Hook Rate Benchmarks 2026: What a Good Hook Rate Is by Platform
Jonathan TapieroJune 16, 202613 min read
If you run paid social, hook rate is the first number you check on every new ad. It tells you whether your opening seconds stopped the scroll before anything else can happen. But "what is a good hook rate" is one of the most mis-answered questions in performance marketing, because the metric is custom, the thresholds come from third-party tools rather than the platforms, and the same percentage means different things on Reels versus Feed and on Meta versus TikTok. This is a reference roundup of the hook rate benchmark ranges that actually circulate among creative-analytics teams in 2026, what they mean, and where they break.
Every number here is attributed inline to a real source with the year it was published. We have also flagged, up front, that there is no Meta-official "good hook rate" threshold to cite. Meta's documentation defines the underlying 3-second view metric but does not publish a benchmark range, so every tier below comes from creative-analytics vendors and agencies. Treat these as a starting point, not gospel. If you want the broader scorecard that hook rate lives inside, UGC Ad Metrics That Matter is the companion piece.
The headline takeaway: a roughly 25 to 35% hook rate is a workable "solid" zone on Meta Feed, with 30 to 45% considered genuinely good and 45%+ exceptional, per AdManage.ai (2025). But hook rate measures attention, not value. Two ads can share an identical hook rate and produce very different ROAS, so always read it alongside hold rate and downstream CPA.
What hook rate actually measures
Before any benchmark means anything, you have to agree on the formula. Hook rate (often called thumbstop rate) is the share of impressions where the viewer watched past the platform's short-view threshold.
The standard calculation is 3-second video plays divided by impressions, times 100, according to Vaizle (2025). AdSights (2025) frames the same formula as (3-Second Video Plays / Impressions) x 100, and notes the duration threshold is 3 seconds for Meta but 2 or 6 seconds for TikTok. Motion (2025) uses an identical definition: 3-second video views divided by impressions, times 100.
A few definitional points decide whether two benchmarks are even comparable:
- What counts as a 3-second view. Meta counts a 3-second video view as a play of at least three seconds, or the full length (97%) if the video is shorter than three seconds, with replays excluded and views counted separately from each impression, per Metric Labs (2025).
- The denominator varies by tool. Hook rate is most commonly 3s views divided by impressions, but some practitioners and tools divide by reach instead, which yields a higher number for the same ad. Always note which denominator a benchmark uses before citing it.
- The platform threshold varies. Meta uses 3 seconds, while TikTok hook rate is typically calculated on 2-second views, per Billo (2025). A TikTok 2-second hook rate is not directly comparable to a Meta 3-second one.
This is why an impression is not the same as attention. Meta counts rapid scrolls and autoplay starts as impressions, so hook rate can be inflated by low-intent autoplay impressions, warm retargeting audiences, and sound-off placements. Segment before you judge a number.
Meta hook rate benchmarks
Here is where the tool-published ranges cluster. We have laid them side by side so you can see how much they overlap, and where they disagree.
| Source | "Needs work" | "Solid" | "Good / strong" | "Exceptional" |
|---|---|---|---|---|
| Motion (2025) | below 25% | 25 to 35% | above 35% | target baseline 30 to 40% |
| Vaizle (2025) | below ~20% | ~20 to 25% | above 30% (top ads) | n/a |
| AdManage.ai (2025) | below 20% | 20 to 25% | 30 to 45% | 45%+ |
| Billo (2025) | below 25% | n/a | 35 to 45% | n/a |
The consensus is tighter than it looks. Below roughly 20 to 25%, every source agrees you have a problem: AdManage.ai (2025) calls sub-20% the "fix-it zone" where the opener is basically invisible, and Billo (2025) treats below 25% as requiring a creative overhaul. The "solid but improvable" band sits around 25 to 35%, per Motion (2025). Genuinely good performance starts around 30 to 35% and runs into the mid-40s, and 45%+ is exceptional, according to AdManage.ai (2025).
What about the average, not the aspiration? Billo (2025) put the average Meta hook rate at 24%, with top performers pushing past 28% and optimized campaigns climbing to 30 to 50%. That lines up with Vaizle (2025), which gives a practical working range of roughly 20 to 25% across most accounts, with top-performing ads exceeding 30%. So when someone quotes you a glossy "good hook rate is 40%," remember that the realistic account-level average is closer to 24%, and most ads live in the 20s.
