What is hook rate?
Hook rate is the percentage of video ad impressions that turn into at least a 3-second view. It measures how well the first few seconds of a video, the hook, stop someone from scrolling past.
The standard formula is:
Hook rate = (3-second video views ÷ impressions) × 100
If an ad gets 20,000 impressions and 5,000 of them become 3-second views, the hook rate is 25%. The metric is most associated with Meta ads (Facebook and Instagram), where it is also called thumb-stop rate or thumb-stop ratio. As of 2026, Meta Ads Manager does not show hook rate as a default column, so advertisers usually build it as a custom metric from 3-second video plays and impressions.
How hook rate works
In a feed, people decide whether to keep watching almost instantly. Hook rate captures that decision. A view counts only once someone has watched for 3 seconds, so a low hook rate means most people saw the first frame and kept scrolling.
A few things shape the number:
- The first frame: a clear subject, motion, or a face tends to grab more attention than a logo or a slow fade in.
- The opening line: a question, a bold claim, or a direct call out of the audience ("If you run a small shop...").
- On-screen text: many people watch with the sound off, so a text hook in the first second carries the message.
- Format fit: a vertical 9:16 video fills the phone screen, while a horizontal video shows up small in a vertical feed.
- Audience targeting: the same creative can hook one audience and miss another.
Different platforms and tools define the threshold slightly differently. Some use 2 seconds for a thumb-stop metric, and TikTok reports its own 2-second and 6-second view counts. Compare hook rates only when they use the same definition.
Hook rate vs hold rate
Hook rate tells you whether people started watching. Hold rate tells you whether they kept watching. The two are usually read together.
| Hook rate | Hold rate | |
|---|---|---|
| Question it answers | Did the opening stop the scroll? | Did the rest of the video keep attention? |
| Common formula | 3-second views ÷ impressions | ThruPlays (or 15-second views) ÷ 3-second views, or ÷ impressions |
| Part of the video it judges | First 1 to 3 seconds | The body and payoff |
| Typical fix when low | New first frame, text hook, or opening line | Tighter pacing, earlier payoff, shorter cut |
Hold rate formulas vary more than hook rate formulas. Some teams divide ThruPlays by impressions, others divide by 3-second views, and some use 75% completion. Pick one and stay consistent.
A high hook rate with a low hold rate usually means the opening promised something the rest of the video did not deliver. A low hook rate with a high hold rate means the video is good, but few people get far enough to see it.
What counts as a good hook rate
There is no official benchmark. Industry write-ups often describe roughly 25% to 30% as a solid hook rate on Meta, with higher numbers considered strong, but results vary a lot by industry, placement, audience, and whether the ad targets cold or warm traffic. The most useful benchmark is your own account: compare new creatives against the hook rates of your past winners.
Where hook rate applies
- Paid social creative testing: comparing several versions of one ad that differ only in the first 3 seconds.
- UGC ads: choosing which creator opening or which line to lead with.
- Organic short-form: the same idea applies to Reels, Shorts, and TikTok, where the platform tracks how many viewers swipe away early.
- Creative briefs: writing several hook options per script so editors can test them.
Why hook rate matters
Every later metric depends on the hook. If people do not stop scrolling, they never see your product, your offer, or your call to action, so click-through rate and conversions are capped by how well the opening performs.
Hook rate is also one of the fastest signals you get. It stabilizes after a modest number of impressions, long before you have enough conversions to judge an ad. That makes it a practical way to cut weak creatives early and put budget behind the openings that work.
The practical takeaway is to test hooks, not just whole ads. Make several versions of the same video with different first frames and opening lines, then keep the ones that stop the scroll. An AI video tool makes it quicker to produce those variations, so you can test more openings with the same budget.