Brand Identity in AI Ads: Scene Bibles, Outfit Locks, Product Fidelity

Brand consistency in AI ads is not a model setting — it is a production discipline. The method that works in practice is a Scene Bible: before generating anything, you lock one location, one outfit, and a short list of look constants, and you feed real product reference images into every single shot so the model never gets to imagine the product. Skip this, and you get eight shots that are each individually fine and collectively incoherent.

Why do AI ads fall apart as a set?

Generative video models have no memory between shots. Every generation starts from zero: a new kitchen, a new sweater, a new color grade, and — this is the dangerous one — a new version of your product. Each shot can pass review on its own. The failure only appears when you cut them together and the ad reads like stock footage from six different libraries.

We learned to describe the failure mode with one phrase: “four drinks on four different tables”. You ask for a six-shot ad about one beverage, and you get six technically competent shots in which the cup changes shape, the lid changes color, the table changes material, and the daylight changes season. No single shot is wrong. The set is.

A 30–40 second ad is typically 6–12 shots in our pipeline. That means 6–12 independent chances for the model to reinvent your world. Consistency does not emerge from better prompts shot by shot — it has to be imposed from above, by constants that every shot inherits.

What is a Scene Bible?

A Scene Bible is a short document — often under a page — that fixes everything the campaign is not allowed to vary. Ours has three sections: one location, one outfit, and the look constants. Every prompt for every keyframe in the campaign carries these constants verbatim. The Scene Bible is written once, before the first generation, and it does not get edited mid-campaign.

One location

Pick a single, specifically described setting and keep the whole ad in it: “small sunlit café counter, pale oak surfaces, white tile behind the espresso machine” — not “a café”. Specificity is what makes regeneration converge: when shot 4 fails QA and you regenerate it, a vague location description produces a different room every time, while a specific one produces variations of the same room.

One outfit lock

If a presenter or creator appears in more than one shot, their outfit is locked to a written description with color, garment type, and one identifying detail — and that description ships in every prompt. Outfits drift even faster than faces: a model will happily move your presenter from a black crewneck to a grey hoodie between shots 2 and 3, and viewers read that instantly as a cut to a different person. Face-level identity is its own problem with its own mitigations — we cover that separately in how we keep AI characters consistent — but the outfit lock is the cheap half of the fix, and teams skip it constantly.

Look constants

• Palette: 2–3 named colors that should dominate every frame, usually the brand colors plus one neutral.
• Lens feel: one phrase like “35mm handheld, shallow depth of field” repeated in every prompt — mixing a phone-camera look with a cinema look across shots is one of the loudest continuity breaks.
• Light: time of day and quality (“soft morning window light”) locked once. Light direction changing between shots is the single most common giveaway that an ad was generated piecemeal.
• Energy: a one-line note on motion (“slow push-ins, no whip pans”) so the edit cuts together at one tempo.

How do you build a Scene Bible? Step by step

1. 01Collect real product reference images first — clean shots of the actual product from 2–3 angles, with labels readable. These get fed into every generation; they are not optional inspiration.
2. 02Write the location in one sentence with at least three concrete physical details (surfaces, fixtures, what is behind the subject).
3. 03Lock the outfit: garment, color, fit, one identifying detail. If you use a persistent brand character, this is part of its permanent definition — the same logic behind virtual influencers.
4. 04Pick the look constants: palette, lens feel, light, motion energy. Four lines, no more.
5. 05Generate 3–4 test keyframes of different moments using the full constant block, side by side. If they look like frames from one film, the bible holds. If not, tighten the vaguest line and retest.
6. 06Freeze it. Paste the constant block into every shot prompt for the rest of the campaign, unedited.

This works because our whole process is keyframe-first: we generate still keyframes, grade them, and only then animate the survivors via image-to-video. The Scene Bible is enforced at the keyframe stage, where a rejected image costs cents and seconds — not at the video stage, where a rejected clip costs a render queue slot and 1–5 minutes of waiting. The full flow is described in our AI ad production pipeline.

How do you keep the product faithful?

The product is the one element where “close enough” is a defect. Three rules cover most of it.

First: always generate from real product reference images. Reference-aware image models can hold a product’s shape, color, and label remarkably well — but only if you give them the real thing to hold onto. A text description of your packaging is an invitation to invent.

Second: never let the model imagine the product. If a shot does not need the product visible, fine. But the moment the product is in frame, the reference images go into the generation. There is no shot where “roughly our bottle” is acceptable.

Third: check labels and logos in QA, every time. Models love to redraw type. A label that reads correctly at a glance will, on inspection, have a mangled letterform, a doubled word, or a logo that is 90% right — which is worse than 0% right, because 90% right ships. Image-to-video adds a second layer of risk: even a perfect keyframe label can drift over the course of a clip, which is one reason we keep identity-critical clips short, 4–8 seconds, and re-anchor every shot on a fresh keyframe.

Why does “don’t invent SKUs” need to be a written rule?

Because the model will do it, cheerfully and plausibly. Ask for “a customer enjoying a drink from the brand” without constraints and you may get a frappé with whipped cream, an iced matcha, a smoothie in a branded cup — none of which the brand sells. The model is pattern-matching to “beverage brand ad”, not to your actual menu. The same applies to flavors, sizes, colorways, and bundle packs in any category.

The fix is explicit negative scope in the Scene Bible: list what the brand actually sells, name the hero SKU for the campaign, and state that no other products may appear. This sounds pedantic until the first time a client asks why the ad features a drink they have never made.

The continuity checklist we run on every shot

Every keyframe and every clip is graded against a checklist before it enters the edit — machine-graded first, human spot-checked after. Roughly speaking we generate 2–4 candidates per shot to keep one. This is the checklist; copy it.

• Product matches the reference images: shape, proportions, cap/lid, material.
• Label and logo are legible and correct — read the actual letters, do not glance.
• No invented SKUs, flavors, or packaging variants anywhere in frame, including the background.
• Location matches the bible: same surfaces, same fixtures, same room.
• Outfit matches the lock: garment, color, identifying detail.
• Light direction and quality match the bible and the neighboring shots.
• Palette holds: brand colors present, no stray dominant hue.
• Lens feel consistent: same focal-length character and depth of field as the rest of the set.
• No artifacts: extra fingers, warped text in background signage, melted edges on the product.
• For video: identity and label hold for the full clip, not just the first frame.

What does skipping this actually cost?

The trap is that the cost is invisible at the shot level. Each generation looks fine, so nothing flags it. The cost lands at assembly, when the edit refuses to feel like one piece, and the only honest fix is regenerating half the shots — now under constants you should have written on day one. We have watched the no-bible path roughly double the generation volume on a campaign: the same 2–4 candidates per shot, but run twice, once before the bible existed and once after.

There is also a softer cost: trust. Viewers cannot articulate “the light direction flipped between shots”, but they feel it, and what they feel is cheapness. For AI UGC ads, where the entire format trades on feeling real and native, incoherence is not a style problem — it is a credibility problem.

What we would do in your place

Before generating a single frame: write the bible. One location sentence with three concrete details, one outfit lock, four look constants, the real SKU list, and a folder of clean product reference photos. It takes under an hour and it is the highest-leverage hour in the whole production. Then enforce it at the keyframe stage, where rejection is cheap, and run the continuity checklist on everything that survives.

Or hand it to a pipeline that already works this way. Scene Bibles, reference-fed product fidelity, and machine-graded continuity QA are built into how we produce AI UGC ads — every campaign gets one world, not eight shots from eight different ones.