The study

The AI Discovery Study.

An AI Discovery Study records what five AI tools say when your buyers ask who to hire. It shows who they recommend instead of you, what they relied on, and what to fix first.

Scope
10 buyer situations, 4 questions each
Tools
ChatGPT, Gemini, Claude, Perplexity, Google AI Overviews
Price
From US$250, paid upfront

The short version

  • What it is. A recorded study of what five AI tools tell your buyers when they ask who to hire.
  • What it covers. 10 buyer situations, each asked four ways, across ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews. About 200 answers.
  • What you get. A private study site, a PDF, and the fixes ranked by what to change first.
  • What it costs. From US$250, paid upfront once you approve the scope.

What the study answers

Buyers now ask an AI tool for a shortlist before they visit a website or call anyone. The study records those answers and reads every one, so you can see three things.

  • Are you recommended? For which buyer situations, by which tools, and how often.
  • Who is recommended instead? The firms that appear in your place, and the pages that earned them the mention.
  • What is the answer built on? The sources each tool leaned on: your own site, a directory profile, a review page or a competitor's pricing page.

What you receive

PartWhat it contains
Private study siteEvery question we asked and what every tool said, so you can check any finding yourself.
ShortlistsWho each tool named for each buyer situation, and how often.
Source trailsThe pages each answer relied on, traced one by one.
Competitor pagesThe pages that got competitors recommended, and what they say that yours don't.
Ranked fixesWhat to change first, second and third, with the evidence behind each.
PDF and walkthroughA PDF of the findings, and a walkthrough in chat.

How it runs

  1. Intake. You send your website, who you sell to and two competitors.
  2. Situations. We agree 10 buyer situations with you, with four questions for each.
  3. Recording. We ask ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews, and trace every source.
  4. The file. You get the study site, the PDF and the ranked fixes.

The scope is fixed and agreed before we record anything. Read the full method, including how we count and what a study can't tell you.

What studies have found

FirmWhat the answers showed
UK payroll recruitment firmRecommended in 27 of 32 answers for an everyday hire, and in 4 of 32 for its highest-fee search.
US leadership training firmNamed once in 160 answers. Its closest competitor was named 63 times.
US commercial contractorRecommended in 15 of 50 answers, always under its parent company's name and never under its own brand.

All three are anonymised. See the case files.

Who it's for

Service businesses whose buyers ask around before they hire: recruiters, training firms, contractors, consultancies, agencies and other firms that win work through a shortlist.

It isn't for anyone who wants a guaranteed position in an AI answer. Nobody can sell that honestly.

What it costs

A study starts at US$250, paid upfront once you approve the scope. Larger scopes are quoted before anything starts. Monthly tracking is US$400 to US$600 a month. See full pricing and terms.

After the study

  • Implementation. We close the gaps the study finds, on your website and beyond it: articles, LinkedIn and Reddit content, video, directories, reviews and press. See every way we get you cited.
  • Monthly tracking. Your agreed questions tracked three times a month, with two runs per AI tool, so you see trends and not one-off answers.
  • Market research. If the study raises a bigger question about your market, we research that too.

What we won't promise

  • A ranking. AI answers don't have one.
  • Guaranteed placement in any AI tool.
  • That money buys recommendations. OpenAI says paying doesn't change them.

Guides by industry

More in insights by industry.

Questions

How is this different from an SEO audit?

An SEO audit checks how your pages rank in Google. A study records what AI tools actually say when your buyers ask for help, who they recommend instead of you, and which sources they relied on. The fixes follow from that evidence.

AI gives a different answer every time. Is this reliable?

Single answers vary. That's why we ask each buyer situation four ways across five AI tools, about 20 answers per situation. Patterns that hold across that many answers are the ones worth acting on. With monthly tracking, every question runs twice per tool, three times a month, so you see trends rather than one-off answers.

What do you need from me?

Your website, who you sell to, two or three competitors and the markets you care about. We agree the 10 buyer situations with you before we record anything.

Do you make the fixes too?

If you want us to. Implementation goes well beyond page rewrites: decision pages, articles, LinkedIn and Reddit content, video, directory and review profiles, and press, wherever the study shows evidence is missing. It's quoted per scope. Many clients make the first fixes themselves.

Does Businesses in ChatGPT replace this?

No. OpenAI's dashboard counts a mention only when ChatGPT cites your website, and it covers ChatGPT alone. In our research, ChatGPT used the firm's own site in about a third of its recommendations. A study covers five AI tools, including Google AI Overviews, and shows the competitors and sources the dashboard doesn't.

Next step

See why AI recommends your competitors.

Tell us who your buyers are. We reply by email within one working day with the scope and the price.