What is a brand’s visual DNA?
A brand’s visual DNA is the set of visual habits that repeat across its best images: the palette with hex codes, the light, the lens and framing, composition, texture and grain, styling, casting, era and typography. You find it by studying 10 to 20 of your own strongest images and writing down what they share. It carries across campaigns; a mood board sets the feeling of one.
Visual DNA in plain words
Brand guidelines usually cover the logo, the fonts and the tone of voice, and some say little about how the photos are made. Visual DNA fills that gap. It is written as measurements and habits: "light from camera left, soft, slightly warm", "product on the lower third", "fine grain, blacks never fully black". You can check each line with an eyedropper, a grid, or a look at where the shadows fall.
AI tools use the idea too. Google’s Pomelli, a free Google Labs marketing tool launched in October 2025, builds a "Business DNA" from your website and existing images: tone of voice, fonts, images and color palette. Visual DNA as used here is narrower and more detailed. It covers only the pictures, and someone who knows which of those pictures are right does the measuring.
The ten fields of a visual DNA
| Field | What to record | How to find it |
|---|---|---|
| Palette | Three to six colors as names plus hex codes, and the share of the frame each takes | Eyedropper on flat, evenly lit areas; average several spots |
| Light | Direction, softness, color, and how dark the shadow side goes | Look where shadows fall and how sharp their edges are |
| Lens and framing | Wide, standard or long-lens look; camera height; distance | Wide lenses stretch near objects and bend lines at the edges; long lenses flatten the background |
| Composition | Where the product sits, how much empty space, crop habits | Lay a 3x3 grid over each image and note where the product lands |
| Texture and grain | Clean or film grain, sharpness, how much retouching | View at 100% and compare the same kind of flat area across images |
| Color grade | Warm or cool, saturation, how deep the blacks go | Sample the darkest and brightest areas of each image |
| Styling | Surfaces, props, fabrics, tidy or lived-in | List every prop across the set and count repeats |
| Casting | Ages, looks, poses, gaze, hands | List who appears and what they are doing |
| Era | Which years the world looks like | Note clothes, cars, phones and interiors |
| Typography | Typefaces, and where text sits if it sits on photos | Note font names, sizes against the frame, positions |
How to extract it from 10 to 20 images
- Choose 10 to 20 images your brand made and would make again: best performers, favorite campaign frames, approved product shots. Leave out fan photos, competitor images and anything you would not reshoot.
- Put them in one grid at the same size, so repeats are easy to see.
- Sample the colors with an eyedropper on flat, evenly lit areas, and write each one down as a name plus a hex code. The tools and settings are in keeping brand colors accurate.
- Read the light from the shadows: which way they fall, how soft their edges are, how dark they get.
- Lay a grid over each image and note where the product sits and how much space is left empty.
- List every surface, prop and person across the set.
- Keep a habit if it shows up in most of the images. Anything that appears in only two or three belongs to a campaign.
- Test it. Give the DNA and one product photo to a model, render five images and drop them into the grid. The ones that stand out show what the DNA is missing.
A visual DNA template
VISUAL DNA Brand: ________ Built from: __ images, from ________ to ________ Version: ________ PALETTE (name, hex, where it appears, share of frame) 1. ________ #______ ________ __% 2. ________ #______ ________ __% 3. ________ #______ ________ __% LIGHT Direction: ________ Softness: ________ Color: ________ Shadow depth: ________ LENS AND FRAMING Look (wide / standard / long): ________ Camera height: ________ Distance: ________ COMPOSITION Product position: ________ Empty space: ________ Crop habit: ________ TEXTURE AND GRAIN Grain: ________ Sharpness: ________ Retouching: ________ COLOR GRADE Warmth: ________ Saturation: ________ Blacks: ________ STYLING Surfaces: ________ Props that repeat: ________ Never seen: ________ CASTING Ages: ________ Poses: ________ Gaze: ________ Hands and nails: ________ ERA ________ TYPOGRAPHY ON IMAGES ________ ONE-SENTENCE SUMMARY ________
A filled example, for a made-up brand
This one is for an invented brand of canned sparkling tea, to show the level of detail. It does not describe a real company.
