Hook rate is the first number that tells you whether the opening two seconds earned a look. This hub defines the maths, the 30%+ target, how Meta and TikTok differ, what actually fixes a weak thumbstop, and when a high hook rate is lying to you.
01
Definition & Maths
1.What is hook rate?
Hook rate is the share of impressions that earn a short view — on Meta, typically 3-second video views divided by impressions. It is a thumbstop metric, not a sales metric. A viewer who watches three seconds has not bought, clicked, or even agreed with you; they have only not scrolled. That is still the gate: if the first two seconds fail, nothing downstream in the script can save the ad. Read hook rate next to hold rate and outbound CTR, never alone. A 35% hook with a 6% hold means people looked and left; a 22% hook with a 16% hold and a 1%+ CTR can still be a winner on cold traffic.
#Definition & Maths
2.How do you calculate hook rate on Meta?
Use 3-second video plays ÷ impressions, and keep the breakdown at the same attribution and placement window you use for the rest of the test. Do not mix thruPlay into that fraction; thruPlay is a different, longer-view metric and will make the hook look worse than it is. Pull the number at ad level, not campaign level, or a strong creative will hide a dead one. Exclude audience network if it distorts the mix, and compare Reels to Feed separately when the gap is large. The formula is simple; the error is almost always mixing placements or mixing warm and cold traffic in one ad set.
#Definition & Maths
3.What is a thumbstop, and is it the same as hook rate?
A thumbstop is the behaviour; hook rate is the measurement of that behaviour. In a 9:16 feed the viewer is moving fast, usually on mute, and the first one to two seconds decide whether the motion of the thumb pauses. Hook rate counts how often that pause happened, using a platform's short-view definition. They are not identical because a 3-second view can include autoplay that the viewer barely registered. Treat thumbstop as the creative job — pattern interrupt, face, on-screen line — and hook rate as the scoreboard. If the scoreboard is green and the rest of the funnel is red, the pause was curiosity, not relevance.
#Definition & Maths
02
Benchmarks
4.What hook rate should I aim for?
Target 30%+ on cold traffic. Below about 20% the opening is weak and you should fix the first two seconds before you touch the body script, the avatar, or the offer. Those two lines are the working benchmarks we use across accounts; they are not a published platform average and they will move with placement mix and how warm the audience is. Retargeting will read higher because people already know you. Prospecting is the number that matters. Do not celebrate 40% hook rate if hold rate is outside a healthy 12–20% (15-second views ÷ 3-second views) and outbound CTR is under 1% on cold traffic.
#Benchmarks
5.Is a 25% hook rate a failure?
Not automatically. 25% sits in the grey band: above the weak <20% line, below the 30%+ target. Read the rest of the funnel before you kill it. If hold rate is a healthy 12–20% and outbound CTR is 1%+ with an acceptable CPA, keep the ad and iterate the opening rather than discarding the whole thing. If hold and CTR are also soft, the 25% is a warning that the first two seconds are not doing enough work. Grey-band ads are where most waste happens — teams either scale them too early or kill a salvageable body script. New hook, same body, is the right next test.
#Benchmarks
6.Why does retargeting hook rate look so much better than prospecting?
Because the viewer already knows the brand, the product, or the face. Recognition is a thumbstop. A 45% hook on a 1% conversion audience does not mean the creative would do that on cold traffic, and using it as your benchmark will make every prospecting ad look like a failure. Split reporting: cold versus view/engagement/purchaser retargeting, always. Judge new creative on the cold number. Use the retargeting lift as a consistency check — if the same avatar keeps winning on warm traffic, that is a casting signal you can reuse, not proof that the hook is globally strong.
#Benchmarks#False Signals
7.What hold rate should sit next to hook rate?
