Hook Rate: The Complete Guide
Hook rate is the percentage of viewers who watch past the first 3 seconds of a video ad. It is the earliest creative health signal — it predicts CPA before you spend enough to measure CPA directly. This guide covers the formula, benchmarks, hook types, and how to improve it.
Why Hook Rate Matters
3 Seconds Decide Everything
If the first 3 seconds don't stop the scroll, the rest of the ad is irrelevant. Hook rate measures this critical moment.
Earliest CPA Signal
Hook rate predicts CPA before you've spent enough to measure conversions. It's your leading indicator.
Diagnostic Tool
Bad hook rate + bad CPA = fix the opening. Good hook rate + bad CPA = fix the body or offer.
The first 3 seconds are the entire ad
- Hook rate = 3-second video views ÷ impressions × 100
- Also called "thumb-stop ratio" or "thumb-stop rate"
- The strongest early predictor of overall ad performance
In a feed-based ad environment, most people scroll past ads in under a second. The hook — the first 3 seconds of a video ad — is the only thing that determines whether someone stops scrolling and watches. Everything else in the ad (the body, the offer, the CTA) only matters if the hook works first.
Hook rate measures this critical moment. It tells you what percentage of people who saw your ad actually stopped and watched. A high hook rate means the opening is doing its job. A low hook rate means you're paying for impressions that never become views.
This metric goes by several names: hook rate, thumb-stop ratio, thumb-stop rate, and 3-second retention rate. They all measure the same thing: did the first 3 seconds stop the scroll?
- Metric: 3-second video views ÷ impressions × 100
- Benchmark: 30–40% on Meta, varies by platform
- Signal: earliest predictor of CPA performance
- Use case: diagnostic — tells you why an ad is or isn't working
What is hook rate?
- The percentage of impressions where the viewer watched at least 3 seconds
- Measures the hook's ability to stop the scroll
- Available as a custom metric in Meta Ads Manager and TikTok Ads
Hook rate is the percentage of viewers who watch past the first 3 seconds of a video ad. It measures the effectiveness of the hook — the opening of the ad that must capture attention before the viewer scrolls away.
Platforms like Meta report "3-second video views" as a standard metric. Hook rate is not a default column — you create it as a custom metric by dividing 3-second video views by impressions. Most performance media buyers add this column to every report because it's the first metric that becomes reliable, often within 24 hours of launching an ad.
Why 3 seconds?
Three seconds is the industry standard because it's the threshold where attention becomes intentional. In the first 1–2 seconds, an ad might appear in someone's feed as they scroll. By 3 seconds, the viewer has made a conscious decision to stop and watch. This is why 3-second views — not impressions, not 1-second views — are the meaningful threshold for measuring initial attention.
Hook rate vs other retention metrics
- Hook rate (3s) — measures whether the opening captures attention. Diagnostic for the hook itself.
- Hold rate (15s or 50%) — measures whether the body retains attention after the hook. Diagnostic for the middle of the ad.
- Completion rate — measures whether viewers watch the full ad. Less useful for performance creative because most conversions happen before the ad ends.
The hook rate formula
- Hook Rate = (3-Second Video Views ÷ Impressions) × 100
- Create as a custom metric in your ad platform
- Reliable with 1,000+ impressions
Example: 3,500 three-second views from 10,000 impressions = 35% hook rate
Setting up in Meta Ads Manager
- Go to Columns → Customize columns. In Ads Manager, click the Columns dropdown and select "Customize columns."
- Create a custom metric. Click "Create Custom Metric." Name it "Hook Rate" or "Thumb-Stop Ratio."
- Set the formula. 3-Second Video Views ÷ Impressions. Format as percentage.
- Save to your default view. Add it next to CPA and CTR so you see it on every report.
Minimum data for reliable hook rate
Hook rate becomes directionally reliable at 1,000+ impressions, which most ads reach within the first 24 hours. This is why it's the earliest signal — you can read hook rate a full day or two before you have enough conversion data to evaluate CPA. At 5,000+ impressions, the signal is strong enough to make decisions.
