Which engines are measured, and how?

Supported engines: ChatGPT (OpenAI), Gemini, Claude, Perplexity, Google AI Overviews and Google AI Mode. The Engines section of your brand settings shows which ones your brand is measured on.

We measure ChatGPT, Claude, Gemini and Perplexity through the providers' official APIs. Google AI Overviews and AI Mode have no official API, so we read those two engines through a licensed search-data provider (SerpApi LLC, USA). We never scrape any engine's consumer interface ourselves. An answer may not be identical to what you see in the same engine's app. Each prompt is asked in your brand's language with web search enabled. ChatGPT, Claude, Perplexity and Google also get your brand's country; the Gemini API offers no country setting. We extract the brands mentioned and the sources cited in each answer.

The same prompt is not asked of the same engine twice within 24 hours: "Rescan" on an interrupted scan only fills the gaps. Ai answers change over time; read the trend, not a single scan.

The first scan starts as soon as you approve the prompts while adding a brand. If you choose "I'll do this later", the prompts are saved and the first scan runs on its own by the next morning (03:00 UTC); to start it right away, use "Start the first scan" on the Overview. Later scans follow the brand's frequency, weekly or daily. If your credit doesn't cover a full scan, it runs as far as the remaining credit allows; with no credit left, no scan starts.