Brand1 Index · Brand consistency methodology
How the Index is made.
The Brand1 Index methodology scores public brand touchpoints from 0 to 100 against six weighted measures. Senior designers review at least 10% of scores and every serious failure. It is published before the results, so anyone can check our work.
The brand research methodology
6 steps, in order.
Open a step for one worked specimen example.
1ChooseAbout 100 leading international brands across sectors. Selection rules are being drafted now and will be published with method version 1 (planned March 2027), before collection begins.
Worked example · Specimen
Brand 037, a food and beverage company, enters under the draft rules. Its slot is fixed before any touchpoint is collected.
2CollectPublic touchpoints only (sites, social posts, ads, packaging images), within each source's terms. Target: about 5,000 in edition 1.
Worked example · Specimen
For Brand 037: its website, 30 social posts, 12 ads and 9 packaging images, each logged with its source and date of capture.
3Outside viewFor each brand, a “public brand profile” built from its own published material, labelled as an outside view.
Worked example · Specimen
Brand 037's profile records its logo versions, palette, typefaces and voice cues, taken from its own site and annual report.
4ScoreBrand Check's 6 measures and weights, where they fit the touchpoint. A serious failure caps a score at 69.
Worked example · Specimen
A print ad from Brand 037 scores 74, but the required AI label is missing. That is a serious failure, so the score is capped at 69.
5ReviewSenior designers review at least 10% of scores and every serious failure.
Worked example · Specimen
A senior designer confirms the failure against the original ad and signs the review, with a note on the record.
6PublishPatterns by sector and channel. Each brand is offered its scorecard.
Worked example · Specimen
Brand 037 appears only inside the food and beverage and print findings. Its own scorecard goes to the brand, privately.
Touchpoint scoring
6 measures, weighted.
Each touchpoint is scored from 0 to 100 against the brand's outside view.
- Bands
- Pass: 85 and above · Review: 70–84 · Fail: below 70
- Serious failures (score capped at 69)
- Logo misuse, such as a retired or rival mark · a missing legal line · a missing required AI label · any rule the brand itself marks as non-negotiable
The measures and weights come from Brand Check, Brand1's brand compliance checker. Read why the AI label matters in our guide to AI label rules.
| Measure | Weight | Checks |
|---|---|---|
| Logo use | 20 | Right version, clear space, minimum size, no distortion or recolouring |
| Colour | 15 | Palette within tolerance, share of brand colour, legibility contrast |
| Typography | 15 | Approved typefaces, hierarchy, sizes and spacing |
| Tone of voice | 20 | Copy against voice rules and examples; banned words |
| Required lines | 15 | Legal lines, disclaimers, regulated claims, required credits |
| AI labels | 15 | Synthetic-content label and metadata where required; Content Credentials present |
Sampling
About 100 brands. About 5,000 touchpoints.
Public touchpoints only: sites, social posts, ads and packaging images, collected within each source's terms. Senior designers review at least 10% of scores and every serious failure.
Channels
Edition 1 samples 6 channels. More in later editions.
- WebSites and landing pages
- SocialPosts and short video
- PackagingPack shots
- RetailShop, shelf and signage
- FilmAd films and frames
- PrintPress and outdoor
Timeline
Method first. Results after.
- NowMethod drafted, published here for comment.
- Planned · Mar 2027Method version 1 published, with selection rules.
- To be announcedCollection window: public touchpoints only.
- To be announcedScoring, and a person reviews every serious failure.
- To be announcedEdition 1: findings published, scorecards offered.
Limits
What the Index does not say.
- It measures consistency with a brand's own public identity, not value or preference.
- An outside view can miss a brand's internal rules.
- AI makes mistakes, which is why people review every serious failure.
- A score is a snapshot of the collection window.
Data and privacy
The Index runs in its own isolated workspace. It holds public touchpoints, not profiles of people. Data from licensed client work enters only in aggregate, and only with written opt-in.
The Index method, answered
How does the Brand1 Index score brands?
Each public touchpoint is scored from 0 to 100 against an outside view of the brand, built from its own published material. Six measures are weighted: logo use 20, colour 15, typography 15, tone of voice 20, required lines 15 and AI labels 15. Pass is 85 and above, review is 70 to 84, and fail is below 70.
Can brands pay to be included in the Index?
No. No brand can pay to be included in the Brand1 Index, to be left out, or to change its score. Brands are chosen under selection rules published before collection begins. A brand that disagrees with a finding can question it; we review it, reply in writing and log any correction publicly on this methodology page.
What counts as a serious failure?
A serious failure is a mistake that matters however good the rest of the work is: logo misuse such as a retired or rival mark, a missing legal line, a missing required AI label, or breaking any rule the brand itself marks as non-negotiable. Any serious failure caps the touchpoint's score at 69, and a person reviews every one.
How is AI used in the Brand1 Index?
AI does the first pass: it reads each touchpoint and scores it against the six measures. People make the calls that matter. Senior designers review at least 10% of all scores and every serious failure, and sign each review. The Index runs in its own isolated workspace and holds public touchpoints, not profiles of people.
How are brands chosen for the Index?
About 100 leading international brands are chosen across eight sectors, from banking and insurance to travel and hospitality. Selection rules will be published before collection begins, and each brand's slot is fixed before any of its touchpoints are collected, so results cannot shape who is included. The rules for edition 1 are still being drafted.
Corrections
If you think we've scored a touchpoint wrongly, tell us.
We review it, reply in writing, and log any correction on this page.
Corrections log
No corrections yet. Nothing has been scored.