How to write hooks that stop the scroll

You can make a genuinely great video and still have almost nobody watch it. Usually the problem isn’t the content. It’s the first second.

People decide whether to keep watching almost instantly. On a feed, your hook is the whole pitch: the first line of a caption, the first frame of a video, the first sentence you say. If it doesn’t give someone a reason to stay, they’re gone before your good part even starts.

The good news: hooks are a skill, not a talent. Here’s how to write them on purpose.

What a hook actually has to do

A hook has one job: make the viewer want the next second. That’s it. Not explain everything, not be clever for its own sake.

A strong hook usually does at least one of these:

• Names a problem the viewer already has
• Promises a specific payoff they want
• Opens a loop they need closed
• Shows something visually unexpected
• Calls out who it’s for, so the right people feel seen

And it does it fast. On short-form video, think of the first one to two seconds. On a text post, the first line before “see more.” On a longer video, the opening seconds before someone decides whether to bail.

YouTube’s own audience retention report tracks what percentage of viewers are still watching after the first 30 seconds, and suggests changing your opening if that number is low. That tells you how much the platforms care about your start.

Hook in three layers

The strongest hooks hit three channels at once:

1. Visual: what’s on screen in frame one. Movement, a surprising object, your face mid-reaction, a before shot. Not a logo, not a slow pan, not you saying “hey guys.”
2. Text on screen: a short line that works with the sound off. Lots of people scroll muted.
3. Spoken or caption: the first words out of your mouth, or the first line of your caption.

They don’t have to say the same thing. The text can set up the topic while your first spoken line adds tension.

10 hook formulas (with examples)

Use these as starting points, not scripts. Swap in your niche.

1. The specific result “How I got my first brand deal with 2,000 followers.” Specific numbers and outcomes beat vague promises.

2. The mistake callout “Stop putting your watermark in the corner.” People hate the idea that they’re doing something wrong.

3. The “you” call-out “If you’re a creator who hates being on camera, watch this.” Tells the right person this is for them, and it’s fine if everyone else scrolls.

4. The contrarian take “Posting every day is killing your account.” Only use it if you can back it up. More on that below.

5. The open loop “I almost didn’t post this, but here’s what happened with my first client.” Creates a question the viewer wants answered.

6. The list promise “3 things I’d do if I were starting from zero today.” Sets clear expectations and signals the video won’t drag.

7. The before-and-after Show the “after” first, then: “Here’s how this started.” Works great for edits, makeovers, and setups.

8. The “nobody tells you” “Nobody tells you that your first 1099 is going to look like this.” Implies insider knowledge.

9. The question “Why do some creators get brand deals with half your followers?” Works best when the answer isn’t obvious.

10. The “POV” or scene-setter “POV: you just opened your first brand contract and have no idea what half of it means.” Drops the viewer straight into a moment they recognize.

Match the hook to the payoff

This is where most creators mess up. A hook is a promise. If the rest of the content doesn’t deliver on it, people feel tricked, and they leave or scroll faster next time they see you.

Before you post, ask:

• Does the content actually answer the question the hook raised?
• Does the payoff come reasonably soon, or am I making people wait through a long intro?
• If I promised “3 things,” are there actually 3 good things?
• Would someone who watched to the end feel it was worth it?

A quick fix for weak payoff: write the payoff first. Figure out the most useful or surprising thing in the piece, then write a hook that points straight at it.

For example, if the best part of your video is a simple email template that got you a reply from a brand, your hook shouldn’t be “Let’s talk about brand deals.” It should be closer to “The 3-sentence email that got a brand to reply in a day.”

How to test your hooks

Don’t guess. Treat hooks like experiments.

Write five, pick one. For every piece of content, write at least five hook options before you film. The first one is almost never the best one.

Test one variable at a time. Post similar content with different hooks and see which holds attention better. If you change the hook, the topic, and the length all at once, you won’t know what worked.

Look at retention, not just views. Views tell you how many people saw it. Retention (or average watch time) tells you whether your hook actually kept them. Most platforms show some version of this in their creator analytics.

Use built-in testing when you have it. YouTube’s Test & compare feature lets you test up to three title and thumbnail options, and picks a winner based on watch time. That’s a useful reminder: a click that doesn’t turn into watching isn’t a win.

Reuse winners. When a hook structure works, save it in a notes file with the stats. Over time you’ll have your own swipe file of hooks that work for your audience specifically.

And when a piece does well, you can take the same core idea and repurpose it into ten more posts, each with a fresh hook.

Common hook mistakes

• Starting with a greeting. “Hey guys, welcome back” costs you precious seconds.
• Starting with context. “So a few weeks ago I was thinking about…” Put the interesting part first and backfill the context.
• Being vague. “This changed everything” means nothing. Changed what? For who?
• Clickbait that doesn’t pay off. You might get the view, but you lose the trust.
• Hooks that only work with sound. Add on-screen text.
• Hiding the hook. If your first frame is a slow fade, a logo, or a wide shot of an empty room, the hook starts too late.
• Copying trending hooks word for word. Borrow the structure, not the exact line, and make it fit what you’re actually saying.

Your hook checklist

Before you hit post, run through this:

• Is the first frame visually interesting on its own?
• Is there on-screen text that works on mute?
• Do the first words give a reason to keep watching?
• Is it specific (a number, a result, a named problem)?
• Does the content deliver what the hook promised?
• Did I write at least five options and pick the strongest?
• Am I tracking retention so I learn from this one?

Hooks get easier the more you write them. Give yourself a month of writing five per post, and you’ll start hearing them in your head before you even pick up your phone.