AI search visibility audit
Your customers don’t just Google anymore.
They ask ChatGPT, Claude, Gemini, Perplexity, Copilot and Google’s AI who to choose. We find out what those answers say about you, and what to do about it.
Evidence-led · every AI answer on record · reviewed by a human
| Customer search | |||||||
|---|---|---|---|---|---|---|---|
| best builders near Leeds | 5/5 | 2/5 | 0/5 | 4/5 | 3/5 | 0/5 | 5/5 |
| loft conversion, Manchester | 2/5 | 0/5 | 0/5 | 5/5 | 0/5 | 1/5 | 3/5 |
| who should I use for a kitchen extension? | 0/5 | 0/5 | 1/5 | 2/5 | 0/5 | 0/5 | 0/5 |
Tested with live web search on, every audit:
ChatGPT
Claude
Gemini
Perplexity
Copilot
Google AI Overviews
Google AI Mode
The shift
The click is disappearing. The answer is the new shop window.
AI assistants don’t show ten links. They give one answer that names two or three businesses, and most people take it.
“Who’s the best builder near me for a loft conversion?”
Discovery
When people don’t know your name yet, does AI put you on the shortlist, or a competitor?
“Is Example Builders any good?”
Reputation
When they do know your name, does AI recommend you, hedge, confuse you with someone else, or steer them to a rival?
“Why doesn’t ChatGPT know us?”
The cause
Blocked AI crawlers, missing review profiles, thin proof on your site. We find the reasons, not just the symptoms.
How it works
From your questions to a clear plan.
A managed service: we run the audit, read every answer and hand you the fixes.
Brief
We map the questions your customers really ask, by service, by area and by your name, plus the rivals you care about.
Run
Every question goes to 7 AI engines, several times each, searched from the towns your customers are in.
Review
We score visibility, brand trust and share of voice, check your website’s AI-readiness, and a human reviews every finding.
Act
You get a live report and a ranked plan. We can deliver the fixes and re-audit to prove what moved.
What you get
One live report. Every answer on the record.
Private, shareable, and built to be acted on, not filed away.
Where you appear, engine by engine and question by question
A matrix of every customer search across ChatGPT, Claude, Gemini, Perplexity, Copilot and Google’s AI Overview, map pack, organic results and AI Mode. Click any cell to read the real answers.
What AI says when people ask about you
Recommends you, hedges, doesn’t know you, or points to a rival, plus the proof it couldn’t find.
You versus every competitor AI names
What share of answers name each business, per engine, and which searches each one owns.
The sites AI trusts in your category
Which review sites and directories the engines cite, and whether you’re on them.
Can AI even read your website?
Crawler access, firewall blocks, structured data and llms.txt, with a fix list.
Critical gaps, quick wins and strategic wins
Every recommendation tagged with the area, the effort (an hour, a day, a few weeks) and what it moves, so you know exactly where to start.
Method
Measured properly, not guessed.
Live answers, not model memory
Every engine runs with live web search on: what a real customer sees today, not what a model learned last year.
Asked more than once
AI answers vary. Each question runs several times per engine, so you get a reliable rate instead of a lucky snapshot.
Searched from where your customers are
A question about Barnet is searched from Barnet; “near me” from your own area.
Nothing paraphrased
Every raw answer is stored and linked from the report. Every finding traces back to the evidence.
Reviewed by people
AI drafts, humans decide. Nothing reaches you without a KINDRED review.
FAQ
Good questions.
What is an AI visibility audit?
It's a test of what AI assistants tell your potential customers. We ask ChatGPT, Claude, Gemini, Perplexity, Copilot and Google's AI Overviews and AI Mode the questions your customers really ask, then show whether you're named, how you're described, who's recommended instead, and exactly what to fix.
How is this different from SEO?
SEO gets you into a list of links. AI engines skip the list and give one answer, often naming two or three businesses. Ranking well on Google doesn't guarantee an AI names you, so we measure the AI answers directly, alongside Google's AI Overview, map pack and organic results.
Which AI engines do you test?
ChatGPT, Claude, Gemini, Perplexity and Microsoft Copilot, plus Google's AI Overviews, local map pack, organic results and AI Mode. Every one runs with live web search on, the way people use them today.
Why do you ask every question more than once?
AI answers change from one ask to the next. We run each question several times on every engine and report a rate, such as 'named in 4 of 5 answers', so you see how reliably you show up, not one lucky or unlucky snapshot.
What do I get?
A private, live report with your discovery score and brand trust score, every raw AI answer, share of voice against your competitors, the review sites AI relies on in your category, an AI-crawlability check of your website, and a ranked plan: critical gaps, quick wins and strategic wins.
Can you fix what the audit finds?
Yes. KINDRED is a creative growth studio, so we can implement the fixes, from structured data and content to reviews and PR, and re-audit at 30, 60 or 90 days to show what moved.
Is it only for London businesses?
No. We're London-based and every search can be made from the exact towns and areas your customers are in, anywhere in the UK.
Find out what AI tells your customers.
Start with an audit. We’ll show you where you stand and what to fix first.