What Is Hook Rate? How to Measure It and What Good Looks Like
Jonathan TapieroJune 17, 20269 min read
Hook rate is the single most diagnostic number in a paid UGC account, and the one most teams either ignore or measure wrong. It tells you, in one figure, whether the first two seconds of your video stopped the scroll. Everything downstream (hold rate, click-through, cost per acquisition) is conditional on that first decision. If the hook fails, none of the rest of the ad ever gets a chance.
The trouble is that "hook rate" is defined three different ways across Meta, TikTok and the dashboards built on top of them, and the differences are large enough to make benchmarks meaningless if you mix them. This article gives you one consistent formula, explains how it relates to the metrics the platforms actually expose, what a realistic hook rate looks like by platform, and why chasing it blindly is a fast way to buy expensive, low-intent attention.
What hook rate actually measures
Hook rate measures how many people who were served your ad watched past the opening beat instead of scrolling. It is an attention metric, not an interest or intent metric. A strong hook rate means the first frame, the first spoken line and the first caption did their one job: they bought you the next few seconds.
The standard formula is simple:
Hook rate = 3-second video views / impressions
That is the version most performance marketers use, and it is the one this article benchmarks against. But the moment you compare numbers across tools, you hit a definitions problem, because the platforms do not all count a "view" the same way.
| Term | Numerator | Denominator | Where you see it |
|---|---|---|---|
| Hook rate (common) | 3-second video plays | Impressions | Meta Ads Manager, most reporting layers |
| 3-second view rate | 3-second video plays | Impressions or reach | TikTok, some dashboards |
| Thumb-stop rate | 3-second plays (sometimes 2s) | Impressions | Agency and tool dashboards |
| Hold rate | Plays to 25%, 50% or 75% | 3-second views or impressions | Both platforms, varies |
The takeaway is not which definition is correct. It is that you must pick one and hold it constant. A hook rate measured against impressions will always read lower than the same ad measured against reach, because one person can generate several impressions. Comparing your impression-based number to a competitor's reach-based number tells you nothing.
How to measure it correctly
Measuring hook rate well is less about the math and more about not fooling yourself with denominators and windows.
Pick impressions, not reach, as your denominator
Impressions count every time the ad was rendered; reach counts unique people. Because the auction can serve the same ad to the same person multiple times, reach-based hook rate flatters your number. Impressions is the stricter, more honest denominator, and it is what Meta uses by default for 3-second video plays. Use it consistently so your week-over-week trend means something.
Know what counts as a "play"
On Meta, a 3-second video play is the standard threshold. TikTok exposes a 2-second and a 6-second view as well, and its autoplay behavior in-feed inflates raw view counts compared to Meta. This is why a 45 percent "hook rate" on TikTok and a 30 percent hook rate on Meta can represent the same creative quality. Never compare a TikTok number to a Meta number as if they are the same metric.
Give it enough impressions before you trust it
Hook rate stabilizes fast relative to conversion metrics, because it fires on almost every impression rather than on rare events. A few thousand impressions per creative is usually enough to trust the order of finish. That is what makes it the ideal early read in a creative testing program for paid social: you can kill a dead hook in hours, long before CPA is meaningful.
Read it per creative, not per campaign
A campaign-level hook rate is an average that hides the spread. The whole point of the metric is to separate the openings that stop the scroll from the ones that do not, so always read it at the ad level and rank your creatives against each other.
What a good hook rate looks like
Here is where honesty matters. There is no universal "good" hook rate, because it moves with platform, placement, audience temperature and how you count. The ranges below are directional, drawn from common patterns in paid UGC reporting, not a guarantee for your account. Treat your own account average as the real benchmark and judge new creatives against it.
| Platform / placement | Weak | Decent | Strong | Note |
|---|---|---|---|---|
| Meta Reels / Stories | Under 20% | 25 to 35% | 40%+ | Impression-based 3s plays |
| Meta Feed | Under 15% | 20 to 30% | 35%+ | Slower scroll than Reels |
| TikTok in-feed | Under 25% | 30 to 45% | 50%+ | Autoplay inflates vs Meta |
A few things to hold in mind when you read those:
- Cold versus warm changes everything. A retargeting audience that already knows you will show a much higher hook rate than a broad cold prospecting audience. Benchmark cold against cold.