Hook rate by placement
A single account-wide number hides a lot. Reels, Feed, and Stories are different viewing modes, and the benchmarks reflect it. AdSights (2025) publishes placement-segmented Meta thumbstop benchmarks for DTC prospecting:
| Placement | Range | Median |
|---|---|---|
| In-feed video | 18 to 28% | ~23% |
| Reels | 24 to 36% | ~30% |
| Stories | 22 to 32% | ~27% |
Notice that Reels medians run higher than Feed in this dataset. That is the opposite of a common practitioner heuristic, which holds that Reels hook rates run several points lower than Feed because viewers are in faster-swipe mode (so a 25% on Reels can be quality-equivalent to a 30 to 35% in Feed). Both can be true depending on the account and the creative, and that is exactly the point: hook rate numbers are only comparable apples to apples, Reels-only versus Reels-only and Feed-only versus Feed-only. Never compare a blended hook rate against a single-placement benchmark. For more on how placements diverge, see TikTok vs Reels vs Shorts.
Hook rate by audience temperature
The other big confounder is who you are showing the ad to. AdSights (2025) reports warm retargeting thumbstop at 30 to 45% (median ~36%) versus cold prospecting at 18 to 28% (median ~22%). That is a roughly 14-point median gap driven purely by audience, not creative quality. A retargeting ad that "beats" a prospecting ad on hook rate has not necessarily earned anything. Always judge a creative against the benchmark for its own audience temperature.
TikTok hook rate benchmarks
TikTok plays by slightly different rules because its short-view threshold is shorter. Hook rate there is typically computed on 2-second views divided by impressions, per Billo (2025). Billo cites an 11-account analysis where the average TikTok hook rate was 30.7%, with top-quartile creatives reaching 40 to 45%.
That higher average partly reflects the shorter 2-second window. Clearing 2 seconds is easier than clearing 3, so TikTok numbers naturally read a few points hotter than Meta numbers for comparable attention. Here is how the two platforms tier against each other, per Billo (2025):
| Tier | TikTok | Meta |
|---|---|---|
| Elite | 40%+ | 35 to 45% |
| Competitive | 30 to 39% | 30 to 39% |
| Fix-it zone | below 25% | below 25% |
The practical implication: do not hold your TikTok creative to a Meta number, or vice versa. A 32% on TikTok and a 32% on Meta are not the same achievement. If you are building hooks specifically for the platform, TikTok Ad Hooks That Work and UGC Ad Hook Examples cover the creative patterns behind those numbers.
Why the opening seconds matter at all
Hook rate is only worth optimizing if early attention actually drives outcomes, and the strongest public evidence for that comes from TikTok's own research. TikTok Marketing Science (2024, via Social Media Today) found that 50% of the impact from a TikTok ad is realized in the first 2 seconds, and the first 6 seconds capture 90% of cumulative impact on Ad Recall and roughly 80% for Awareness. That is the mechanistic case for caring about the hook: most of the lift happens before most viewers have decided whether to keep watching.
On the buying side, Motion (2025) notes that Meta rewards content that quickly engages by serving it more efficiently, which translates to lower CPMs and cheaper cost per action, making hook rate a leading indicator of downstream CPA. Note the honest framing here: that relationship is asserted qualitatively (cheaper delivery for engaging creative), not as a published numeric hook-rate-to-CPA coefficient. No Meta Marketing Science study we could find directly quantifies a hook-rate-to-ROAS link, so treat the CPA connection as directional.
Hook rate is not the win condition
This is the most important section, and the one most teams skip. Hook rate measures attention, not value. Practitioner Barry Hott explicitly warns that a higher CTR or hook rate (thumbstop) does not mean an ad will perform better, cautioning against treating it as a vanity metric (Barry Hott on LinkedIn, 2023). Two ads can share an identical ~30% hook rate and produce wildly different ROAS, which is why you read hook rate alongside hold rate and profit metrics (CPA, ROAS, CAC) rather than scaling on 3-second views alone.
Hold rate is the natural partner metric. It is computed as ThruPlays divided by 3-second video plays, per Vaizle (2025), and answers the second question: once you stopped the scroll, did you keep them? Motion (2025) benchmarks hold rate at 40 to 50% as average, over 60% as strong, and under 30% as needing improvement. Vaizle (2025) gives a similar practical range of roughly 40 to 50%, with top performers exceeding 50%.
| Metric | Formula | "Good" range | Source |
|---|---|---|---|
| Hook rate (Meta) | 3s views / impressions x 100 | 30 to 45% | AdManage.ai (2025) |
| Hold rate | ThruPlays / 3s plays | over 60% strong, 40 to 50% avg | Motion (2025) |
A high hook with a collapsing hold means a great opener attached to a weak body. A modest hook with a strong hold and good CPA is a better ad than a flashy hook that never converts. The competitive-benchmarking category is maturing to support exactly this kind of multi-metric read: Varos (2024, via Digital Journal) launched the first benchmarking dashboard dedicated to video ads, covering hook rate, percent viewed over 50%, average watch time, ThruPlays, and CTR across its data co-op.
