Hook analysis

What it is

Hook analysis classifies the first three seconds of an ad and assigns a single primary archetype from the five that account for nearly every winning DTC creative: pattern-interrupt, bold-statement, question, direct-address, and proof-first. It also scores hook strength on a 1–10 scale, flags whether the hook is earning a sound-off viewer's attention (on-screen text, visual interrupt), and notes a Schwartz-stage match or mismatch against the inferred audience.

The five archetypes aren't taxonomy for its own sake. Each archetype has a different mechanism for stopping the scroll, and each performs differently by platform and awareness stage. A pattern-interrupt earns thumbstop through a visual break against feed norms — the wide patio shot that doesn't look like an ad. A bold statement earns it through a claim that triggers disbelief: the viewer stays to see whether the proof arrives. A question earns it by starting an internal conversation the viewer can't mute. Knowing which one you're using tells you what you're betting on.

How AdRevila scores it

Paste a Meta Ad Library or TikTok Ad Library URL. One Gemini model call receives the full video and watches from the first frame. It is not transcribing and pattern-matching keywords. It sees the visual cut, the on-screen text, the audio, the pacing — the whole hook as a viewer would experience it.

From that single call the report prints:

• Archetype (primary) — one of the five, with a confidence note when the opening straddles two
• Hook strength — 1–10, calibrated against Schwartz's awareness stages and Hormozi's value equation (does the hook's promise set up a numerator that the body can pay off?)
• Sound-off legibility — pass/fail: does the first 1.5 seconds communicate a thesis to a viewer with audio off?
• Stage match — does the hook's framing match the inferred Schwartz stage? A proof-first hook opening with a revenue number is Solution-Aware framing; showing it to Unaware cold traffic is the most common waste in DTC media buying

No transcript guessing. No keyword matching on the audio. The model watches the ad.

How to read it

The archetype tag at the top of the report is the starting point, not the conclusion. The useful read is: archetype + strength + sound-off + stage match, together.

A pattern-interrupt scoring 8/10 on hook strength with a sound-off fail is a hook that's winning on visual novelty but losing half the feed — the ~50% of Meta viewers who scroll with audio off see no thesis in the first 1.5 seconds and keep scrolling. The fix is surgical: add on-screen text to the opening frame, not a new hook.

A direct-address hook scoring 6/10 with a stage-mismatch flag ("hook implies Problem-Aware framing; audience likely Solution-Aware") is telling you the copy is talking to the wrong person. The hook uses "are you struggling with X?" language against an audience that already knows the category and wants a reason to pick your brand. Reframing to solution-aware language — leading with selection, price, or social proof — is typically a 15–30% thumbstop lift for that traffic type.

The strength score is calibrated against benchmarks from Cialdini social proof, Hormozi's value-equation numerator, and platform-native aesthetic norms. A 7+ usually means the hook earns the next five seconds. Below 5 usually means the body is performing against a leak that starts before anyone reads the copy.

Why it matters

The hook is the leakiest part of most ad funnels. A 10-point thumbstop lift compounds through every metric downstream — CTR moves because more viewers reach the body, CVR moves because a stronger hook self-selects a higher-intent cohort. Fixing a weak hook is the highest-leverage creative edit available.

Most operators don't know which of the five archetypes they're running — they know they have "a UGC ad" or "a testimonial." That description doesn't generate a brief. Knowing you're running a proof-first hook that scores 4/10 because the proof arrives at second 6 instead of second 1 generates exactly one edit: move the number forward. That's the difference between a revision that takes three hours and a revision that takes twenty minutes.

The archetype vocabulary also makes A/B testing structurally cleaner. Run the same body and CTA against three hooks — one pattern-interrupt, one bold-statement, one proof-first. The winning archetype tells you what your audience trusts. Run that insight into your brief template and the next creative round starts from signal instead of preference.