Skip to main content
Buyers now ask AI engines which product to choose. Citino asks those engines the questions your buyers ask, every day, keeps every answer in full, and shows you who gets named, which websites the engines read before they decide, and what that is worth in visits to your site.

​
What happens every day

1

We ask

Every question in your project is asked on every engine you track, once a day, from the country your project is set in. The questions are phrased the way a buyer would ask them, and each one is asked for you alone.
2

We keep the answer

Each answer is stored in full, exactly as the engine gave it. You can read any of them on the Answers screen, so a number never has to be taken on trust.
3

We read it

We note which brands the answer named, in what order and how favourably, and which websites the engine read or cited while writing it.
4

We count

That becomes your visibility, share of voice, position and sentiment, each with how many answers sit behind it and how far it could be off.
By default a project is asked on ChatGPT, Google AI Overview and Perplexity. Other engines can be switched on per project: see Engines and opt-ins.

​
What you can do with it

The app’s rail starts with two groups. Measure is where you stand:
  • Overview: your four headline numbers, what moved, and the questions you are losing.
  • Prompts: every question we ask for you, and how you do on each.
  • Answers: every answer the engines gave, in full.
  • Brands: every brand named in your answers, ranked by visibility.
  • Attribution: visits to your site from AI engines, from your Google Analytics, traced back to the questions that sent them.
  • Claims: whether the engines say what is true about you.
  • Objections: the reasons the engines give a buyer not to choose you.
Act is what to do about it:
  • Sources: the websites the engines read, and which of them you can do something about.
  • Gaps: things worth doing to get named more often, ranked by how much each should move your visibility.
  • Experiments: whether a change you published worked, measured against your competitors over the same days.
  • Crawlability: whether your site lets the AI engines’ crawlers in.
  • Crawler activity: which AI crawlers fetched which of your pages, from your own server logs.

​
How we keep the numbers honest

  • Every number says what it rests on. It carries n, the answers behind it, and a ± range. See The ± band and n.
  • A small change is not news. A change smaller than its ± range is shown muted as “within noise” and never raises an alert. See Within noise.
  • Engine changes are labelled, not hidden. When an engine changes how it answers, the day is marked on your charts, and no answer is dropped from any figure. See Drift labels.
  • Your headline numbers count the questions that matter. Only questions that name no brand, in the segments you compete in. See Scope and segments.
  • Your history starts with your first collection. Nothing is inherited or filled in before the day we started asking for you.

​
Private by design

No workspace can read another workspace’s questions or answers, even when two competitors both use Citino.

​
Who it is for

Citino is built for a brand’s own growth or search team, and for agencies who track many client brands, each as its own project. If you are an agency, read Clients after the quickstart.

​
For developers

The same numbers are available through the REST API, and Claude and ChatGPT can read them through the MCP server.

Quickstart

Sign up and see your first figures.

Visibility

The number most people look at first.

What we do not measure

The limits of asking AI engines on your behalf.

Glossary

Every term the app explains, in its own words.