Meta hook rate below 25% triggers algorithm deprioritisation and inflates CPMs. The 2026 benchmarks: 25-30% healthy, 30-40% strong, 40%+ unicorn. How Andromeda scores hook rate and the structural fix.
Meta's Andromeda algorithm uses three-second video view rate as a primary creative-quality signal in 2026. Below 25%, the auction surcharges your impressions and CPMs spike. Above 30%, delivery improves and costs fall. Here are the benchmarks, how Andromeda scores them, and the structural fix.
Key takeaways
Hook rate is the percentage of people who watch the first 3 seconds of your video ad, divided by total impressions. It is the most important early signal the algorithm receives about your creative quality.
On Meta, the algorithm uses hook rate (alongside other engagement signals) to determine how much to charge for your ad. A high hook rate earns lower CPMs because the algorithm classifies your content as valuable. A low hook rate triggers higher CPMs because the algorithm classifies your content as low-quality and charges a premium to show it.
Andromeda is Meta's 2024+ ad ranking system. It uses early engagement signals — including three-second video view rate — to predict creative quality before an ad has accumulated meaningful conversion data. Low scores trigger CPM surcharges and reduced delivery weight in the auction.
This is why hook rate has moved from a “nice to know” into a primary bid-pricing input in 2026. Meta has not published Andromeda's full feature set, but every advertiser-visible signal points to three-second video view rate being one of the top creative-quality inputs to delivery prioritisation. Two ads with identical targeting and offers can deliver at dramatically different CPMs based on hook rate alone.
Algorithm actively deprioritises delivery. CPMs will be significantly inflated. The hook is failing.
The algorithm is delivering but charging premium CPMs. Your hook is not earning attention efficiently.
Competitive hook performance. You are earning reasonable CPMs and the algorithm sees your content as quality.
Above-average hook performance. Lower CPMs and efficient delivery. The algorithm rewards you with better placements.
Exceptional. These are rare ads that stop the scroll consistently. Analyse the hook structure — this is a formula worth protecting.
These are fundamentally different metrics that answer different questions. Confusing them leads to bad creative decisions.
| Hook Rate | Hook Scaling Rate | |
|---|---|---|
| measures | How well ONE ad hooks viewers | How well a hook TYPE performs across contexts |
| scope | Single ad performance | Cross-brand, cross-category reliability |
| Use case | Optimising a specific creative | Choosing which hook archetype to use for your next ad |
Improving hook rate is not about writing better copy for the same type of hook. It is about choosing the right hook archetype for your category and executing it with structural precision.
Match hook archetype to category
Different categories respond to different hook types. Skincare responds to Transformation Reveal hooks. Software responds to Pattern Observation hooks. Use data, not intuition.
Front-load the value proposition
The hook must deliver value in the first 1-3 seconds. Not a logo. Not a brand intro. The value.
Create pattern interrupts
The scroll is a pattern. Your hook must break it. Unexpected visuals, counter-intuitive statements, and direct address all create interrupts.
Test structurally different hooks
Do not test five versions of the same hook type. Test five different hook archetypes: Diagnostic Question, Challenge Statement, Shock Reframe, Object Intro, Authority Setup.
Analyse decoded hook performance
Study which hook archetypes perform in your category by looking at decoded ads, not surface-level swipe files.
25-30% is healthy, 30-40% is strong, and 40%+ is unicorn territory. Below 25%, Meta's Andromeda algorithm classifies the content as low-value and inflates CPMs. Above 30%, delivery improves and costs fall. These benchmarks apply across most D2C verticals in 2026.
The single biggest non-targeting driver in 2026 is hook rate below 25%. Andromeda uses three-second video view rate as a content-quality signal and surcharges low-hook ads in the auction. Other contributors are creative fatigue, audience saturation, and seasonal CPM cycles — but hook rate is the lever you can move fastest.
Three-second video views divided by impressions. If 1,000 people see your ad and 300 watch past 3 seconds, your hook rate is 30%. Meta surfaces this directly inside Ads Manager — the underlying metric is "3-second video views" relative to "impressions."
Meta has not published Andromeda's full feature set, but every advertiser-visible signal points to three-second video view rate being one of the top creative-quality inputs to bid pricing and delivery prioritisation. Ads with strong hook rates earn lower CPMs and better placements; ads with weak hook rates are surcharged in the auction.
TikTok doesn't surface a named "hook rate" metric — the equivalent is 2-second or 3-second video view rate. TikTok is less forgiving than Meta: you have 1-2 seconds to prove value, where Meta gives you closer to 3. The benchmark bands are similar but the time pressure is tighter.
Test structurally different hook archetypes (Diagnostic Question, Challenge Statement, Shock Reframe, Object Intro, Authority Setup), not five variants of the same hook type. Different categories respond to different archetypes — skincare responds to Transformation Reveal hooks; software responds to Pattern Observation hooks. Use category data, not intuition.
Yes. Advantage+ Shopping and Advantage+ App campaigns lean even harder on creative-quality signals than legacy campaign types because targeting is largely automated — creative is the primary differentiator. Weak hook rate inside Advantage+ shrinks the audience the algorithm can match you to and inflates CPA.
Create your first AI UGC video ad in minutes — no filming, no actors, no editing.
Try Klip Kanvas freeA diagnostic guide to the five most common causes of ROAS decline on Meta for D2C brands: hook fatigue, creative structure decay, weak mechanism clarity, generic proof, and scaling fragility. Includes diagnosis framework and structural fix playbook.
Why generic stock models are dying in 2026 and what consistent, brand-specific AI talent is replacing them with — the signals, the contributing factors, and what to do about it.
Paste a product link and Klip Kanvas writes the script, casts the creator and renders the ad — no filming, no actors, no editing.
