Technical access and indexing
I check what search engines and AI crawlers can read, along with canonical URLs, the sitemap, robots rules, and core structured data.
The website and its technical signalsAI visibility audit: find out what assistants actually discover about your business.
When a customer asks ChatGPT, Gemini, or another AI for help with a problem, your business may not appear at all—or the assistant may describe it inaccurately. The audit investigates why and shows you what is worth changing.
Discuss an AI visibility audit
I look separately at access, understanding, trust, and the final answer.
My experience spans data, websites, SEO and GEO checks, and practical testing of AI answers. That helps me separate indexing or access issues from problems with content, credible sources, and unclear positioning.
- Multiple layers
- crawling, indexing, sources, meaning, citations, and buying situations
- No algorithm guarantees
- recommendations backed by evidence of a plausible cause
AI visibility crosses marketing and technology.
The audit connects the customer’s buying situation, public evidence about the business, and the website’s technical accessibility. Each role needs a different answer.
- Owner / CEO I need to know whether AI understands why a customer would choose us.You see how your business appears in real buying questions, what supports the answer, and where the brand is unclear or missing.
- Marketing / sales I need to know which questions and sources deserve our content effort.You get gaps tied to specific customers, offers, and credible sources, rather than a generic GEO checklist.
- CTO / web team I need to separate technical blockers from gaps in content or reputation.I check crawling, indexing, canonical URLs, and structured data separately from business identity, offers, citations, and model answers.
Access, understanding, trust, and the answer customers actually see.
AI visibility is not a single status light. A crawler may reach your site without it being indexed. An indexed site may not establish a clear business identity. An assistant may understand the business but lack sources it can cite.
A clear business and offer
I assess whether public pages explain who you are, who you serve, what problem you solve, and why someone should trust you.
Business identity, services, and evidenceCredible public sources
I check which first-party and third-party sources confirm key facts about the business, its offer, and its experience—and where sources contradict one another.
Consistency and public evidenceAnswers to buying questions
I test a set of real questions, compare mentions, descriptions, and sources, and distinguish observations from hypotheses.
A test panel, rather than a single promptA set of priorities, without a catalog of every possible GEO tactic.
I separate technical blockers, unclear explanations, missing evidence, and dependencies on third-party sources. Each recommended step explains what it should change and how to check it later.
- A baseline
- Where and how your business appears for the agreed questions, and which sources the answers draw on.
- A diagnosis of the gaps
- Technical, content, and credibility barriers distinguished by the available evidence.
- An order for the fixes
- The smallest set of steps with the highest expected value, plus a way to measure progress afterward.
The audit starts with the questions customers ask when choosing a provider.
We start with the business context. Without it, the result would be a generic technical checklist disconnected from the customer.
- 01
Buying situations
Products, typical customers, competitors, and the questions that lead to choosing a provider.
- 02
Technical and content review
Your website, public sources, and a repeatable set of tests in AI assistants.
- 03
Priorities and measurement
Recommended changes, their order, and a way to track progress without false certainty.
A good fit if
- You know which customer questions you want to be relevant to.
- You can update the company’s website, content, or public sources.
- You see the audit as a baseline for deliberate improvement, rather than a ranking guarantee.
Not a good fit if
- You want a guaranteed mention or recommendation from a particular model.
- It is unclear who your business helps and with what.
- No one in web, marketing, or leadership will own the findings.
What the findings can actually tell you.
Model answers change. I distinguish repeatable findings from a snapshot that, on its own, proves very little.
What will we agree on before starting?
The goal and scope, the people involved, the materials or access needed, the deliverable, timing, price, and any follow-up work that would be a separate engagement. We start only when we have a shared understanding of all of these.
How will we handle nonpublic data?
Before working with nonpublic data, we will agree in writing on which sources may be used, where they may be processed, who has access, and whether external AI tools are allowed. I do not work with nonpublic data without that agreement.
Which AI assistants and questions does the audit cover?
We choose them based on your customers, countries, languages, and buying situations. The proposal will specify the test panel’s scope, measurement date, and how the tests will be repeated.
Does the audit include website and content fixes?
It includes a diagnosis, priorities, and a verification approach. We can agree on small fixes as a follow-up; larger content, development, or PR projects are not automatically included.
Can you guarantee that AI will start recommending us?
No. We can improve technical access, the clarity of the business, and the quality of public sources. We cannot control another company’s model rankings or recommendations, so the findings make the evidence and level of confidence clear.
Want to know what AI finds about you before your customers do?
Tell me which business, service, and buying situation you want to examine. I will propose test questions, an audit scope, and how you will receive the results.
Prefer email? jsem@radekduha.cz