Skip to main content
Methodology

Limits of interpretation

We document response variability, model changes, location constraints, causal uncertainty, and the limits of available data.

Read the limitations See how confidence is reported
Answers can vary Model changes outside our control Geographic variance
Visualization for the page: Limits of interpretation
Definition and protocol

Answers can vary

The same model can answer an identical question differently. This is a normal property of generative systems, not necessarily a measurement error.

We reduce its impact through repeated queries and stability analysis, but we do not pretend the variation can be removed entirely.

Methodology rule

Model changes outside our control

Providers can change a model or API behaviour without notice. Methodology and collector versions are stored, but model identity is not currently complete for every answer.

A score jump may therefore come from the environment. Review it with date, provider logs, and full answers instead of automatically attributing it to brand actions.

Methodology rule

Geographic variance

AI answers vary between locations, and a measurement from one location does not perfectly reflect the experience of every user in a country. We define the measurement location explicitly, but it is a sample of the environment, not its complete picture. Results should not be generalised to locations we did not measure.

Methodology rule

Source volatility

The ecosystem of sources AI draws on changes constantly: content disappears, rankings reshuffle, services change access rules. The source picture in a report is current as of the measurement and can go stale faster than the score itself. That is why we treat source analysis as directional guidance, not a permanent map.

Methodology rule

No automatic attribution

Run comparison shows correlation over time: answers changed after a period of work. Models, sources, competitors, and data scope may have changed at the same time.

GeoFolks does not currently calculate the causal impact of an individual task. Such a conclusion is an analytical judgement, not a platform feature.

Methodology rule

A visibility score is not revenue

The GEO Visibility Score measures the visibility and presentation of a brand in AI answers — not sales, traffic, or conversions. The link between AI visibility and business outcomes depends on the category, the market, and many factors outside the measurement. We do not present a score increase as a revenue forecast and advise against that interpretation.

Methodology rule

No guaranteed AI recommendation

No one — not us, not any service provider — controls what AI models answer to users. GEO Readiness recommendations improve the measurable preparedness of a brand, but do not guarantee that AI will start recommending it. Any offer of a "guaranteed position in AI answers" should be treated with great caution — including ours, if it ever appeared.

Methodology rule

Privacy restrictions

We do not measure private user conversations with AI — measurement relies on controlled queries executed under the protocol. We therefore do not know how often real users ask about a category, nor exactly what they see in their own sessions. Results describe the behaviour of AI environments under defined conditions, not actual user traffic.

Continue exploring

Related GeoFolks areas

Compare plans
Methodology A traceable GEO methodology

The score uses 12 components, versioned weights, and source-quality labels. Some signals are direct, while others are proxy-based or declared.
Methodology Measurement confidence

Measurement confidence combines coverage, source fidelity, repeatability, freshness, and validation. These are technical indicators, not guarantees that an answer is true.
Methodology Providers and measurement channels

GeoFolks uses configured API integrations. A model name, consumer application, and API channel are not the same environment.
Next step

Interpret the result responsibly

We document response variability, model changes, location constraints, causal uncertainty, and the limits of available data.

Read the limitations Discuss the scope