So you can put it in front of whoever approves the £500 without having to describe it yourself.
£500 founding price · five working days · one ranked list · credited in full against your first month on a plan started within 30 days
Order today and it lands by
Most AI visibility work fails at the measurement, not the fixing. If the supplier picks the prompts, the supplier picks the score. So the prompt set is agreed with you first, in writing, and every number carries the date it was pulled.
| What | Detail |
|---|---|
| 15 buying questions | Agreed from your category, your service pages and your sales language before any work starts. Your company name appears in none of them. |
| Three runs per question | One AI answer is noise. Retrieval is probabilistic, so a single run proves nothing either way. |
| Six named surfaces | ChatGPT, Perplexity, Google AI Overviews, Google AI Mode, Gemini and Copilot. If your buyers use something else, we swap it in and say so before starting. |
| Three named competitors | Chosen by you, measured on exactly the same method as you, on the same dates. |
| The citation list | The pages the models actually cited when they named you, or named someone else instead. This is the part you act on. |
| A dated baseline | Every figure carries the tool, the mode and the pull date, so you can repeat the measurement yourself later. |
AI search is where it starts because that is where your buyers went. It is not the whole answer, and a supplier who only looks there is selling you half a picture at full price.
| Channel | What gets read |
|---|---|
| SEO | Indexation and crawl access including the AI crawlers, technical blockers, the overlap between who ranks and who gets cited, and the pages most likely to earn the answer. |
| Paid search | Google Ads and Microsoft Ads: account structure, search terms, conversion actions, spend concentration and tracking health. Read from account data where read-only access exists. |
| Conversion | Whether the page answers the query that was bought, visible friction, form failures and tracking gaps. |
Each one carries the evidence behind it, the commercial reason it sits where it does in the order, who inside your business should own it, and the next action.
An audit tells you everything. The useful version tells you what to do on Monday. That is why it stops at five: a list of forty findings is a way of transferring the hard decision back to you.
One PDF and one ranked list of the three to five fixes that matter most, each with the evidence behind it, the commercial reason it ranks where it does, who inside your business should own it, and the next action. Not a backlog. Anything over ten items is a report, not a decision.
Your URL, plus read-only access to GA4, Search Console, Google Ads and Microsoft Ads where those exist. Read-only is enough throughout. If an account does not exist or access cannot be granted, that section is marked unavailable rather than guessed at.
Access is read-only and used solely to produce your Diagnostic. We do not export your account data into any third-party tool beyond the analysis itself, and access can be revoked the moment the document lands. The full detail is in the privacy notice.
It is a fixed first read, not implementation. AI does the collection, a human reviews and ranks the findings, and the deliverable stops at the ranked list. That is the whole reason it can be sold at this price. It is also a founding price, and it rises as the published proof accumulates.
The guarantee is testable: at least three prioritised fixes, each with the evidence behind it and the commercial reason it ranks where it does. If the document does not contain that, you get the £500 back, and you do not have to argue the point.
The Diagnostic ends with the document. If the answer happens to be ongoing search management, that is sold separately at a published price with published limits, and the Diagnostic will say plainly if it is not worth your money.
£500 founding price. Five working days. The work and the price, upfront.
Request your Diagnostic