Why twelve dimensions rather than a single score?

A single number tells you how you are doing but not what to do. The dimensions separate distinct failures, so being unreadable to automated systems is never confused with being readable but unconvincing.

Is this just SEO with different labels?

No. Traditional SEO asks whether a page can rank. AEO and GEO ask how a business shows up in direct answers and generative search. Agent Readiness measures the conditions underneath all of them: whether systems can find the business, understand it, trust the information around it, and use it in an answer or recommendation.

Can I buy the Agent Readiness Audit on its own?

It is not sold separately. It is performed inside a Goldeneye Check: the Existing Presence scope for a business with a live site, and the Pre-Build scope when a new site or rebuild is being planned.

What do I actually receive?

A Composite Agent Readiness Score, a dimension by dimension breakdown, the findings behind each one, and a prioritised action plan. The criteria and weighting are proprietary. The findings and the score are yours.

What we measure

Can search and AI systems actually use what they find about your business?

What does Goldeneye actually measure?

A good website can still leave your business out of the answer. Agent Readiness is a structured diagnostic of the conditions underneath participation: twelve dimensions, evaluated against a fixed rubric applied identically to every business. They cover whether search and AI systems can find the business, understand what it is, encounter consistent and credible information about it, and use that information in an answer or a recommendation.

See Where You Stand Talk With Us

The situation that makes no sense

“Our website looks better than anyone else's around here.”
“We rank fine, but nobody is calling.”
“When I ask an AI assistant, it names our competitors.”
“Everything about us online is technically correct, and it still is not working.”

A business can have a well designed website and still be invisible to the systems customers now ask. Agent Readiness is a structured diagnostic of the conditions that can affect whether search and AI systems find a business, understand what it is, encounter sufficiently consistent and credible information about it, and use that information in an answer or a recommendation. Twelve dimensions, evaluated against a fixed rubric applied identically to every business. They are grouped here by the four questions those dimensions decide.

The model

Which question is your business failing?

Every one of the twelve dimensions exists because it decides one of four things. Start with the question, then open the dimensions that answer it.

Tap a question to see what decides itSelect a question to see what decides it

Can you be found?

Whether search engines and AI systems can reach the business, read it, and hold a record of it that they are willing to use in an answer.

Why a good website can still fail this

A site can look perfect to a visitor and still be unreadable to the systems that answer questions. If pages only assemble in a browser, or automated readers are blocked, or no verified record of the business exists, there is nothing for an answer to be built from.

When it fails

The business is not in the consideration set. It is not being rejected by customers, because it is never being shown to them.

What Goldeneye measures here

See Where You Stand

The evaluation tells you which of these four is actually holding the business back.

Where this shows up

Readiness explains why. Conversations show where.

Two different questions, answered by two different kinds of work. They meet at a finding.

Agent Readiness asks

Can search and AI systems reliably find, understand, trust and use the information about this business?

This is measured against the twelve dimensions, identically for every business. It explains why a business may struggle.

Real conversations ask

What is the customer trying to decide, who gets named in the answer, and what evidence appears to sit behind it?

This is Conversational Intent Intelligence, or CII. It is observed in the decisions your customers are actually making. One decision can appear through many natural conversational variations, so a single prompt is never treated as a benchmark. The conversation shows where you may be getting left out of the decisions your customers are making.

A worked example

What that looks like on one restaurant

Example business, example situation. No real measurements are shown.

Observed in conversations: The restaurant is rarely named when people ask about romantic waterfront dining, although it appears regularly for seafood generally.
What the readiness evidence shows: The site never clearly establishes waterfront dining, occasion suitability is barely stated, independent corroboration of the setting is limited, and reviews describe the food far more often than the atmosphere or the view.
Dimensions involved: Goal-State Content, Citation Authority, Review Signal Strength, Trust Architecture
Priority: Improve, because the decision is high value and the gap is a stated attribute rather than a structural rebuild.
Action: Strengthen the sources and signals that accurately establish the setting and the occasion, then re-evaluate that decision once the changes have had time to be discovered.

