Curtis HowlandFrameworksRequest a conversationFrameworks
Framework / Hook Rate

What is a good hook rate on Meta ads?

A good hook rate on Meta ads is 30 to 40%. Below 25%, kill the ad. Above 40%, scale it. Hook rate is 3-second video views divided by impressions.

Curtis Howland
Published 14 Aug 2026
Exhibit
Where each hook rate sits, and what to do next.
Hook rate thresholdsPlot your ads →
Kill below 25%. Good is 30 to 40%. Scale above 40%.
  • Below 25% · Kill
    The ad isn't stopping the scroll. Fix the first three seconds first.
  • 25–30% · Undecided
    Check hold rate, CTR and CPA before you decide.
  • 30–40% · Good
    The hook is working. If the ad still loses, look at the video body, offer or page.
  • Above 40% · Scale
    You won't see many. Cut new versions off the same opening.
Source: 20,000+ ads and $100M+ in Meta spend over six years. Cold, conversion-optimized video traffic.
01
Definition

What hook rate measures.

Hook rate is the share of people who watched at least 3 seconds of your video ad, out of everyone the impression reached.

It tells you one thing: did the first few seconds stop the scroll? The offer, product and landing page haven't come into it yet, which is why it's useful. When an ad dies, hook rate shows whether it died on the first frame or later. Those need different fixes.

Don't mix it up with hold rate. Hook rate is whether people stopped. Hold rate is whether they stayed. A strong hook with a weak hold means keep the opening and reshoot the rest.

Formula
3-second video plays ÷ impressions
02
Worked example

Two ads with the same CPA.

Ad A is in the good range, so leave the hook alone and judge it on what happens after three seconds. Ad B is below the kill line. It doesn't matter how good the offer is, because almost nobody sees it. It needs a new opening, and a new landing page won't help.

Their CPAs match today. But the 34% ad still has room to find its audience, and the 21% ad is wasting most of its impressions.

The numbers
Ad A: 34,000 plays on 100,000 impressions, or 34%. Ad B: 21,000 plays, or 21%.
03
Limits

When the thresholds don't hold.

There are six cases where these numbers don't apply.

  • 01Statics have no hook rateJudge images and carousels on CTR and downstream metrics.
  • 02Small samples swingWait until an ad has real delivery before you act on it.
  • 03Retargeting inflates itPeople who already know the brand stop for other reasons.
  • 04The objective changes the audienceEngagement campaigns read high. These numbers assume conversion delivery.
  • 05Placement shifts the baselineCompare within Feed or within Reels before comparing accounts.
  • 06A great hook with a bad CPAUsually means the ad promises something the product doesn't deliver.
04
In practice

What to do with it.

Hook rate tells you why an ad is dying. It won't tell you which ads to look at. For that, plot the account on a log-spend scatter, find the ads that are clearly losing, then use hook rate to see why.

Check it for every ad on a regular schedule, sort new launches by it once they have real delivery, and brief your next batch off every hook above 40%.

Hold these numbers to
Cold, conversion-optimized video traffic.
Run it on your account

Plot every ad on spend against CPA, with confidence bands.

Open Ad Scale Scatter

Free, and it runs in your browser. Nothing gets uploaded.

Get the weekly email.

One tactic a week, to 3,475 operators.