Get AI to recommend your business.
MarquePhare is a generative engine optimization (GEO, also called AEO) platform that measures how ChatGPT, Gemini, Mistral Vibe, Claude and Perplexity cite your brand — then writes and optimizes the content they'll cite.
Cancel anytime · Tracks ChatGPT, Gemini, Mistral Vibe and Google AI Overviews
Tracks every answer engine that matters
Built on SERP data and LLM citations we actually collect — not estimates. Every number is dated.
monthly users on ChatGPT — and on Gemini (OpenAI, Google, 2026).
of AI users now start their searches with an assistant instead of Google (Eight Oh Two, 2026).
Traffic from ChatGPT converted at 15.9%, versus 1.76% for Google (Seer Interactive case study).
of those recommendations show up in your Google Analytics.
Your customers now ask AI who to choose. When it recommends a competitor, you lose your most qualified lead — and none of your current tools can see it.
From blind spot to citation
One loop, four steps. Other tools stop at the audit. MarquePhare takes you all the way to the citation.
Step 1 · Audit
40+ checks: can AI crawlers read your site, and are your pages worth citing?
Step 2 · Measure
Where models name you, who they cite instead, and which sources they rely on — with sample size and margin of error.
Step 3 · Write and optimize
The content studio writes the citable content that closes the gap, and scores it against the same checks as the audit before you publish.
Step 4 · Re-measure
Publish, re-measure your citation share, and the action plan automatically re-prioritizes the next step.
Other tools tell you you're losing.
We fix it, all in one place.
Monitoring tools hand you a dashboard and an invoice. MarquePhare closes the loop: spot the gap, then write and optimize the content AI engines cite — without leaving the app.
See the full report before you commit
The exact report a client receives, for a fictional brand: what its customers ask AI, who AI recommends instead, the 40+ check audit, the site's topic map, and an action plan ranked by impact.
Excerpt — sample data
AI already recommends you for your core business
Cited in 7 of 24 answers, 3 of them in first position (Mistral Vibe, ChatGPT).
Frequently asked questions
What is GEO (generative engine optimization)?
GEO — also called AEO (answer engine optimization) — is the practice of getting your brand cited in the answers generated by ChatGPT, Gemini, Mistral Vibe, Claude and Perplexity, instead of (or as well as) ranking in classic Google results.
How is this different from Profound or Peec?
Those tools only measure — they show you a dashboard of where you stand. MarquePhare measures AND acts: it audits your site against 40+ technical checks, finds the authority platforms you're missing, then writes and scores the content that closes the gap, all in the same app. Tracking is the first step, not the product.
Which AI engines do you track?
ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews — with real share of voice, position and sentiment per engine, based on dated SERP data and LLM citations refreshed at regular intervals.
Do I need to be technical to use it?
You enter a topic, the GEO content studio does the research and writes a citable article, and a chain of agents tells you exactly what to improve. You stay in control; the tool does the work.
Does it work for my language and market?
Yes — intent classification and SERP research run per topic and per market, so the analysis reflects how AI answers in your specific context. The interface is available in English, French and Spanish.
What exactly does the GEO audit check?
More than forty checks across four areas. Technical: AI crawler access in robots.txt, bot by bot (OAI-SearchBot, GPTBot, PerplexityBot, ClaudeBot, Google-Extended, Bingbot), server-side rendering, JSON-LD structured data, sitemap and llms.txt. Content: direct answer, sourced data, structure, FAQ, comparison table. Mentions: your real share of voice per engine. Authority: your link profile, the platforms AI cites without linking to you, and the domains that reference your competitors but not you.
Can I deliver reports to clients under my own brand?
Yes. Reports carry your agency's logo and colors, and export to CSV. They break down share of voice engine by engine, the competitors cited instead of you, the sources AI relies on, and the authority you're missing — every number shown with its sample size.
What is the topic map, and what is it for?
Tracking tools start from keywords: they tell you what ranks. We start from the site itself. A crawl maps the entities the brand claims — every service, product, technique and service area — then links each one to the pages that cover it. Anything left unlinked is a topic you claim without a page for it, so an answer engine has nothing to cite. Measured on a construction client: 71 such topics, each one named, plus 14 pages no internal link reached. That's a complete editorial plan, drawn from what the site already says it is.
Can I query my data from Claude or Cursor?
Yes. MarquePhare exposes an MCP server: create an access key in settings, then query your clients' visibility straight from Claude Code, Claude Desktop or Cursor — without opening the app.
A control center, not just a dashboard
Seven layers that put you in control, from site audit to client report.
GEO site audit
40+ checks, page by page: AI crawler access, server-side rendering, structured data, citability.
AI mention tracking
Your share of voice across six engines, with its date, market and sample size.
Authority and link building
The platforms AI cites instead of you, and the domains linking to your competitors.
Site topic map
What your brand claims, versus what no page covers. A named list, not a score.
Prioritized action plan
Every finding becomes an action ranked by impact and effort. You know where to start.
GEO writing and scoring
The editor writes a citable article, then scores it against the same checks as the audit — before you publish.
White-label reports
The client deliverable in your colors: share of voice, competitors, cited sources. Exports to CSV.
Three metrics that decide AI search
Measured on ChatGPT, Gemini, Mistral Vibe, Claude and Perplexity.
Visibility
Your share of voice — how often AI names your brand when it answers. Market share, for AI conversations.
Position
The order in which AI names you when it lists several brands: being named first is worth more than being mentioned last.
Sentiment
The tone models use when they cite you — positive, neutral or negative. Only measurable on answers where you appear.