Read your AI Visibility results
Understand mentions, recommendations, citations, and question coverage without treating them as universal AI rankings.
Explain a visibility metric using its question set, underlying answers, and sources.
On this page
Start with the measurement scopeUnderstand the main metricsInspect a specific answerCompare like with likeTechnical details and methodologyStart with the measurement scope
Open KI-Sichtbarkeit and check the brand, market, language, question set, and available providers. Results describe the answers collected for that scope. The same provider may answer a different wording or a different session differently; your measurement is a sample of responses, not a guarantee about every shopper's experience.
Understand the main metrics
Open the underlying answers when a headline percentage needs explanation. Different metrics can have different eligible samples.
| Metric | How to read it |
|---|---|
| Brand mentions | How often your brand appears in the eligible measured answers. A mention can be neutral and is not necessarily a recommendation. |
| Recommendations | How often an answer recommends your brand or offer in the measured context. Read the answer to understand the reasoning. |
| Own-source citations | Whether the answer links to an identified source belonging to your business. A citation and a brand mention are separate observations. |
| Question coverage | How broadly your brand appears across the monitored questions. More answered questions does not automatically mean broader brand coverage. |
| Measured answers | The amount of usable evidence behind the rates. Check missing or incomplete observations before comparing percentages. |
Inspect a specific answer
Open Antworten, choose a question or provider, and read the actual answer and linked sources. Confirm that a mention belongs to your business rather than a similarly named company. Distinguish your retailer brand from a product brand that many shops sell.
Compare like with like
Keep a stable core question set and note changes in providers, market, language, or competitors. A higher percentage after removing difficult questions is a different sample, not automatically an improvement. If a run is still processing, wait for its results or use the last completed measurement.
Technical details and methodology
For the collection steps, eligibility rules, calculation units, and limits behind this workflow, use the implementation-reviewed technical guides below.
Still need help?
Tell us what you expected and what happened.