Mood is not a prompt.
It is a brand kit.

Generative image tools have trained a generation of designers to think of "mood" as a prompt — a paragraph you type into a text box, with comma-separated stylistic references and a couple of emoji for emphasis. "Vogue editorial 1998, golden hour, slight grain, cinematic, 35mm." That kind of paragraph.

This is a misunderstanding. Mood is not a prompt. Mood is a set of commitments. Specifically: a palette, a light direction, a roster of models you have decided you trust, a category of secondary objects, a typographic register, and a relationship to grain, vignette, and saturation. A real mood is what your art director would write down on a single page if you asked her to define the house, then sign at the bottom.

Why prompts fail at mood

A prompt is a single instance of an intention. The next prompt is a slightly different instance. Two prompts written by the same designer fifteen minutes apart will produce campaigns that are 80% similar and 20% different, and the 20% is exactly where mood lives.

You can see this in any portfolio of mid-2024 generative campaigns from independent brands. The visual identity drifts from one drop to the next. The model in the spring campaign has a slightly different jaw than the model in the summer campaign. The lighting in October feels warmer than in September. The brand reads as inconsistent — not because the designer is inconsistent, but because the prompts are.

A prompt is a sentence. A brand kit is a grammar. You ship campaigns in a grammar, not a sentence.

What a brand kit actually contains

In Drape, a locked brand kit holds six things. We chose these six because they are the six that, when locked together, eliminate the visual drift between campaigns.

Palette. Four to six colors, hex-specified, with weights (primary 60%, accent 25%, neutral 15%). Used as soft constraints during generation — the model will compose backgrounds, lighting, and accessories that respect the palette without you typing it every time.

Signature models. Three to five identity-trained models that become the recurring faces of your house. Same person, same nose, same eye shape, every campaign forward. This is the single largest contributor to a coherent brand identity, and it is the thing that prompts fundamentally cannot do.

Mood baseline. One of our base moods (Editorial Brutalist, Soft Atelier, Sun-Soaked, Berlin Grey, Cinematic, Pastel Calm) or a custom one you train. This sets light, location archetype, styling cues, and secondary-object category.

Type lockup. If you want generated frames to include your wordmark as a soft watermark, or to compose with negative space for typography overlay, this is where you tell the kit. Most brands skip this. The ones that do not, ship cleaner editorial.

Grain and finish. Drape can apply a learned film-emulation pass over every generated frame — Kodak Portra 400, Fuji Pro 400H, push-processed Tri-X, or a custom curve. Once you lock this, every campaign reads as shot on the same stock. This is the second largest contributor to brand coherence.

Drape-specific overrides. Custom instructions to the underlying model about how to handle your specific fabric library. If you ship a lot of textured wool, you tell the kit to bias generations toward shorter focal lengths that flatter wool weave. If you ship a lot of silk, you bias toward longer focal lengths. These are the kind of nudges a creative director would whisper to a photographer mid-shoot.

How long it takes to build one

The first time, about ninety minutes. You sit down with our brand kit wizard (or an Atelier success engineer who walks you through it), and you make six decisions. The wizard generates a sample editorial from each decision pass so you can see the cumulative effect.

The second time, about twelve minutes. By the second kit, you have internalized which decisions matter most for your house. Most brands end up with two or three locked kits — one for their core line, one for resort/special drops, one for bridal or limited collaborations.

The third time and forever after, brand kits become reusable. You can branch a kit (copy and modify), export a kit for a partner agency, or version a kit to A/B test a refresh. The kit is the durable artifact. The campaigns are instances of it.

You spend ninety minutes once. You stop drifting forever.

The deeper point

The reason mood-as-prompt produces brand drift is the same reason that fashion houses with strong creative directors produce coherent decades and fashion houses without them produce confused decades. Coherence is a function of having a single locked grammar that everyone working in the house respects.

For most of fashion history, that grammar lived in the head of one person — Coco Chanel for Chanel, Yves for Saint Laurent, Phoebe Philo for Céline. The brand was coherent because the grammar was singular and persistent.

When that creative director leaves, the brand drifts until the next creative director re-codifies the grammar. This is how Gucci became confused between Frida Giannini and Alessandro Michele. This is how Burberry kept finding itself.

A brand kit in Drape is the same thing as the grammar in the creative director's head, except externalized, codified, and durable across hires. It is the modern equivalent of writing your house language down so the next person who picks up the camera shoots in the same dialect.

This sounds dramatic for a software feature. We promise we are not being dramatic. The brands inside Drape who have locked their kits ship 38% more campaigns per quarter than those who have not. They also report — anecdotally, but consistently — that their internal teams stop arguing about creative direction. The arguments live in the kit, not in every Slack thread.

That is the value. Stop fighting about mood. Lock the kit. Generate forever.