How to write hooks that stop the scroll without clickbait

By Ranjit Bhinge, founder of Blur Studio · Updated · 7 min read

Short answer: A hook is the first line (or first 2 seconds of a video) that makes someone stop and read on. A good one promises something the post then delivers. Write three versions, post each on a few separate days with the same body, and keep the one that holds attention best. A tool such as Blotato, a software subscription that writes, designs and schedules social media posts, can draft hook options and schedule the test, but only your own numbers tell you which hook works for your audience.

Ranjit Bhinge

You write decent posts that nobody opens, and you suspect the first line is the problem, but you do not want to sound like a click farm.

I'm Ranjit, and I run Blur Studio: 15+ years building software, 120 apps taken live, and clients who have gone on to raise $10M+. I build AI systems and processes that help businesses do a lot more with a lot less time. I also work differently from most of what else is out there, more on that further down this page. Say hello.

Most advice on hooks is a list of tricks. A list cannot tell you which trick works for your audience, and some tricks slide into clickbait that costs you trust. This guide does three things: it explains what a hook does, gives patterns with clearly made-up examples, and then gives you a test table so you learn from your own posts. For related openers see our hooks library of 40 post openers. Where a tool helps I name it, including Blotato, and where it does not fit I tell you. Product limits are as of November 2026. TikTok is a short video app, and Claude is an AI chat assistant from Anthropic.

This page may contain affiliate links from which Blur Studio earns commissions.

What a hook does

On LinkedIn (a business social network) and Instagram (a photo and video app), most of a post is hidden until someone taps "see more" or swipes. The hook is the only part that is always visible. Its job is narrow: earn the next 10 seconds of attention. It does not need to explain the whole post. It needs to do three things:

  1. Be about the reader. A reader asks "is this for me" before anything else.
  2. Open a gap. It raises a question the rest of the post answers.
  3. Be true. The post must pay off the promise. A hook that the post does not pay off is clickbait by definition.

Where a hook ends and clickbait starts

Use this test: cover the hook, read the post, then read the hook again. Would a reader who finished the post say the hook was fair? If the hook promised a secret and the post gives a tip everyone knows, that is clickbait. If the hook promised a number and the post has no number, that is clickbait. If it only works by hiding the subject ("You will never guess what this plumber did"), it fails. A fair hook can be curious and still deliver on every word.

Seven patterns, with made-up examples

The examples below are invented to show the shape. None of them has been tested and none is a claim about results. Swap in your own facts.

PatternMade-up exampleHonesty check
Number"We fixed 3 things in our quote email. Replies doubled." (invented)The number must be real and you must be able to show it
Mistake"The most common mistake in a kitchen quote is leaving out the waste bin."The mistake must be common in your experience, not a guess
Question"Why does your invoice get paid last?"The post must answer it
Before and after"Last year: 14 emails to book a visit. Now: 2."Only use real figures from your own work
Contrarian"I stopped offering free quotes. Here is what happened."It must be what actually happened
Specific story"A customer rang me at 7pm on a Sunday about a leak."It must have happened; change names if needed
Plain promise"A 5 step checklist for your first week of social posts."Count the steps. There must be 5

For a video, the hook is the first 2 to 3 seconds: a spoken line plus on-screen words. Put the key claim in both. Some viewers watch with the sound off, so on-screen text helps carry the hook.

Honesty rules

  • No invented results. Do not write "I made $10,000 in a week" unless it is true and you can show it. Advertising regulators such as the US Federal Trade Commission (FTC) and the UK Advertising Standards Authority (ASA) take misleading claims seriously, so check their current guidance before you publish a result as a hook. This is information, not legal advice. If a post is paid or gifted, also read the FTC's page on disclosing paid relationships, linked in the sources.
  • No fake urgency. "Only 2 spots left" must be true.
  • No hidden subject. Say what the post is about within the first line.
  • Pay off in the first third. If the answer arrives at the end, tell the reader in the middle that it is coming.
  • Do not copy someone else's hook word for word. A format is fine, the exact words are theirs. See how to use viral posts as templates without copying.

Common hook mistakes

  • Starting with your name or your business. Readers do not care yet. Start with their problem.
  • Burying the subject. If a reader cannot tell the topic from the first line, they scroll on.
  • Using a long first line. On LinkedIn the preview cuts off after a few lines, so a hook that needs 40 words loses its payoff. Aim for one line you can read in a breath.
  • Using the same pattern every day. Readers learn to ignore it. Rotate patterns.
  • Using all capitals or stacked emojis. It reads as shouting and as spam.
  • Promising a result you cannot deliver in the post. "How I got 10,000 followers" followed by a vague list is a broken promise.

Test design: a hook test table, not a list of tricks

Social data is noisy, so keep the test small and honest about it. This plan needs only 6 posts over three to four weeks, all of them posts you would publish anyway. Six posts cannot prove anything, so treat the result as a hint and repeat it.

