Why Meta's Hook Window
Is Now Under 2 Seconds
Viewers decide in 1.7 seconds. Here are 5 hook formulas that stop the scroll and the 4 hook mistakes silently killing your Meta performance.
1.7 Seconds: The Death of the Slow Burn Ad
The 5-second brand intro is dead. Meta internal engagement data is unambiguous: viewers decide within 1.7 seconds whether your content deserves their attention.
The engagement decision -- scroll past or continue watching -- happens within 1.7 seconds. No signal means no distribution means wasted impression. Andromeda uses this signal to determine ad quality score and distribution priority.
Andromeda uses 3-second video views as a primary quality signal. Every impression failing to generate a 3-second view is scored negatively. Enough negatives and CPMs rise, reach contracts even with no changes to targeting or budget.
Users scroll Reels at roughly one frame per second. The brain makes a pattern-match decision before conscious evaluation begins. Hooks that interrupt this pattern trigger a pause reflex that is neurologically involuntary.
Feed hooks can be slightly slower at 2-3 seconds because users are already paused. Reels and Stories hooks must be sub-2-seconds because the next piece of content is one swipe away. The same hook often fails in Reels after working in Feed.
Sound-on Reels means audio can be your hook: a surprising sound effect, recognisable music intro, or bold voiceover line. A strong audio hook stops scrolling before any visual even registers.
Test hooks independently before testing full ads. Create 5-10 hook variations with the same body and CTA. The hook with the highest 3-second view rate is your winner. Build your full creative library around it.
5 Hook Formulas That Stop the Scroll in 2026
Open with a specific surprising result stated immediately. Example: We cut our client CPL by 62% in 30 days. No setup, no brand intro. The number creates instant credibility curiosity and viewers want to know how. Works across all placements and audiences.
Directly challenge the viewer current belief. Example: Your Meta ads are probably broken. Here is why. Creates mild discomfort that demands resolution. Highest CTR for performance marketers and B2B audiences who self-identify as competent practitioners.
A visual that does not fit the expected context. Someone doing something unexpected in the wrong setting. The brain registers this does not compute and pauses to resolve the mismatch. High completion rates across all audience types.
Open with a specific recognisable situation. Example: If you are spending $10k per month on ads and still guessing what is working. The viewer self-selects -- if it resonates they continue. Self-selection makes everything downstream more relevant.
Hook Performance Benchmarks
The 4 Hook Mistakes Killing Your Meta Performance
Mistake 1: Starting With Your Logo
Brand intros are a TV-era convention. On Meta, starting with your logo signals this is an ad before the viewer has any reason to care. You have used your 1.7-second window on brand recognition rather than attention capture. Start with the hook. Put your logo at the end.
Mistake 2: Slow Text Reveals
Animated text that fades in character by character feels polished in a presentation. On Reels it is a scroll trigger. Text must appear instantly and be readable in a single glance. Use bold high-contrast text communicating the hook in one look.
Mistake 3: The Did You Know Opening
Never start with "Did you know?" or "Have you ever?" These openers signal educational content incoming. Bold claim, direct challenge, or pattern interrupt always outperform question-based openers on conversion-objective campaigns.
Mistake 4: Skipping the Sound Design Brief
Most creative briefs specify visuals in detail and mention audio as an afterthought. In a sound-on environment, audio is the hook for a significant portion of your audience. Brief the audio hook with the same specificity as the visual hook in every Reels creative brief.
The hook diagnostic: Watch your ad with the sound off and cover everything after the first 2 seconds. Would a stranger stop scrolling based on what they see? If not, rewrite the hook. A strong hook works silently. Great audio makes it better.
Want a Hook Audit for Your Current Meta Ads?
I will review your existing creative library, score each hook against Meta engagement criteria, and brief 5 replacement hooks for your lowest performers.
Meta Hook Window FAQ
Meta data shows viewers decide within 1.7 seconds. For Reels treat 1.5 seconds as your practical target. Feed ads have slightly more tolerance at 2-3 seconds.
A good hook does one of three things in under 2 seconds: states a surprising result, challenges a current belief, or creates a visual pattern interrupt that the brain needs to resolve.
Yes. Create 5-10 hook variations with the same body and CTA. The variation generating the highest 3-second view rate is your winning hook. Build your full creative library around it.
Yes but differently. Static image hooks are visual -- the first element the eye lands on must create a pattern interrupt or communicate a benefit instantly.
3-5 hooks per ad set maximum. More than 5 dilutes impression volume per variation, extending time to statistical significance.