What is measured
Two different things, kept apart because they answer different questions. A mention is the brand named in an answer's text. A citation is the answer pointing at a page on that brand's domain, in the sources an engine lists or links. An answer can mention a brand without citing it, and can cite a domain it never names.
Which engines
The answer engines Cituna tracks: ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode. A note says which of them it used. When a note gives one number across engines, it says how that number was combined.
How often, and on what
Every measurement comes from a set of prompts run against those engines and recorded with the date they ran. A note states its prompt set, the window it covers and the market and language it was run in, because all three change the answer. Numbers in a note are a snapshot of that window, not a running average, unless the note says otherwise.
What this cannot tell you
- Answer engines are not deterministic. The same prompt, asked twice, can return different text and different sources, so a re-run will resemble these numbers rather than reproduce them.
- Answers vary by account, location, language and time. A sample says what these prompts returned, not what every reader sees.
- A prompt set is a sample. A larger one moves a number, so a note says how large its sample was and you can judge how much weight the number carries.
Sources and corrections
A note that leans on something published elsewhere lists it at the foot, with the date it was read, because a page can change after it is cited. Mistakes are fixed in the note itself and the date at the top changes when that happens; the corrections policy says how.