How AI
shops for Britain

We ask ChatGPT, Claude and Gemini the questions real UK consumers ask — every week, across 67 categories — and record exactly which brands they recommend and which sources they trust. This is what the answers say.

0 AI answers · 0 brands tracked · measured, never estimated

Used Cars is a monarchy. Consumer Electronics & Appliances is a bazaar.

The share of brand mentions captured by each category's top three. In used cars, three brands take 100% of everything AI says; in consumer electronics & appliances, 27 brands split the answers and the top three manage just 21%.

Every category has a king — and 37 thrones changed hands.

The most-named brand in each category's answers. In 37 of 67 categories the AI answer leader is not the market leader we seeded — Motors outranks Auto Trader in used cars.

Some categories AI recommends on love. Others on habit.

Every answer describes a brand in three ways: how much it stands out, how much it belongs, and how much it is returned to. The balance is not the same everywhere. Watches & Jewellery leans hardest on attraction (35% of its desire mix); Price Comparison Sites hardest on attachment (37%).

Shares are of each category's own desire and sum to 100, which is what makes categories comparable. See the full AI Desire Index →

Britain has a citation establishment.

reddit.com and youtube.com are cited in 61 of 67 categories — from credit cards to skincare. A small set of trusted referees decides what AI believes, whatever the topic. Each green cell is one category where the domain is cited.

AI reads your website. It cites it 26% of the time.

Across every category, Claude and ChatGPT opened 5,564 brand-owned pages while answering — and put just 1,435 of them in their cited sources. Household Cleaning Products brands convert best (57%); fragrance & beauty worst (0%).

Beauty behaves like one market. Health behaves like 6.

Every category here is really several — Finance is credit cards and mortgages and pensions. Alikeness scores how tightly a category's sub-categories cluster on concentration, leader visibility, model agreement and read→cited, against the spread across the whole programme: 100 is indistinguishable, 0 is as far apart as any two categories picked at random. In 1 of 20 categories one brand leads more than one sub-category — Bupa holds 2 of Health's 6 crowns.

In 23 categories, every model names a different winner.

Consensus is the exception: the models agree on the leading brand in only 14 of 67 categories, and disagree most in pensions. A brand that owns one assistant can be invisible in another — which makes "our AI visibility" a question with three different answers.

There is no single formula. Every category rewards something different.

Percentage-point difference in how often a page gets cited, with a feature versus without, within the pages models actually read. Structured data travels furthest — worth +17pp on average — but read across each row and the picture changes category by category.

Some categories run on guides. Others run on the crowd.

What AI actually cites in each category, by kind of page. The mix is the brief: it tells you what to publish to be quotable here — and it is not the same twice.