Why Audience Intel

Four things this system cannot do.

Invent a number. Pass a simulated finding off as a quote. Approve its own work. Hide its worst reading.

Four refusals

Built into the code, where copy cannot reach.

Each one is a property of the code, so it stays true whoever writes the next page.

A simulated finding cannot carry a quote.

The type that holds modelled output has no field for a verbatim quote, a respondent count or a prevalence share — and it is sealed, so none can be added at runtime. Simulation and observation cannot be mistaken for one another here, because they cannot be written down the same way.

The reviewer has no “approved” state.

Published work passes an adversarial check whose first question is whether the claim’s subject is the thing the method actually measured. It never returns approved, and human sign-off is permanently required. We built it after publishing something that was false and had already passed four thousand seven hundred tests, a typecheck, a build and a human review.

No model ever produces a number.

Models classify; code computes. Every share, every interval, every score is calculated from those classifications rather than generated. A model that cannot emit a figure cannot invent one.

We publish our own worst reading.

We ran the method on our own firm and published what came back, including the parts that went against us. It is dated. The findings that have since closed are marked closed, and the one we cannot verify from our own records is marked, in as many words, not claimed.

Read our own report

The difference

The usual report, and this one.

The usual way

A report that reads well and cannot say where its numbers came from
Your site judged alone, with no competitor beside it
A list of fifty issues of equal weight
A rebuild recommended before anyone checks what is missing

With Audience Intel

Every finding labelled measured or inferred, with its source
Your site read beside the competitors you name
The first actions ranked by effort and by the buyers they help
The missing pieces named before any spend on a new site

Scope

What it reads, and what it does not.

The free report and the Snapshot, side by side, including what neither reads.

Scope of the free report and the Snapshot today.
Feature	Free report	AI Visibility Snapshot
Reads
Your website's pages	Included	Included
The competitors you name	Included	Up to three
Method
Fixed rubric, the same for every firm	Included	Included
Assistants used to evaluate	One	One (Claude), named
Live search rankings or review platforms	Not claimed	Not included
Numbers computed in code, not generated	Included	Included

FAQ

Questions people ask

Who is this for?

Owners and managers who make decisions about their market without a research team: professional firms, trades and online stores who want to know why buyers choose the firm beside them.

How is this different from an SEO audit?

An SEO audit asks how a search engine reads your site. This asks how buyers read it beside your competitors, and where each kind of buyer would stall. Speed and basic SEO appear where they affect that, not as the whole report.

Do you use AI?

Yes, to read pages and classify what they say. Models classify; code computes. No score, share or count in a report is written by a model.

Can I see a report before I use it?

Yes. The sample report on a constructed test firm is public, and so is the report we ran on ourselves, including the findings that went against us.

Judge it on your own site.

The free report needs no account and no card.

Run your free report