AI Visibility
Do AI answers mention me?The share of valid observed AI answers mentioning the brand. It measures presence, not preference or website traffic.
See how your brand appears in observed ChatGPT, Gemini and Google AI answers. Compare competitors by topic and prompt, distinguish mentions from measurable recommendations and source citations, and find gaps worth investigating.
Four metrics answer four different questions: does the brand appear, how prominent is it, is it recommended, and is its website cited as a source?
The share of valid observed AI answers mentioning the brand. It measures presence, not preference or website traffic.
The weighted share of mentions among monitored brands, based on query weight, position and mention type. It is not market share.
The weighted share of competitive recommendations attributed to the brand. A mention alone does not count. When recommendations cannot be measured, the result should not be read as zero.
The weighted share of citations pointing to the brand domain compared with monitored competitors. A visible citation does not prove the content was used to generate the answer.
Percentages describe only the observed sample and period, not all AI users. A citation is not a recommendation; an unmeasured metric must not be confused with a measured zero.
Tracking starts from prompts aligned with market needs. Telescop separates valid observations from unavailable AI surfaces, so a provider failure is not silently counted as zero visibility.
Illustrative scenario: two software providers are both mentioned, but only one is explicitly recommended in some answers.
This illustrative Telescop dashboard preview is reconstructed from an observation dated October 15, 2026, with competitor names blurred. Presence, weighted share of voice, recommendations and citations are distinct signals.
The brand appears in a portion of observed answers. Presence alone does not prove preference.
Weighted share of voice captures the brand’s relative prominence within the monitored panel.
No explicit brand recommendation was measured in this sample; inspect buying prompts for the gap.
Some source citations are recorded, but citations and brand recommendations measure different outcomes.
Illustrative reconstruction, not a pixel-exact capture of the production interface. Figures describe a historical sample; provider coverage and results may change. Recommendations and citations are not guaranteed.
A single ChatGPT search cannot tell you how visible a brand is. You need a stable panel of questions reflecting customer needs, with language, market, dates and AI surfaces clearly documented. Telescop observes available answers and separates valid measurements from collection failures, so missing data is not mistaken for an absent brand.
AI answers can change over time. Visibility percentages describe the monitored sample, not every conversation people have with AI.
A brand may appear frequently in AI answers while remaining weak at the moments when buyers compare options. That is why it helps to interpret visibility alongside topics, search intent and available demand signals. The aim is not to invent another isolated score, but to find competitive gaps that deserve investigation and an actionable next step.
Reading signals together does not mean every metric is available on every surface, nor does it imply predicted leads or sales.
Before comparing pricing plans and dashboards, examine how the data is collected. Reliable AI visibility software should make the prompt panel, actually observed AI platforms and metric definitions clear. A feature list cannot replace an auditable methodology: ask what each number measures and which decision it can realistically support.
The right choice depends on your market, the number of brands monitored and the quality of evidence—not simply the number of charts.
In Telescop's AI Energy Choice Italy 2026 study, the Exposure leader changes across AI surfaces: Edison Energia leads on Gemini in the observed panel, while Plenitude leads on Google AI Overview. This illustrates why one prompt or one platform cannot describe the whole competitive picture.
These percentages refer to a defined research sample and different AI surfaces. They are not market shares, sales figures or guarantees for other brands.
Research examples and documented methods behind the metrics.
Examine mentions, share of voice, cited sources and recommendations across the prompts that matter to your market.