Methodology

The Recommendation
Readiness Hypothesis

Recommendation Readiness measures how easy it is for an AI assistant to justify recommending your business.

Not find. Not rank. Not crawl. Justify. That changes everything, and the whole product is built around it.

The central hypothesis

AI assistants don't pick the best company. They pick the one they can most confidently recommend.

When an AI assistant recommends a supplier, it isn't choosing the “best” company. It chooses the company it can build the most confident, evidence-backed recommendation for.

The product doesn't measure quality. It measures recommendability.

Definition

Recommendation Readiness is the probability that an AI assistant can confidently recommend your company when a buyer asks a relevant question.

The extended hypothesis

Every AI recommendation comes from three things working together.

Discoverability

Can the AI find evidence about you?

If not, you don't exist.

Answerability

Can the AI answer a buyer's questions from the evidence it found?

If not, it lacks confidence.

Verifiability

Can the AI justify recommending you?

If not, it recommends someone else.

Every score in your audit rolls up to these three principles.

The reasoning chain

Every recommendation follows the same path.

Each score in the report shows where this chain breaks for your company.

1
Buyer asks a question
2
AI retrieves evidence
3
AI evaluates confidence
4
AI constructs an answer
5
Company is recommended (or not)
Why companies disappear

Companies don't disappear because they're bad.

They disappear because one of four things failed. This is the diagnostic behind every recommendation we make.

1
Evidence doesn't exist
2
Evidence exists but isn't public
3
Evidence is public but isn't discoverable
4
Evidence is discoverable but isn't convincing
The evidence model

Evidence has four dimensions.

Exists

Did we find it?

Accessible

Can AI retrieve it?

Relevant

Does it answer buyer questions?

Trustworthy

Would an AI rely on it?

Every piece of evidence in your audit is scored on these four dimensions. Any recommendation you receive can be traced back to specific evidence that passed or failed on one of them.

The methodology

Every audit tests four assumptions.

01

A buyer asks a realistic question.

02

The AI retrieves publicly available evidence.

03

The AI synthesises an answer using that evidence.

04

The company recommended is the one supported by the strongest accessible evidence.

The full model

From truth to recommendation.

Break any link and the recommendation weakens. The audit shows which link failed and how many Recommendation Readiness points you'd gain by fixing it.

Truth
Evidence
Discovery
Understanding
Confidence
Recommendation
The formal hypothesis
The likelihood of an AI recommending a company is proportional to the quantity, quality, accessibility and consistency of independently verifiable evidence available to answer buyer intent.

That's a testable hypothesis. Our software is the apparatus for testing it, one company and one buyer question at a time.