A healthy hold rate is 12–20%, measured as 15-second views ÷ 3-second views. Below that, people who stopped are not staying; above that on short ads can simply mean the video is 15 seconds long and everyone who hooked completed it. Hold is the honesty check on hook. A 30%+ hook with 8% hold is a false thumbstop. A 22% hook with 18% hold and 1%+ cold CTR is often a quieter winner. Read the pair every time you open Ads Manager. If you only have time for two creative diagnostics, make them hook and hold; CTR and CPA come next.
#Benchmarks#Definition & Maths
03
Platform Differences
8.How is hook rate different on TikTok versus Meta?
The job is the same — stop the scroll in the first two seconds — but the meters are not. Meta's common hook proxy is 3-second views ÷ impressions. TikTok's short-view and 6-second / 15-second completion metrics do not map 1:1 onto that fraction, and For You distribution behaves more like entertainment than a right-hand-column ad. Native TikTok pacing (jump cuts, on-screen text, a spoken hook that sounds like a comment) usually beats a Meta UGC pasted over. Do not copy a 30% Meta hook-rate target onto a TikTok dashboard and call the ad dead. Compare TikTok ads to other TikTok ads in the same account, same optimisation, same time window.
#Platform Differences
9.Does Reels hook rate compare cleanly to Feed?
No. Reels is full-screen, sound-optional, swipe-vertical, and closer to TikTok than to a static Feed unit. A creative can hook at 34% in Reels and look ordinary in Feed because the unit itself supplies motion and face size. Always break hook rate out by placement before you declare a winner. If 80% of spend is Reels, the campaign-level 32% hook rate is a Reels number wearing a campaign badge. Build 9:16 first. If you need Feed, export a 1:1 or 4:5 cut rather than letterboxing, and judge that cut on Feed stats, not on the Reels halo.
#Platform Differences
10.Should I use thruPlay as my hook metric?
No. ThruPlay is a 15-second (or completion, if shorter) view metric. It is closer to hold than to hook. Using thruPlay ÷ impressions as a “hook rate” will make almost every ad look weak and will push you to rewrite the middle of the video when the actual problem is the opening, or the reverse. Keep two ratios: hook as 3-second views ÷ impressions, hold as 15-second views ÷ 3-second views. Target hook 30%+, hold 12–20%. ThruPlay is useful as a delivery and quality signal; it is the wrong denominator partner for thumbstop.
#Platform Differences#Definition & Maths
04
Diagnosing Low Hook Rate
11.What does a low hook rate actually mean?
The first two seconds did not earn the look. That is usually one of five things: a slow or silent open, a face that does not match the buyer, a line no human would say, a setting that reads as an ad, or an audience so mismatched that nothing would hook. It is not automatically a bad product, a bad offer, or a bad landing page — those show up later, in CTR, CPC and CPA. If hook is under ~20% on cold traffic, do not open the landing page, do not rebuild the offer, and do not add five new avatars. Change the opening shot and the opening line, then re-read at 48–72 hours or about 1,000 impressions.
#Diagnosing Low Hook Rate
12.How do I tell a hook problem from an offer problem?
Look at the rest of the funnel. Weak hook (under ~20%) with decent hold and CTR on the people who stay is a hook problem. Strong hook (30%+) with weak hold (well under 12%) is a mismatch between the opening promise and the next ten seconds. Strong hook and hold with weak CTR is a creative-to-offer or creative-to-page problem. Strong click and terrible CPA is usually offer, landing page, or tracking. Fix in that order: hook, body, click, offer. Teams waste weeks rebuilding a product page when the 9:16 open is a logo fade and a smile. Read the ratios before you touch the store.
#Diagnosing Low Hook Rate
13.My hook rate is low only on one avatar — what is that?
A casting signal. If three hooks all die on avatar A and live on avatar B, the face, age band or setting is wrong for that audience, not the copy. That is exactly why the starting batch is 6 creatives — 3 hooks × 2 avatars on one body script. Two avatars is enough to separate casting from copy without slicing budget into noise. Do not average them. Kill the weak face, keep the winning face, and add hooks against it. Preview taste is a terrible casting tool; the cold hook rate is the tool.