VISUAL DNA Brand: made-up canned sparkling tea Built from: 16 images Version: 1 PALETTE 1. Chalk #F2EEE6, backgrounds and table tops, about 60% 2. Sage #9DAA8C, walls and fabric, about 20% 3. Tea amber #C8893A, the drink and the band on the can only, under 10% 4. Charcoal #2B2B28, shadows and type LIGHT Low morning sun from camera right. Hard-edged but not harsh. Long shadows falling left. Warm highlights, cool shadow side. LENS AND FRAMING Standard-lens look. Camera at table height. The can fills about a third of the frame height. COMPOSITION Can on the left third. Right third empty for copy. Table edge on the lower third line. TEXTURE AND GRAIN Fine film grain in every frame. Sharp on the can, soft behind it. Condensation as a few drops, never a full sweat. COLOR GRADE Warm highlights, neutral midtones, blacks lifted to charcoal. Low saturation except the amber. STYLING Linen, pale stone, one glass at most. Lemons appear in 3 of 16 images, so they are optional. Never seen: ice buckets, straws, neon. CASTING Adults 25 to 40. Hands and forearms more often than faces. Short, unpainted nails. Nobody looks at the camera. ERA Present day. No phones in frame. TYPOGRAPHY ON IMAGES None in the photo. Headlines set later in the brand serif. ONE-SENTENCE SUMMARY Quiet early-morning table scenes where the only warm color belongs to the tea.
Visual DNA, mood board and brand guidelines compared
| Question | Visual DNA | Mood board | Brand guidelines |
|---|---|---|---|
| What it describes | How your pictures look, measured | The feeling of one campaign | The whole brand: logo, type, voice, sometimes photography |
| Where it comes from | Your own best images | Images from anywhere, other brands included | The brand team or an agency |
| How long it lasts | Years, checked each season | One campaign | Until the next rebrand |
| What the model gets | Words: hex codes, light, lens, grain | Words about light, mood and energy | Only the lines that change a picture |
| Who keeps it | Brand manager | Art director | Brand team |
Campaigns get their variety from the mood board, and the DNA keeps them looking like the same brand. See how to use a mood board with AI, and how to write brand rules an AI can follow for turning the DNA into rules.
When to update it
Check the DNA each season against the newest approved images. If the last 20 approved images no longer match it, either the brand has moved or the images have drifted, and someone should decide which. Rebuild it after a rebrand. Keep old versions with their dates, so you can see how the look changed and roll back a change that did not work.
Questions people also ask
- Can an AI tool extract visual DNA for me?
- Partly. Tools can pull colors and describe light and composition, and Google’s Pomelli builds a Business DNA of fonts, colors, images and tone from your website. Check what any tool extracts against your own grid, because it reads whatever is on your site, old or off-brand photos included. See what Google Pomelli is.
- What if we have fewer than 10 good images?
- Use what you have and mark the DNA as a draft. Where you cannot measure, decide: pick the light direction, the grain and the crop. After your first AI campaign, rebuild the DNA from the images you approved.
- Should the logo be part of the visual DNA?
- Only where the logo appears inside photos: on the product, the packaging or a sign. Clear space and minimum size belong in the brand guidelines. In AI images, the safest logo is the real one on your real product photo, or one placed in a design tool after the render.
Where Overs fits
Visual DNA is one of the ten steps Overs runs before any photo is made, covering colors with hex codes, light, lens, grain and styling. Brand Watch reads a brand’s recent Instagram posts so the look stays current: weekly on Pro, daily on Team, and on request at most once a month otherwise.
Free for 40 photos a month. The AI that makes the photos is billed separately, on your own key, with no markup from Overs.