Script breakdown

What it is

Script breakdown maps the body of an ad — everything after the hook — to a recognizable copy framework, identifies which Cialdini levers fire and when, measures the pacing of the persuasion arc, and pinpoints the moment where attention is most likely to drop after the hook has done its job.

The body's one job is to turn earned attention into intent. The hook stopped the scroll. The body has to answer: "OK, now make me care." How that case gets made — whether it follows AIDA (Attention → Interest → Desire → Action), PAS (Problem → Agitate → Solve), BAB (Before → After → Bridge), PASTOR, or no named structure at all — determines whether the viewer reaches the CTA or drops out somewhere in the middle.

The five frameworks aren't equally useful across all traffic types. PAS defaults to retargeting and Problem-Aware cold traffic where the pain is already loaded. AIDA defaults to cold prospecting. BAB earns its keep on transformation products — Athletic Greens, Liquid IV — where the contrast state does heavy lifting. PASTOR handles long-form supplement-style arguments where testimony and offer specificity matter. FAB serves spec-heavy products (Ridge wallet, Mejuri pieces) where features actually differentiate. Knowing which one you're running tells you whether the body is matched to its audience.

How AdRevila scores it

One Gemini model call watches the entire video from first frame to last. The script breakdown section of the report prints:

• Copy framework — primary framework tag (AIDA, PAS, BAB, PASTOR, FAB, or "narrative/none"). When the ad straddles two, both are noted with the dominant one first
• Cialdini levers — up to three, ranked by visibility in the creative, with the timestamp range where each fires
• Pacing assessment — does the body move at the right pace for the platform? A 45-second PAS argument that belongs in a 20-second TikTok is flagged
• Leak point — the specific window (in seconds) where hold-rate is most likely to drop, and the reason: brand transition too early, no proof before the desire phase, agitation that runs too long before the solve

The model is watching pacing, visual reinforcement, on-screen text that supports the copy argument, and audio delivery — not just words. An AIDA body that's correctly structured but never delivers the social-proof beat visually (review count, before/after, specific number) scores lower than one that reinforces each phase with the matching visual.

How to read it

Start with the framework tag and ask: is this the right framework for the Schwartz stage implied by the hook?

A PAS body on an ad with a proof-first hook is a mismatch. The hook promised a result — the viewer's first question is "is this real?" — and the body's opening with a problem frame, which restates a pain the viewer didn't ask about. That mismatch costs hold-rate in the 3–8 second window, which is where most of the leak occurs.

The Cialdini lever list tells you which mechanisms are doing the persuasion work. A body that fires zero levers is descriptive, not persuasive — it's telling the viewer what the product does without giving them a reason to trust it. The most reliable DTC bodies use two to three: social proof (review count, specific number, recognizable brand) plus liking (relatable creator, unpolished moments) plus one more depending on the offer (scarcity for flash-sale SKUs, authority for supplement efficacy claims).

The leak-point flag is the most actionable output. If the model tags seconds 6–9 as the likely drop window — "brand transition before interest is established" — the edit is precise: delay the logo/product reveal until after the Interest phase has landed, or insert a micro-proof beat (one number, one face reaction) at second 5 to bridge the hook's promise into the body's argument.

Why it matters

Most creative analysis stops at the hook. Hook strength and thumbstop are easier to measure — platform dashboards report 3-second views directly. The body's failure is harder to see because hold-rate data is noisier and the drop is gradual rather than binary.

But the body is where CTR is won or lost. A hook that earns a 40% thumbstop into a body that drops to 15% completion at the 10-second mark is a worse ad than a hook with 30% thumbstop into a body that holds 35% completion through to the CTA. The first ad wastes more of the media budget it earns; the second converts a higher share.

Script breakdown makes the body's persuasion structure as auditable as the hook. When a DTC team briefs a new creative round, knowing the last ad's body ran a PAS structure that fired social proof and liking but dropped viewers at second 8 because the solve arrived too late is a precise brief: "move the solve earlier, keep the agitation tighter, add a specific number at second 5." That brief produces a testable hypothesis, not a preference.