On Tuesday night, out of curiosity, your CEO asks ChatGPT to recommend the best providers in your market.
On Wednesday morning you get an email asking why your company isn’t showing up in AI search results.
What do you tell them?
The truth: a single response from one AI platform is not a meaningful measure of any organisation’s AI visibility.
So, why does one search proves so little?
On this page:
AI answers change
Unlike conventional search results, an AI response is generated afresh and may change when the same question is repeated.
A recent study of commercial recommendations from ChatGPT, Gemini and Google AI Overviews found that recommended products often changed across repeated requests. The researchers concluded that isolated responses should not be treated as representative of what users encounter.
Your CEO could repeat the same search and receive a different answer.
Small changes to the prompt matter
“Who are the best providers?” is very different from asking for providers with experience in a particular sector, location, use case or technical environment.
Research comparing Google Search, Gemini and Google AI Overviews found that generative results were less consistent across repeated searches and sensitive to minor changes in the query.
A single broad prompt is unlikely to represent the detailed questions asked by buyers with a genuine requirement.
Platforms do not agree
Visibility in ChatGPT does not guarantee visibility in Gemini, Perplexity, Copilot or Google AI Overviews.
The commercial recommendations study found that the ChatGPT and Gemini interfaces shared only 5.4% of displayed source domains for equivalent queries. In more than three-quarters of comparisons, they had no source domain in common.
Testing one platform provides evidence about one response, not the wider buying journey.
What should Marketing measure instead?
A credible AI Search Optimisation reporting approach should start with a structured library of commercially relevant prompts.
The library should cover how buyers define their problem, explore possible solutions, identify and compare suppliers, and validate expertise before making contact.
These prompts should be tested across the AI platforms your buyers are most likely to use and repeated over time.
Reporting should then show:
- how frequently your organisation appears
- which competitors appear alongside it
- how prominently and accurately it is represented
- which websites and sources influence the answers
- where visibility is improving or declining
This is broadly consistent with the IAB’s AI visibility measurement framework, which distinguishes between a brand’s presence, prominence, portrayal and ability to influence action.
Our B2B Marketer’s AI Search Action Plan explores why realistic prompts and buying-journey analysis are essential to understanding AI visibility.
Turn that one search into the right conversation
Your CEO’s search cannot establish whether there is a persistent visibility problem. But it can open an important conversation about how the organisation approaches AI search.
The question is not: “Did we appear in this one answer?”
It is: “Do we have a strategy and reporting approach to ensure we are consistently visible, accurately represented and credibly sourced throughout the AI-assisted buying journeys that matter to our business?”
How about booking that conversation? Book a free 30-minute consultation with me to talk about your AI discoverability.
On this page:
Subscribe
Receive our biweekly newsletter and stay up to date with the latest B2B digital marketing news and insights.