The benchmark behind the benchmarks
For scale context on where these Meta numbers come from, Motion's 2026 Creative Benchmarks report analyzed an anonymized dataset of more than 550,000 ads launched by over 6,000 advertisers, representing roughly USD 1.3 billion in spend across Facebook and Instagram between September 2025 and early January 2026. That is a large, recent sample, which is why Motion's thresholds carry weight. But it is still one vendor's account base, with its own client mix and category skew.
Which brings us to the single most useful benchmark of all: your own. Because hook rate is a custom, advertiser-computed metric with no Meta-official threshold, every public range should be treated as a starting point, and every number is account-relative. The benchmark that actually predicts your next winner is your own account's historical median for the same audience, the same category, and the same creative style. Public tiers tell you roughly where you stand. Your own median tells you whether this specific ad beat the last one. The volume of testing that produces a reliable internal median is the real project, and How Many Ad Creatives to Test covers how much you need.
What this means for your testing
Pulled together, the picture is coherent. On Meta, aim for a hook rate in the 30s, treat sub-20 to 25% as a fix, and read 45%+ as exceptional (AdManage.ai, 2025). On TikTok, the bar runs a touch higher because of the 2-second window, with ~30% as the average and 40%+ as elite (Billo, 2025). Always segment by placement and audience before judging a number, never compare a 2-second TikTok rate to a 3-second Meta rate, and never scale on hook rate alone, since attention is not value (Barry Hott, 2023).
The through-line is that hook rate rewards volume. You do not engineer a 35% hook on the first try, you find it by testing many openers and keeping the ones that stop the scroll. That is a production problem before it is an analytics problem. For the system that turns hook data into a pipeline, see The Creative Testing Framework for Paid Social.
SepiaLab exists to make that volume affordable. With it, you generate on-brand AI UGC video at the cadence paid social demands, so you can test enough hooks to find the ones that clear these benchmarks instead of betting everything on one opener. Get started and generate enough hooks to test against your own hook-rate history.
FAQ
What is a good hook rate for Facebook ads?
On Meta, a hook rate of 30 to 45% is considered genuinely good and 45%+ is exceptional, per AdManage.ai (2025), while 25 to 35% is "solid" with room to improve per Motion (2025). The realistic account-level average is lower, around 24%, according to Billo (2025). There is no Meta-official threshold, so these come from third-party creative-analytics tools and should be treated as a starting point.
How is hook rate calculated?
Hook rate is 3-second video plays divided by impressions, times 100, per Vaizle (2025) and Motion (2025). Meta uses a 3-second threshold, while TikTok hook rate is typically computed on 2-second views, per Billo (2025). Note that some tools divide by reach instead of impressions, which produces a higher number, so always check the denominator before comparing.
What is a good hook rate on TikTok?
TikTok hook rates run a few points higher than Meta because the threshold is 2 seconds rather than 3. Billo cites an 11-account analysis where the average TikTok hook rate was 30.7%, with top-quartile creatives reaching 40 to 45%, per Billo (2025). Billo treats 40%+ as elite on TikTok and below 25% as the fix-it zone. Do not hold TikTok creative to a Meta 3-second benchmark.
Is hook rate or hold rate more important?
They answer different questions and should be read together. Hook rate measures whether you stopped the scroll, while hold rate (ThruPlays divided by 3-second plays) measures whether you kept attention, with 40 to 50% as average and over 60% as strong per Motion (2025). A high hook with a low hold is a strong opener attached to a weak body. Neither is the win condition, since both can be high while ROAS is poor (Barry Hott, 2023).
Why does my hook rate vary so much across placements?
Because each placement is a different viewing mode and the benchmarks reflect it. AdSights (2025) reports DTC prospecting medians of ~23% for in-feed video, ~30% for Reels, and ~27% for Stories, and warm retargeting can run 14 points higher than cold prospecting. A blended account number hides all of this, so segment by placement and audience temperature before judging any single creative.