Profound
Archipel is among the first Profound partner agencies in Europe and contributed to its agency mode. We also use it to estimate prompt volumes.

AI visibility tracking
AI visibility tracking measures, prompt by prompt and engine by engine, whether ChatGPT, Claude, Gemini, Copilot, Perplexity and Google AI Overviews cite your brand, at what rank and against which competitors. Archipel runs it in Profound or Peec AI, tool included, with a dashboard open to your teams: transparent measurement of your AI search optimization.
Updated
50 to 100
prompts tracked per account, typically
2
partner tools: Profound and Peec AI
Included
the tracking tool, in the GEO program
1+ year
of history on some accounts
Why measure
AI answers change from one day to the next, one user to the next, one engine to the next. Typing your question into ChatGPT once tells you nothing about your real visibility. Only daily tracking, on a stable list of business prompts, tells you whether an action worked.
That is the foundation of our GEO work: every prompt in your GEO structure is run every day on the selected engines, in a professional tool, in every language and country you target. Our method itself is updated every week from the combined data of all our active clients, who typically track 50 to 100 prompts each, sometimes with more than a year of history.
Metrics
| Metric | Definition | What it tells you |
|---|---|---|
| Visibility rate | Share of answers in which your brand appears | Whether AI engines know you on this topic |
| Average rank | Average position of your brand when it is cited | Whether you are recommended first or last |
| Share of voice | Your mentions relative to your competitors’ | Your real weight in the category |
| Cited sources | Pages and domains the answers rely on | Where to act to earn citations |
| Sentiment | Tone of the answers that mention you | How AI engines describe you |
Tools
Archipel is among the first Profound partner agencies in Europe and contributed to its agency mode. We also use it to estimate prompt volumes.
We are a Peec AI partner, with agency accounts. It is the tool that measured our own jump from 14% to over 80% visibility in ChatGPT in October 2026.
Bought directly, a tracking tool can cost up to €4 to €5 per tracked prompt. At Archipel, it is included in the program, from €990 per month excl. VAT.
Reporting
You have access to the tool, to the same data as we do, in real time. No cherry-picked screenshots, no flattering averages: you see every answer, every competitor and every source. Our automated reports summarize the trends, but the raw data stays at your disposal.
Results are read by topic and by tag, so you can track your offers, cities, countries or personas separately, and watch the "Main Activity" tag first: the prompts where you must be first. To roll out one of these tools in-house, see Profound and Peec AI setup.
Cadence
Your GEO consultant reviews prompt trends, the actions taken and those planned for the next period.
Every month, a full report: visibility, rank, share of voice, content published, citations earned, priorities.
Every quarter, a strategic review: new topics, new countries or languages, new engines, link to revenue.
We also come to see you, in Paris and Lyon, to work with your marketing and sales teams. International clients are followed remotely, in English or French.
All the way to the contract
Visibility is only a means. What counts is meetings. Yet a prospect who discovered your brand in ChatGPT often arrives by typing your name, which analytics tools classify as direct traffic or branded search.
So we add a "How did you hear about us?" field to your forms, with ChatGPT, Claude, Perplexity and the other engines as options. Matched with your CRM, it lets you count AI-sourced leads and connect your AI visibility to your sales pipeline.
FAQ
Free GEO audit: your rank on ChatGPT, Claude and Google AI Overviews, your competitors and your first actions. Reply within half a business day.