- Placement changes everything. Full-screen vertical placements like Reels and TikTok in-feed produce higher hook rates than feed placements, because the format itself demands more attention.
- Your trend beats any benchmark. A hook rate climbing from 22 to 31 percent across a testing cycle is a clearer signal of progress than hitting some external "good" number once.
If your hook rates sit consistently in the weak column across placements, the problem is almost never the targeting. It is the first two seconds, and specifically the patterns covered in TikTok ad hooks that convert: a missing pattern interrupt, no relevance signal, or an opening that screams "ad" with a logo or a polished studio intro.
Why a high hook rate is not the goal
This is the trap that costs accounts real money. Hook rate is an attention metric, and attention is cheap to buy if you do not care who you attract. A shocking visual, a clickbait claim or a misleading open loop can drive a huge hook rate while pulling in viewers who will never convert. You will have optimized for the wrong thing.
The fix is to read hook rate inside a ladder, never on its own:
- Hook rate (3s views / impressions): did the opening stop the scroll at all?
- Hold rate (plays to a meaningful point): did the body keep the people the hook attracted?
- Cost per acquisition or ROAS: did the watch turn into the action you care about?
A hook with a great hook rate but a collapsing hold rate is a curiosity trap: it earns the stop and then loses everyone because the body does not pay off the promise. A hook with a modest hook rate but a strong CPA is a quiet winner worth scaling. Always let the bottom-of-funnel number break the tie.
The diagnostic power of the ladder is that it tells you what to fix. Strong hook rate, weak hold rate means keep the opening and rebuild the body. Weak hook rate across the board means the body might be fine but nobody is getting there, so the openings are the problem.
Turning a weak hook rate into a fixed one
When the metric ladder points at the hook, the answer is volume, not agonizing over a single opening. The reason most teams plateau on hook rate is production economics: a traditional UGC shoot yields one or two usable openings, so they test two hooks, crown one by default and never find the angle that would have doubled the number.
The efficient move is to hold the body, voice and offer constant and vary only the first two seconds, then let hook rate rank the openings. When the opening is cheap to swap, testing fifteen hooks becomes routine instead of a luxury. This is exactly the workflow AI UGC pipelines enable: from one product photo and a short brief, the system generates a batch of finished 9:16 ads where each video opens on a different hook, with AI footage, AI voice, captions and music already assembled. Sepia is built around this many-hooks-from-one-product motion, so finding the opening that lifts your hook rate stops depending on a shoot calendar.
To be fair about the trade-off, a great human creator can deliver a hard-to-replicate authenticity that sometimes wins on hold rate even with a modest hook rate. AI generation wins on the breadth of openings you can test cheaply. Most disciplined teams use AI to find the winning angle, then decide whether to scale it as is or hand it to a creator.
FAQ
What is the formula for hook rate?
The most common formula is 3-second video views divided by impressions, expressed as a percentage. Some tools use reach instead of impressions as the denominator, which produces a higher number, and TikTok counts views differently from Meta. The math is less important than picking one definition and applying it consistently, otherwise your benchmarks and your week-over-week trend become meaningless.
Is hook rate the same as 3-second view rate?
They usually point at the same idea, but the denominator can differ. Hook rate is typically 3-second views over impressions, while some dashboards compute 3-second view rate over reach or even over total plays. Always confirm the denominator before comparing two numbers, because impression-based and reach-based rates are not interchangeable.
What is a good hook rate on Meta versus TikTok?
Directionally, strong Meta Reels hook rates sit around 40 percent or higher on an impression basis, while TikTok in-feed can read 50 percent or more because autoplay inflates view counts. These are not the same scale, so never compare them directly. The most reliable benchmark is your own account average, judged cold against cold and placement against placement.
Can a high hook rate still lose money?
Yes, and this is the most common mistake with the metric. Attention is cheap to buy with a shocking or misleading opening, so a high hook rate can attract viewers who never convert. Read it inside a ladder: hook rate, then hold rate, then cost per acquisition. A great hook rate with a bad CPA is a curiosity trap, and scaling it just buys cheaper attention from people who will not buy.
Hook rate is the fastest, most honest early signal you have, but it is a thermometer, not a destination. Use it to kill dead openings in hours and to rank the openings that survive, then let hold rate and CPA decide which of those survivors is actually worth your budget. The teams that win are not the ones chasing the highest hook rate; they are the ones who can read it in context and put the next ten openings in market before the current winner fades.