#Diagnosing Low Hook Rate
14.Can frequency make hook rate fall even if the hook is good?
Yes. On cold audiences, creative fatigue typically shows at frequency 2.5–3.5: hook rate and CTR slip together, CPA rises, and the same opening that used to stop the thumb becomes wallpaper. That is not a new diagnosis of a bad hook; it is the audience having seen it. Confirm by splitting frequency buckets if you can, and by checking whether new people still hook. The cheap refresh is a new first line on the same body and avatar. Rebuilding the whole ad because frequency cooked a winner is the most common waste we see in creative teams.
#Diagnosing Low Hook Rate
15.Does a weak hook mean I should hire a human creator instead?
Only after you have tested openings, not before. Most “AI UGC does not hook” reports we see are slogan scripts, slow opens, or a face that does not match the buyer — all of which also fail with a hired creator. Run 3 hooks × 2 avatars, read 48–72 hours or ~1,000 impressions, and look at hook against the 30%+ / <20% bands. If several decent avatars still cannot clear ~20% on cold with specific, spoken copy, you may have an offer or category-fit problem, and a human shoot will not magically fix that. Film real only when the failing ingredient is something AI cannot do: hands-on demo, fit, or a founder the audience already trusts.
#Diagnosing Low Hook Rate
05
Fixes
16.What actually fixes a weak hook rate?
A new first two seconds. Start talking or showing the problem in the first sentence, put a face and on-screen text on frame one, cut any logo sting, and write a line a real person would say out loud. Then generate 3 hooks × 2 avatars on the same body script and read hook rate at 48–72 hours or ~1,000 impressions per variant. The fixes that rarely work: a new music bed, a new colour grade, a longer intro “for context”, or swapping the CTA while leaving the open intact. If you cannot explain the video's job in one spoken clause before second two, the hook is not finished.
#Fixes
17.Which hook types are worth testing first?
Start with three families, not fifteen: a problem call-out (“if your …”), a pattern interrupt (unexpected visual or confession), and a specific proof or demonstration. Those three cover most of the work a 9:16 open has to do. Question hooks and “wait, don't scroll” lines are overused and often read as ads instantly. Native-sounding, specific, slightly unfinished lines beat slogans. Write them under 15 words, put the same three against two avatars, and let cold traffic pick. Once a family wins, make more variants inside that family rather than adding a fourth unrelated type into the same test.
#Fixes
18.Does on-screen text in the first two seconds help hook rate?
Usually yes, because a large share of feed viewing starts muted. A short, high-contrast line that matches the spoken hook gives you a second chance at the thumbstop. Keep it inside safe zones, under about seven words, and do not duplicate a long paragraph the avatar is already saying. Mismatched text — a joke on screen, a claim in the voiceover — burns trust in the same two seconds you are trying to win. Captions for the full video are a separate job from the hook card. If you only have time to design one text element, make it the opening line.
#Fixes
19.Will a more realistic avatar automatically raise hook rate?
No. Realism helps only when the current face is the reason people bounce — obvious artefacts, wrong age band, or a setting that screams studio. A slightly more realistic avatar delivering a slogan will still lose to a merely decent avatar delivering a specific, human line. Casting and copy both sit in the first two seconds; copy is usually the cheaper test. Run the 3 × 2 batch before you commission a custom clone. If both stock faces hook under 20% on cold traffic with good scripts, then you have a casting or offer problem, not a render-quality problem.
#Fixes
06
False Signals
20.When is a high hook rate a false signal?
When people stop and then leave. A 40% hook with hold well under 12% means the open over-promised or used a curiosity trick the body could not pay off. Clickbait faces, fake arguments, “don't make this mistake” and shocking first frames manufacture hook rate and destroy CPA. Another false signal is a high hook on a tiny, warm audience or a placement mix you will not scale. Always pair hook with hold (healthy 12–20%) and cold outbound CTR (1%+). If those two are healthy, the hook is real. If they are not, you bought a pause, not attention.