The dimension score tells you what is weak. The conversation shows where you may be getting left out of the decisions your customers are making.

A traveller at a Southwest Florida waterfront cafe reading her phone while deciding where to go

Customers now ask a machine the questions they used to ask you.

The twelve dimensions in full

One model, one rubric, applied identically to every business measured. The criteria and weighting are proprietary. The findings and the score are yours.

01 Schema Completeness
Quality, coverage, validation, nesting, and implementation of structured data.
02 Entity Definition
Unambiguous brand identity, descriptions, sameAs relationships, and attributes.
03 Knowledge Panel
Existence, verification, accuracy, completeness, and claimed status.
04 Citation Authority
Press, reference sites, directories, backlink context, and descriptive depth.
05 Brand Consistency
Naming, description, category, contact, and location consistency across the web.
06 AI Assistant Visibility
Whether the brand surfaces accurately and competitively in goal-state AI queries.
07 Goal-State Content
Whether content matches real user goals across the journey in conversational, structured form.
08 Review Signal Strength
Volume, recency, sentiment, response behavior, and platform spread.
09 Topical Authority
Cluster coverage, content depth, freshness, internal linking, and external citation.
10 AI Crawler Access
Crawler access, rendering, crawl efficiency, and sitemap/indexability.
11 Transactability
Machine-accessible services, booking/action clarity, pricing data, integrations, and conversion friction.
12 Trust Architecture
Author bios, E-E-A-T signals, credentials, press/media proof, security/compliance signals.

Reading the score

What does the score actually tell you?

It is a starting position, not a grade. A baseline and trajectory measure, not a pass or fail badge. The Composite Agent Readiness Score sits in one of five bands, and a low band is an opportunity set rather than a failure. The value is in the movement between measurements.

Critical
0–39
At Risk
40–59
Developing
60–74
Strong
75–89
Leading
90–100

Critical

Baseline with major structural and signal gaps. The emphasis is prioritized opportunity, not failure.

At Risk

Foundation exists; important signals remain missing or inconsistent.

Developing

Solid base; focus on the specific gaps that improve recommendation consistency.

Strong

Well positioned; focus shifts to leadership, monitoring, and adaptation.

Leading

Exceptional readiness; maintain, monitor, and protect the position.

What happens while it is running

Status is visible throughout, so there is never a question about where the evaluation stands.

Intake requiredInformation or access is still needed.
Ready for auditRequired intake is complete.
In analysisGEI is conducting the audit.
Quality reviewScores and report are being reviewed.
Ready for debriefReport is available and the session can be completed.
CompleteBaseline delivered; recommended next path available.

What is observed, and what is not

The answers search and AI systems give
Citations and sources where they are shown
Recurring recommendation patterns across a decision
The evidence that exists about your business, and where it disagrees
Which competitors participate in which decisions
How all of that changes over time

Goldeneye measures the environment around the decision. It does not pretend to see inside the model making it. Findings distinguish between what is observed, what is supported by evidence, and what is likely contributing. Nothing is presented as proven cause.

Where this sits among SEO, AEO and GEO

These disciplines overlap, but they are not the same question. A business can be strong in one and weak in another.

SEO: Whether your pages and business can earn visibility in traditional search results.
Technical SEO: Whether the content that matters can be crawled, rendered, interpreted and indexed reliably.
Local SEO: Whether the business is accurately understood and surfaced for relevant locations and local intent.
AEO: Answer Engine Optimization. Whether your content and business information are structured, supported and usable in direct-answer environments such as Google AI Overviews and AI assistants.
GEO: Generative Engine Optimization. How clearly a business is understood and represented in generative search, including whether it is cited, included, or considered when people ask for recommendations.
Agent Readiness: Goldeneye's measurement layer underneath all of them: whether systems can find the business, understand it, trust the information around it, and use it in an answer or recommendation.