Weak: Most founders get pricing wrong.
Strong: We raised prices by 40% in March. Two of our 31 clients left.
A number tells the reader there's a real case behind the post. Use the actual figure, including the unflattering part.
2. The result
Weak: Here's how we doubled our pipeline.
Strong: We had 9 open deals in August and 21 in October. The only change was who ran the demos.
State the result and hint at the cause. The reader opens the post to learn the cause.
3. The decision
Weak: Sometimes saying no is the best decision.
Strong: Last week I told our biggest client we won't renew their contract.
A decision with consequences creates its own tension. You don't need to add "and it was the best thing I ever did"; the reader will want to know whether it was.
4. The quote
Weak: Feedback is a gift.
Strong: "Your demo answered questions nobody asked," a buyer told me on Tuesday.
Real words from a client, colleague or candidate are hard to fake and easy to picture. Quote only what was actually said, and ask before you name anyone.
5. The moment
Weak: Let me tell you a story about failure.
Strong: At 11 pm before launch, we found that the import deleted every second row.
Start inside the scene, with a time, a place or an object. This is the "story" opener from the AuthoredUp data, and it doesn't need an announcement that a story is coming.
Openers to retire
These show up constantly and mostly cost you attention:
- "Here's what nobody tells you about..." It promises a secret and says nothing yet.
- A question as the first line, unless it's the exact question a client asked you. Both studies above put question openers last.
- "Unpopular opinion:" followed by an opinion that isn't unpopular. If you disagree with common advice, open with what happened when you followed it.
- Three fragments in a row, like "No team. No funding. Just a laptop." It's one of the most recognizable AI patterns.
- "I'm excited to announce..." The announcement is fine; put the specific news in the first line instead of the excitement.
Elynd helps you write posts like this from your own material - try it free
A quick test for any first line
Before you publish, read only the first line, as if it were all you'll see in the feed, and check:
- Does it say what the post is about? A reader should be able to guess the topic.
- Does it contain something only you could write? A number, a name, a date, a quote, a decision.
- Does the post deliver what the line promises? If the hook mentions a result, the body explains how it happened.
- Would you say it out loud to a colleague? If it sounds like an ad, rewrite it the way you'd tell it.
If the line fails the second check, rewriting it rarely helps: the post doesn't have its specific detail yet, and no opening formula will add one. Go back to the episode and find it.