2026 hook rate benchmarks
- Below 20%: problem — the hook is not stopping the scroll
- 20–30%: average — room for improvement
- 30–40%: good — the hook is working
- 35%+: strong — high-performing creative
| Hook Rate Range | Assessment | Action |
|---|---|---|
| Below 20% | Hook is failing — ad is invisible | Replace the hook entirely. The opening is not working. |
| 20–30% | Average — functional but not competitive | Iterate. Try faster product reveal, stronger opening line, text overlay. |
| 30–40% | Good — the hook is doing its job | Focus on optimizing the body and CTA. The hook is not the bottleneck. |
| 35%+ | Strong — top-performing creative | Scale. Produce variations with this hook style on different body narratives. |
Benchmark context
These benchmarks are for Meta (Facebook + Instagram) feed placements in 2026. Hook rate varies by category — entertainment ads tend to have higher hook rates than B2B SaaS ads — but the 30–40% range holds as a useful general benchmark across most consumer and DTC categories.
Important: hook rate is relative to your own account's historical performance. If your best ads consistently hit 28%, an ad hitting 35% is a strong signal even though 35% is "good" on an absolute scale. Track your own benchmarks alongside industry averages.
Hook rate as a diagnostic tool
- CPA tells you what happened. Hook rate tells you why.
- Bad hook rate + bad CPA = problem is the opening
- Good hook rate + bad CPA = problem is the body or offer
The most valuable use of hook rate is as a diagnostic tool, not just a performance metric. When CPA is bad, hook rate tells you where the problem is — in the opening (hook) or in the body and offer that follows.
The diagnostic framework
| Hook Rate | CPA | Diagnosis | Fix |
|---|---|---|---|
| Low (<20%) | Bad | Hook is the problem — ad never gets seen | Replace the hook entirely |
| Good (30%+) | Bad | People watch but don't convert — body or offer is weak | Keep the hook, fix the body/offer/CTA |
| Low (<20%) | Good | Small but highly qualified audience converts — rare but possible | Improve hook to increase volume while maintaining conversion |
| Good (30%+) | Good | Winner — the entire creative is working | Scale. Produce variations to extend lifespan. |
The most common mistake
When CPA is bad, most teams change everything — new script, new creator, new format. This destroys any signal from the original creative. Use hook rate to isolate the problem. If hook rate is good, keep the hook and only change the body. If hook rate is bad, only change the hook and keep everything else. This is the foundation of the creative testing discipline: isolate variables.
Want OKAD to test hooks for your brand?
We produce 5–10 hook variants per concept, test with $50–$100/day, and find the openings that stop the scroll. You get data on what hooks work for your product and audience.
Hook types that stop the scroll
- Different hook types trigger different psychological responses
- Test multiple hook types per concept — you can't predict which wins
- The best hooks combine a strong opening line with an unexpected visual
| Hook Type | Example | Why It Works |
|---|---|---|
| Question hook | "Why is nobody talking about this?" | Creates a curiosity gap that demands an answer |
| Stat hook | "I cut my ad costs by 47% in one week" | Specific numbers create credibility and intrigue |
| Before/after hook | "My skin 6 months ago vs today" | Visual contrast stops the scroll instantly |
| Controversy hook | "Your morning routine is actually making you tired" | Challenges existing beliefs, triggers engagement |
| "I tried X" hook | "I tried this $12 tool for 30 days" | Personal experiment format creates trust and curiosity |
| Result-first hook | "I made $4,200 in passive income this month" | Leads with the outcome, makes viewer want the method |
| Pattern interrupt | [Unexpected visual + "Wait, what?"] | Breaks the expected feed pattern, forces attention |
| Social proof hook | "2 million people use this and I just found out" | Social validation triggers FOMO |
Combining visual and verbal hooks
The best-performing hooks combine a strong opening line with a visual that doesn't match what the viewer expects. A creator saying "I can't believe this is only $12" while holding up an unexpectedly premium product works better than either element alone. The mismatch between expectation and reality is what creates the thumb-stop moment.
How to improve hook rate
- Show the product within the first 2 seconds
- Use specific numbers instead of vague claims
- Add bold text overlays that reinforce the verbal hook
Seven techniques to improve hook rate
- Faster product reveal. Show the product or result within the first 2 seconds. Don't waste time on intros or context before showing what the viewer came to see.
