What we do Pricing FAQs Contact Get started

The Search and AI Audit

Exactly where you stand: how you rank on Google, and what ChatGPT, Gemini, Perplexity, Claude and Copilot say when a customer asks for a business like yours. It is the first thing we do when you start, with evidence and a plan.

In short

You cannot improve what you cannot see. The Search and AI Audit measures both halves of modern search at once: your position on Google, and what six AI engines say about you when real customers ask. We capture the evidence and date every answer, check whether the facts are even right, and hand you a prioritised plan in plain English. It is the first thing we do when you start the plan: one flat price, no lock-in, cancel any time.

Most business owners have never seen what an AI assistant actually says about them. They can picture their Google ranking, roughly. But when a customer opens ChatGPT and types "who is the best in my area for this", a specific answer appears, names get mentioned, and someone gets the enquiry. If that someone is not you, you usually never find out. There is no page two to check, and no analytics line that reads "lost to a competitor inside an AI answer".

The audit exists to end that blind spot. It is the honest starting point for everything else we do: the first thing you get when you join the plan, so you know exactly where you stand before anything else changes.

What a Search and AI Audit actually is

It is a structured, evidence-based read of how discoverable your business is across the whole of modern search, delivered as a report and a call rather than a login to yet another dashboard. Half of it is familiar: where you rank on Google for the terms that bring you customers, and whether your site is technically healthy enough to compete. The other half is newer and rarer: what AI assistants say, whether they say it accurately, and why. We set out what this kind of audit includes and why it matters in our guide to AI visibility audits.

We treat both halves as one picture on purpose. Ranking and being recommended are related but not the same, and looking at only one gives you a false sense of safety. An Ahrefs study in 2025 found that only about 12% of the pages cited in AI answers also rank in Google's top ten for the same query, so a strong Google position tells you surprisingly little about whether AI will name you (Ahrefs, 2025).

What we measure

Your position on Google and classic search

We map where you currently rank for the searches that actually lead to enquiries, not vanity terms. We check that your important pages are indexed, that the site is crawlable and quick, and that the basics that still decide rankings are in order. This is the foundation, because AI systems learn about you from the same web that Google indexes, so a broken site holds back both halves at once.

What AI actually says about you

We ask ChatGPT, Gemini, Perplexity, Claude, Copilot and Google's AI the questions your customers ask, in the words they use. Then we do it again, several times, because these systems are probabilistic and a single run is not evidence. We record whether you are named, how you are described, who gets recommended ahead of you, and how consistent the answers are across engines and across repeat runs.

Whether machines can read you

We check the signals that decide whether an engine can confidently extract and repeat information about you: structured data, a clean page structure, a sensible robots and llms setup, and whether AI crawlers are allowed in at all. Plenty of sites are quietly invisible to AI because a default setting blocks the crawler, and they have no idea.

Which sources are shaping your answers

When an assistant recommends a business, it leans on sources it trusts: directories, review platforms, industry publications and reference sites. We trace which sources are feeding the answers in your category, and where you are present, absent or out of date in them. That map is what later turns into targeted citation work.

Whether the facts are even right

AI systems confidently state things that are wrong. We flag every place an engine gets your prices, services, location, opening hours or specialism incorrect, because a wrong answer given with confidence can cost you a customer just as fast as no answer at all.

How we run it, and why it is trustworthy

The method is deliberately boring, because boring is what makes it honest. Every buyer question is run multiple times across every engine, so we report the pattern rather than a lucky or unlucky single result. Every claim in the report is backed by a dated record of the answer, so you are reading evidence, not our opinion. And where the picture is mixed, we say so, rather than rounding up to a tidy story.

We write the whole thing in plain English and then talk you through it on a call. You should finish the call understanding exactly where you stand and what would move the needle first, whether or not you ever work with us again.

Only 12%

of pages cited in AI answers also rank in Google's top ten for the same query (Ahrefs, 2025). Ranking well and being recommended by AI are two different results, which is exactly why we measure both.

What you get

The deliverable is a report you own outright, plus a call to walk through it, covered by the same flat monthly price as everything else we do. No separate fee, no quote to chase.

  • Your Google ranking position for the searches that bring customers
  • A visibility read across six AI engines, with repeat-run sampling
  • A dated record of every answer, so it is evidence not opinion
  • A technical and machine-readability check of your site
  • A map of the sources shaping AI answers in your category
  • An accuracy check flagging every wrong fact about you
  • A prioritised action plan, most important fixes first
  • A call to talk it through, in plain English

Honest limitations

An audit is a snapshot, not a promise. AI answers shift over time and vary by phrasing, user and day, which is precisely why we sample repeatedly rather than checking once and why we recommend re-measuring monthly if you act on the findings. We will never guarantee a ranking or a place in any AI answer, because nobody can control what these systems say. What we can do is show you the truth today, and the clearest path to improving it. From the second month, we implement that path and re-measure the same picture, covered in the monthly report.

Questions people ask

How long does the audit take?

Usually five to seven working days from the moment we have what we need. We run each buyer question several times across every engine to average out the variation, capture the evidence, then write it up in plain English. You get a scheduled call to walk through it, not just a file.

Do you need access to my website for the audit?

Not for the audit itself. It measures what is already public: your Google rankings, what AI assistants say, your structured data and your citations. Once you are on the plan, we ask for access by a single secure key so we can start implementing the fixes it finds.

Is this just an SEO audit with a new name?

No. A classic SEO audit looks at Google rankings and technical health, and we do that too. This audit adds a second half that most do not: what six AI engines actually say when a customer asks, whether the facts are right, and which sources are shaping those answers. You see both pictures side by side.

What happens after the audit?

You own the report. From the second month we implement the fixes on your site, build on them and re-measure your position every month, covered by the same flat price. Cancel any time, no lock-in.

Sources

Every statistic here is attributed inline to the source below. We link to primary material so you can check each claim yourself. If a number is not attributed, treat it as our professional judgement rather than a measured fact.

See where you actually stand

Start with the audit. The full picture across Google and AI in the first month, one flat price, no lock-in.

Get started