  1. Pick one platform and one post type. For example LinkedIn text posts. Mixing formats ruins the comparison.
  2. Pick 3 hook patterns from the table, for example Number, Mistake and Question.
  3. Write 2 topics you would post anyway.
  4. Write each topic with each hook. That is 6 versions. Keep the body the same and change only the first line. Space the three versions of a topic at least a week apart and change the order between topics, so no pattern always goes first.
  5. Post at the same time of day and record the numbers at the same age for every post (say 48 hours after posting), using the platform's own screen.

At one or two posts a week the test runs for three to four weeks. If you cannot spare 6 posts, run 3 (one topic, three hooks) and treat it as a rough guide only.

Copy-ready test log

Post #TopicPatternDate and timeImpressions at 48 hours"See more" or swipe rate if shownRepliesSaves or sharesNotes
1
Continue to 6. Then compare the three patterns.

Pass and fail rule

A pattern passes the round if it comes first in impressions or replies on both topics. If no pattern does that, the result is a tie, and the lesson is that your topic matters more than the pattern. Six posts is a very small sample, so run the winner against a new pattern in the next round rather than treating it as final. If any post gets a complaint that it was misleading, that pattern fails regardless of its numbers.

Where tools help, and where they do not

  • Drafting options. Any AI assistant can write 10 hook options from your body text in a minute. Ask for a different pattern each time, then edit them in your own words.
  • Blotato. Blotato says its Viral AI Coach scores a draft video you upload (up to 100 MB) on five points, including the opening hook, and can offer 10 hook options if the score is low. Blotato's pages say the trial shows only a limited view of the scorecard. It analyses your video and does not make a new one, and Blotato's claims about what it was trained on are vendor claims. Blotato also publishes free Claude skills, including one called viral-hooks, which are written instructions you load into Claude, not a connection to Blotato. See does the Viral AI Coach help.
  • Measuring. Blotato's analytics cover posts published through Blotato only. On Starter they update on day 1, 7, 14 and 30 after posting, so a 48 hour reading is not available there. The platform's own screens give 48 hour numbers. For this test, use those.

When not to use Blotato for this

You do not need a paid tool to run the test. Skip Blotato for this job if you already schedule posts somewhere else, post under five times a week, or only want hook ideas. A free chatbot and a spreadsheet do the whole job. Blotato fits if you want drafting, scheduling and a video hook score in one place and post to several platforms. Its Starter plan is $29 a month, billed monthly, as of November 2026.

How this page was checked

The Blotato facts come from its help pages on the Viral AI Coach, analytics and Claude skills, and its pricing page. The FTC page covers disclosing paid relationships; the points on misleading claims point you to the regulators to check and are not a legal summary. Last checked 21 November 2026, refreshed every 90 days. Blur Studio has not run Blotato itself, and none of the hook examples here is a tested result, so run the test on your own posts and trial any tool before paying.

How I'd approach this

I would test hooks before buying anything. Week 1: write 2 topics in 3 versions each, and publish the first two. Weeks 2 to 4: publish the rest and fill in the log after each, then compare the three patterns. Then repeat with the winner against a new pattern. If drafting 6 versions feels slow, a trial of Blotato is one way to speed up the drafting step, as long as you edit every hook yourself. For a chat about your own posts, use the free AI brainstorm call or the free tools at /free.

Questions people ask

What makes a good social media hook?

It is about the reader, opens a question and is true. The post must pay off what the first line promises. If it does not, it is clickbait.

What is the difference between a hook and clickbait?

A fair hook can be curious but the post delivers every word of it. Clickbait hides the subject, overpromises or never pays off. Cover the hook, read the post, then read the hook again to test it.

How many posts do I need to test hooks?

Six posts is a workable start: 3 hook patterns on 2 topics, all posts you would publish anyway. It is a very small sample, so repeat the winner against a new pattern instead of treating it as final.

Do LinkedIn and Instagram need different hooks?

Usually yes. LinkedIn text posts show the first lines before "see more", so the first line matters most. Instagram carousels and Reels depend on the first slide or first 2 to 3 seconds. Test each platform on its own.

Can AI write my hooks?

It can draft options fast. Edit each one so it is true to your post and your own voice. See how to keep AI posts from sounding the same.

Does Blotato test hooks for me?

No. Blotato says its Viral AI Coach scores one uploaded draft video, including the opening hook. A hook test across posts you run yourself, using your own numbers.

Is it fine to use a viral post's hook?

Using the same format is usually fine. Copying the exact words is not. See the guide on templates and copying.

More in this topic: Social media: AI images, video, captions and carousels (17 guides) →

Sources

Ready for the next step?

A free 15-minute AI brainstorm. We find 3 broad areas where AI could help your business.

Ranjit Bhinge You talk to me, Ranjit. I reply within a few hours.
Book a free AI brainstorm call

Not sure what you need? Tell me what is going on. I will point you the right way, even if it is not me.

Blur Studio · 30 minute call

Book a 30 minute call

You talk to me, Ranjit. If you do not need me, I will tell you.

  • We talk about your business and what is going on.
  • You leave with a clear next step, even if it is not me.
  • I reply to every message within a few hours.
Loading the calendar…