That may look impressive, but a high score says little if the brand is missing from the questions and decision points that matter to its audience.
Before looking at any dashboard, a few more important questions need to be answered. Which audiences matter most? At which stages of their decision-making should the brand appear? And what actually needs to improve: the number of mentions, visibility among a specific audience or a stronger association with a key differentiator? Those choices are not made by the tool. They come first.
Like many organisations and agencies, we are still getting to grips with GEO. The field is developing quickly and there is still a lot we do not know. That is why we are testing different tools to understand what they really measure, which insights are useful and how those insights can lead to concrete improvements. One thing is already clear: a good GEO analysis does not start with a dashboard. It starts with the questions for which a brand wants to show up for. Those questions form the basis for the benchmark and help determine which KPIs and actions matter.
What does a GEO tool actually measure?
A GEO tool does not measure a brand’s overall visibility in AI search. It analyses answers to a predefined set of questions. The choice and wording of those questions therefore have a major impact on the results. Take a question such as: ‘What is the best cybersecurity solution?’ The answer will probably feature large, well-known providers. Add more context and the result may change: ‘What is the best security solution for a mid-sized healthcare organisation with a small IT team, strict requirements around European data storage and limited capacity for a complex implementation?’
Both questions are about cybersecurity, but they measure something different. The first is mainly about general awareness within the category. The second explores which provider AI considers a good fit for a specific organisation and situation. A brand does not need to come first for every question. It needs to show up when its strengths are relevant.
Start with the right questions
Choosing the right questions may sound like a technical part of the analysis, but it depends largely on market knowledge. What problems is the audience trying to solve? What concerns do they have? Which requirements ultimately shape the decision?
A tool can suggest questions, but it does not automatically understand all the nuances. It does not know which audiences matter most commercially or why customers choose one brand over another. That insight still comes from customer interviews, sales conversations, customer support queries, search behaviour, market research and internal experts.
A brand may appear during the early stages of research, but disappear once someone starts adding specific requirements or comparing providers. A useful analysis should therefore include questions from different stages of the decision-making process. For international brands, language and location add another layer. Search behaviour, terminology and local context vary from market to market, so content that works well in one country may be less relevant or visible in another. Grouping questions by audience, problem and decision stage creates a clearer framework and shows where a brand appears and where it drops out.
Turn the benchmark into action
With the right questions in place, a GEO tool can provide a useful benchmark. The results show not only how often a brand appears, but also for which questions, alongside which competitors and for what reasons. AI may recognise a brand as a provider but fail to recommend it to an important target audience. A distinctive product feature may be missing from the answer. Or AI may simply be relying on outdated information.
Those findings give organisations something concrete to work with. Instead of aiming to become ‘more visible’ in general, they can focus on visibility among a specific audience, a stronger association with a distinctive feature or a more accurate description of the brand.
It also matters whether AI simply mentions a brand, uses it as a source or actively recommends it. That makes GEO KPIs more meaningful than counting mentions alone.
Build authority
Before trying to improve a score, it is important to understand what is driving it. A low score does not automatically mean an organisation needs to publish more content. The right information may already be online but be too generic, unclear or difficult to find. The way information is presented matters too. Concrete answers, up-to-date figures, clear definitions and verifiable claims make content more useful as a source.
What a brand says about its own expertise also carries more weight when other sources support it. Trade media coverage, customer experiences, independent research and expert commentary can all help people and AI recognise the brand as a credible source on a particular topic. The analysis can then point to the right action. Sometimes a product page needs clearer information. In other cases, a customer story may help, or new research or PR around a topic on which the brand can credibly claim expertise. One improvement is not the end of the process. AI answers change as new information appears, different sources are picked up and models are updated. GEO therefore requires regular measurement, analysis and adjustment.
Keep measuring and adjusting
Repeated measurements using the same relevant questions make it possible to track changes over time. Is AI mentioning the brand more often when the right questions are asked? Is the description becoming more accurate? Are the right strengths showing up more clearly? By comparing those results with changes in content and PR activity, organisations can see whether the desired positioning is becoming stronger. The impact of a single article or PR activity cannot be isolated, but patterns over time can show where progress is being made and where further adjustment is needed. That turns GEO into more than a score in a slick dashboard.
The goal is not to be mentioned as often as possible. It is to appear at the right moment, for the right reasons and in front of the right audience.
Petrick de Koning
Petrick de Koning is a senior content creator. He helps technology-driven organisations tell complex stories in a clear, human and credible way. With a sharp eye on emerging developments such as AI and GEO, he explores how technology can help make content more relevant, without losing the human perspective.