#False Signals
21.Can a bad hook still produce a good CPA?
Occasionally, on retargeting or on a tiny converting audience that already wants the product. It is not a strategy. A weak hook on cold traffic starves the rest of the funnel of people, so even a strong offer cannot show you what it can do. You also cannot scale a low-hook ad without paying for a lot of ignored impressions. If CPA is acceptable and hook is under 20% on cold, treat it as an offer that deserves a better open, not as proof that hook rate does not matter. Write three new openings; do not scale the quiet one.
#False Signals
22.Does a 3-second autoplay count as a real hook?
Sometimes no. Autoplay can credit a 3-second view when the viewer never really looked. That is one reason hook rate can look healthy while comments, outbound CTR and hold rate look dead. You cannot fully strip autoplay out of Meta's 3-second metric, so you compensate by never reading hook alone. If hook is 32% and hold is 5%, believe the hold. If hook is 32%, hold is 15%, and outbound CTR is 1.2% on cold, believe the hook. The metric is useful; it is not a lie detector. Pair it, or you will scale wallpaper.
#False Signals
23.Why is my hook rate high on previews and low in the feed?
Because you are not the customer, you are watching on a laptop, and you already know what the ad is. Editor previews and Slack comments measure taste; hook rate measures a cold thumb on a phone. Loud music, a polished grade and a clever line all win in a review meeting and lose in Reels. The only preview rule that helps is: watch it on mute, 9:16, at arm's length, and drop it if the job of the ad is unclear before second two. Then still ship a 3 × 2 test. If the team favourite dies and the slightly messy one hooks at 30%+, believe the feed.
#False Signals#Testing
07
Testing
24.How many hook variants should I test at once?
Three, against two avatars, on one body script — 6 creatives. That is the starting batch we recommend because it separates copy from casting without shredding budget. More than about six to eight variants in one ad set on a modest daily budget turns the test into noise: nothing reaches ~1,000 impressions in 48–72 hours. Once a hook family wins, iterate inside that family (new first line, same structure) rather than throwing eight unrelated concepts into the same set. Volume is useful; simultaneous volume is not the same as learning.
#Testing
25.How long before I trust a hook-rate reading?
48–72 hours or about 1,000 impressions per variant, whichever comes first. Earlier than that, delivery is still bouncing between people and placements, and a 12% versus 40% “winner” is often just who got the first Reels burst. Do not require a statistically perfect test; require a directional one you will actually wait for. If a variant is still under ~20% hook after that window on cold traffic, kill the opening, not necessarily the body. If two variants are both above 30% and CPA is not in yet, let them run rather than crowning a hook on impressions alone.
#Testing
26.Should I optimise the campaign for hook rate?
No. Optimise for the business event — purchase, lead, add to cart — and use hook rate as a diagnostic, not as a delivery goal. Meta will happily find people who watch three seconds if you ask it to; those people are not automatically buyers. Hook rate tells you whether the creative is earning the right to be judged on CPA. It does not replace CPA. The workflow is: launch on the conversion objective, read hook and hold to decide which creatives deserve more budget, then kill or iterate the rest. Vanity optimisation is how accounts fill with entertaining ads that do not pay.
#Testing
27.How do I test hooks without rebuilding the whole ad each time?
Lock the body script, the proof beats and the CTA, and only regenerate the opening scene. That is the point of a scene-based workflow: three new first lines should not cost three full rewrites. Keep the same avatar on a hook test if you already have a winning face; keep the same hook and swap avatars if you are testing casting. Mixing both at once on a small budget is how you learn nothing. Klip Kanvas is built for that iteration loop — new hook, same body — so you can ship the 3 × 2 batch in one sitting instead of a new shoot.
#Testing#Fixes
Ready to put this into practice?
Create your first AI UGC video ad in minutes — no filming, no actors, no editing.
