Hook rate and hold rate
Two ratios that split a video ad's performance into its opening and its body. Hook rate shows whether the first seconds stop the scroll. Hold rate shows whether the rest keeps people watching.
Some teams divide ThruPlays by impressions instead. We divide by 3-second plays so hold rate measures the body on its own.
What are hook rate and hold rate?
CTR tells you a video ad isn't working. It doesn't tell you where it breaks. A video can lose people in the first second, or grab them and then lose them halfway through. Those need completely different fixes.
Hook rate measures the opening: of everyone who was shown the ad, how many watched at least 3 seconds. Hold rate measures the body: of those who were hooked, how many stayed through to a ThruPlay.
How they're calculated
Both come from three video metrics Meta already reports for each ad, over the same date range.
| Meta metric | What it counts | Used for |
|---|---|---|
| impressions | Times the ad was shown | Hook rate denominator |
| 3-second video plays | Plays of at least 3 seconds | Hook rate numerator, hold rate denominator |
| ThruPlays | Plays to completion, or at least 15 seconds for longer videos | Hold rate numerator |
Comparing creatives fairly
Hook and hold rates move with things that have nothing to do with how good the creative is. Compare like with like.
- 1Match video lengthA ThruPlay on a 10-second video means a full watch. On a 45-second video it means 15 seconds. Group creatives by length.
- 2Match placement mixReels, Stories and Feed can behave differently. Compare creatives that ran on similar placements.
- 3Watch the trendHook rate tends to slip as an ad ages and repeat viewers scroll past. A falling hook on a winner is an early fatigue signal.
Worked examples
Example 1: Diagnosing three creatives
Three illustrative 20-second videos for the same product, each with 2,00,000 impressions on similar placements.
| Creative | Impressions | 3-second plays | Hook rate | ThruPlays | Hold rate | ThruPlay rate |
|---|---|---|---|---|---|---|
| A · Founder story | 2,00,000 | 60,000 | 30% | 9,000 | 15% | 4.5% |
| B · Product demo | 2,00,000 | 24,000 | 12% | 8,400 | 35% | 4.2% |
| C · UGC unboxing | 2,00,000 | 56,000 | 28% | 19,600 | 35% | 9.8% |
- AThe opening works, the body loses people. Tighten the middle and bring the proof or offer forward.
- BThe people who stay love it, but few stop. Keep the body and test three new first-3-second openings.
- CBoth halves work. This is the one to scale, and its hook is worth reusing on B.
Example 2: Same ThruPlay rate, opposite problems
On ThruPlay rate alone, A and B look almost identical. Split into hook and hold, they need opposite edits.
Using them to decide
As starting points, treat a hook rate around 25% and a hold rate around 30% as strong for short-form video, then calibrate against your own best creatives. Always confirm with CTR and true ROAS before scaling.
Common mistakes
- Mixing definitionsHold rate over impressions and over 3-second plays aren't comparable. Pick one and label it.
- Comparing different lengthsShort videos get easier ThruPlays. Compare within a length band.
- Scaling on hook aloneA clickbait opening can win the hook and lose the sale. Check true ROAS.
- Using it on static adsThese are video metrics. For images, look at CTR and New Reach.
Frequently asked questions
What is a good hook rate?
It depends on placement, category and audience. Around 25% is a reasonable starting bar for short-form video. Your own top creatives are the best benchmark.
What counts as a ThruPlay?
Meta counts a ThruPlay when a video is played to completion, or for at least 15 seconds if it's longer than that.
Should I optimise campaigns for ThruPlays?
Only for awareness goals. For sales, optimise for purchases and use hook and hold rate to improve the creative itself.
Related metrics
Know which half of every video is failing
BooleanMaths lines up hook rate, hold rate and true ROAS for every video creative, so you know whether to re-cut the opening or the body.
Hook rate and hold rate
Two ratios that split a video ad's performance into its opening and its body. Hook rate shows whether the first seconds stop the scroll. Hold rate shows whether the rest keeps people watching.
Some teams divide ThruPlays by impressions instead. We divide by 3-second plays so hold rate measures the body on its own.
What are hook rate and hold rate?
CTR tells you a video ad isn't working. It doesn't tell you where it breaks. A video can lose people in the first second, or grab them and then lose them halfway through. Those need completely different fixes.
Hook rate measures the opening: of everyone who was shown the ad, how many watched at least 3 seconds. Hold rate measures the body: of those who were hooked, how many stayed through to a ThruPlay.
How they're calculated
Both come from three video metrics Meta already reports for each ad, over the same date range.
| Meta metric | What it counts | Used for |
|---|---|---|
| impressions | Times the ad was shown | Hook rate denominator |
| 3-second video plays | Plays of at least 3 seconds | Hook rate numerator, hold rate denominator |
| ThruPlays | Plays to completion, or at least 15 seconds for longer videos | Hold rate numerator |
Comparing creatives fairly
Hook and hold rates move with things that have nothing to do with how good the creative is. Compare like with like.
- 1Match video lengthA ThruPlay on a 10-second video means a full watch. On a 45-second video it means 15 seconds. Group creatives by length.
- 2Match placement mixReels, Stories and Feed can behave differently. Compare creatives that ran on similar placements.
- 3Watch the trendHook rate tends to slip as an ad ages and repeat viewers scroll past. A falling hook on a winner is an early fatigue signal.
Worked examples
Example 1: Diagnosing three creatives
Three illustrative 20-second videos for the same product, each with 2,00,000 impressions on similar placements.
| Creative | Impressions | 3-second plays | Hook rate | ThruPlays | Hold rate | ThruPlay rate |
|---|---|---|---|---|---|---|
| A · Founder story | 2,00,000 | 60,000 | 30% | 9,000 | 15% | 4.5% |
| B · Product demo | 2,00,000 | 24,000 | 12% | 8,400 | 35% | 4.2% |
| C · UGC unboxing | 2,00,000 | 56,000 | 28% | 19,600 | 35% | 9.8% |
- AThe opening works, the body loses people. Tighten the middle and bring the proof or offer forward.
- BThe people who stay love it, but few stop. Keep the body and test three new first-3-second openings.
- CBoth halves work. This is the one to scale, and its hook is worth reusing on B.
Example 2: Same ThruPlay rate, opposite problems
On ThruPlay rate alone, A and B look almost identical. Split into hook and hold, they need opposite edits.
Using them to decide
As starting points, treat a hook rate around 25% and a hold rate around 30% as strong for short-form video, then calibrate against your own best creatives. Always confirm with CTR and true ROAS before scaling.
Common mistakes
- Mixing definitionsHold rate over impressions and over 3-second plays aren't comparable. Pick one and label it.
- Comparing different lengthsShort videos get easier ThruPlays. Compare within a length band.
- Scaling on hook aloneA clickbait opening can win the hook and lose the sale. Check true ROAS.
- Using it on static adsThese are video metrics. For images, look at CTR and New Reach.
Frequently asked questions
What is a good hook rate?
It depends on placement, category and audience. Around 25% is a reasonable starting bar for short-form video. Your own top creatives are the best benchmark.
What counts as a ThruPlay?
Meta counts a ThruPlay when a video is played to completion, or for at least 15 seconds if it's longer than that.
Should I optimise campaigns for ThruPlays?
Only for awareness goals. For sales, optimise for purchases and use hook and hold rate to improve the creative itself.
Related metrics
Know which half of every video is failing
BooleanMaths lines up hook rate, hold rate and true ROAS for every video creative, so you know whether to re-cut the opening or the body.