Skip to main content

​
Your numbers

Of all the answers we collected, the share that named you. 40% means you were named in 4 of every 10 answers.
It counts your unbranded questions in the segments you claimed, over the period and engines in the screen’s scope. See Visibility.
Three reasons. We ask signed out, with no chat history or memory, so a signed-in answer can differ. We ask from your project’s market. And engines answer differently from one ask to the next, which is why we ask every day and show a ± range rather than trusting one answer. See How answers are collected.
— with “needs one full run” means there are no answers in scope yet, so nothing has been measured. 0% · not named means there are answers and none of them named you. That is a finding, not missing data.
How much this moved since the previous period of the same length, in percentage points. “Within noise” means the move is smaller than the ± range, so it could be chance.
A change within noise is shown muted and never raises an alert. See Within noise.
No. A question that names your brand would name it nearly every time, so your headline numbers count only the questions that name no brand. Questions that name a brand are collected and kept separately. See Scope and segments.
We label the day as drift on every chart and keep every answer in every average. Alerts and experiments compare you with your competitors on those days, so a move that hit everyone is not reported as yours. See Drift.

​
Questions and engines

ChatGPT, Google AI Overview and Perplexity for every project. Gemini, Google AI Mode and Claude Haiku can be switched on in Project settings once your subscription is paid. Claude Sonnet, Grok and Claude Opus are on Enterprise plans. See Engines and opt-ins.
How often we ask your questions. Every question is asked once a day.
That is the same on every plan. See Cadence.
No, you replace it, through the API. As the app says, “A question is never edited, only replaced, so its history stays comparable.” The new wording starts its own history, and the old one stays readable.
No. Archiving a question, which is done through the API, frees its slot and keeps its history readable. Turning off an engine stops future runs of your prompts against it and keeps its answers, but figures count only the engines switched on, so its history leaves the charts while it is off and returns when you switch it back on.
Today every project asks from the United States, in English.

​
Plans

It depends on your plan, and the allowance is shared across every project in the workspace. Archived questions and Citino’s own questions about your brand do not count. See Plans and prompt limits.
There is no Billing page in the app yet. Plans and billing are arranged with us: contact us.
Yes. Both are available on every plan. An owner or admin creates keys in Workspace settings, and any member can connect Claude or ChatGPT by signing in.

​
Your data

No. Your questions, answers and every number worked out from them belong to your workspace alone, including when a competitor is also a customer. Nothing about your answers reaches another workspace, not even as a total.
Read-only access to the one GA4 property you choose for a project, and only totals per day, per referring site and per landing page: sessions, conversions and revenue. Never a person. “Imported sessions are deleted within 24 hours of disconnect.” See Integrations.
Yes. Add https://app.citino.ai/api/mcp as a connector and sign in. The connection is read-only and reads what you can read. See Connect Claude or Connect ChatGPT.
Yes, through the API, which can also return rows as CSV. Every stored answer is readable in full on the Answers screen.
Only if you connect WordPress or Webflow, and then only an approved draft you choose to send, which arrives as a draft. Nothing goes live until you publish it. See Integrations.

Troubleshooting

When something on a screen looks wrong.

Glossary

Every term the app explains.

What we do not measure

The limits of the sample.

Contact

Write to us.