- Specific numbers. "I saved 3 hours a week" beats "This saves so much time." Specificity creates credibility and intrigue. Vague hooks blend into the feed; specific hooks stand out.
- Unexpected visuals. Start with something that doesn't look like an ad. An unusual camera angle, an unexpected object, a visual that creates a "what is that?" reaction. Feed algorithms favor content that generates engagement signals.
- Text overlays. Bold, contrasting text in the first frame reinforces the verbal hook. Many viewers scroll with sound off — if the text overlay doesn't stop them, the audio hook never gets heard.
- Movement in frame one. Static openings get scrolled past. The first frame should have motion — the creator moving, a product being revealed, a transition happening. Motion triggers pattern recognition in the viewer's peripheral vision.
- Address the viewer directly. "You need to see this" or "If you're still doing X, stop" creates an immediate personal connection. The viewer feels spoken to, not marketed at.
- Cut the preamble. Delete the first 1–3 seconds of most creator footage. Most creators start with a breath, a glance at the camera, or a slow build-up. Cut straight to the hook. The raw, abrupt opening often performs better.
Hook rate by platform
- TikTok: higher hook rates (autoplay + sound on), faster fatigue
- Meta: moderate hook rates, sound-off optimization critical
- YouTube: lower hook rates but higher intent once hooked
| Platform | Typical Hook Rate | Key Factor | Optimization Note |
|---|---|---|---|
| Meta (FB + IG) | 25–40% | Many users scroll with sound off | Text overlays are essential. Visual hook must work muted. |
| TikTok | 30–50% | Autoplay with sound on, full-screen | Audio hooks matter more. Native style critical. |
| YouTube Shorts | 25–35% | Higher intent audience, more competitive | Educational hooks outperform pure curiosity hooks. |
| YouTube In-Stream | 15–25% | Skippable after 5 seconds | Must hook in first 5s before the skip button appears. |
Cross-platform hook strategy
A hook that wins on one platform often works on others, but the execution must be adapted. A Meta hook that relies on text overlays needs to be re-produced with a stronger audio component for TikTok. A TikTok hook that relies on trending audio needs a visual substitute for Meta's sound-off environment. The insight (which message stops the scroll) transfers; the format must be platform-native.
Hook testing: the highest-leverage test in advertising
- Same body + same CTA + 5 different hooks = isolated hook test
- OKAD tests 5–10 hook variants per concept with $50–$100/day each
- Hook tests produce actionable data within 3 days
Hook testing is the foundation of the creative testing hierarchy. Because hooks have the highest impact on performance, testing hooks first produces the fastest and most reliable learnings.
How to run a hook test
- Keep the body and CTA identical. The only variable is the first 3 seconds. This isolates the hook's impact.
- Produce 5–10 hook variants. Use different hook types: question, stat, before/after, controversy, "I tried X." Cast with different creators if possible.
- Launch all variants simultaneously. Same audience, same placement, same budget ($50–$100/day per variant).
- Read hook rates after 24–48 hours. At 1,000+ impressions per variant, hook rate is directionally reliable.
- Decide after 3–5 days. Kill hooks below 20%. Iterate hooks in the 20–30% range. Scale hooks above 30%. Then test body narratives against the winning hooks.
Hook rate and CPA correlation
Hook rate and CPA are correlated but not perfectly. A high hook rate means more people see the body and CTA, which generally leads to more conversions at a lower cost. But a hook that attracts the wrong audience (high hook rate from curiosity-seekers who never buy) can have a high hook rate and bad CPA. This is why hook rate is a diagnostic metric, not the decision metric. CPA is always the final judge.
OKAD tests hooks systematically
We produce 5–10 hook variants per concept using modular production, test each with controlled spend, and identify which openings stop the scroll for your specific product and audience.
Key terms
Frequently asked questions
Ready to find hooks that convert?
Tell us about your product and current ads. We'll audit your hook rates, produce tested variants, and find the openings that stop the scroll for your audience.
This guide is based on hook rate data from creative testing programs run by OKAD Agency, supplemented by publicly available Meta and TikTok advertising benchmarks. Hook rate ranges and CPA correlations are directional industry figures based on aggregate campaign data. Individual outcomes depend on product, audience, creative quality, platform, and market conditions.