AI search analytics

Be the answer AI gives.

AEO measures how ChatGPT, Claude, Gemini, Perplexity and Google AI Mode talk about your brand — and how ready your site is for the crawlers behind them. Then it hands the same data to your agents over MCP.

Free teaser scan · no card required · agent-native MCP included on Pro

Visibility

42%

+5.2% this month
2.1Position
18%Share of voice
+68%Sentiment

Tracked across every major AI surface

ChatGPTClaudeGeminiPerplexityGoogle AI Mode
What you get

Four signals that decide whether AI recommends you

The dashboard, the metrics, and the audit — all fed by the same collection pipeline.

AI visibility tracking

Sampled runs per prompt across models and chat surfaces — with Wilson 95% confidence intervals. Track mention rate, position, and share of voice over time.

Competitor benchmarking

Track competitors next to your brand. Know who actually gets named when a buyer asks an AI what the best option is.

Source intelligence

Every citation classified as owned, competitor, editorial or UGC. Find the domains AI trusts — and the ones citing competitors but not you.

AI-readiness audit

Robots.txt for 40+ AI crawlers, llms.txt, structured data, parseability, and raw-vs-rendered content — scored 0–100 with prioritized fixes.

Agent-native

Your AI visibility, exposed to your agents over MCP

Every workspace ships an authenticated MCP server. Authenticate with a scoped Bearer key and let Claude, Cursor, or your own agent query visibility, sources, prompts and site audits directly — no screenshots, no CSV wrangling.

get_projectsget_visibilityget_sourcesget_site_auditlist_competitorsget_prompts
mcp client config
{
  "mcpServers": {
    "aeo": {
      "url": "https://your-host/mcp",
      "headers": { "Authorization": "Bearer aeo_••••••••" }
    }
  }
}
claude code
claude mcp add --transport http aeo https://your-host/mcp \
  --header "Authorization: Bearer aeo_••••••••"
Metrics you can defend

Not one lucky answer — a distribution

AI answers are non-deterministic. AEO treats them that way, so a number you show your team holds up under scrutiny.

See plans
  • Probability-based metrics from N sampled runs per prompt, engine and day.
  • Wilson 95% confidence intervals — never a single lucky answer.
  • Position is conditional on inclusion, so rank is only averaged over real mentions.
  • Deterministic extraction for mentions and citations; calibrated LLM judges for sentiment and claims.
FAQ

Questions, answered

What is AEO?

AEO measures how your brand shows up inside AI answers and how ready your own site is for the crawlers behind them — then exposes that data to both a dashboard and your agents.

Which AI engines do you cover?

Model-level via OpenRouter (GPT, Claude, Gemini, Llama and more), plus chat surfaces: ChatGPT, Claude, Gemini, Perplexity and Google AI Mode.

How is this different from rank tracking?

Rank tracking asks where you sit on Google. AEO asks whether you are mentioned at all when someone asks an AI what is best, where you rank in that answer, how you are framed, and which sources caused it.

Do I need an API key for the agent integration?

Yes. The MCP server is authenticated with a per-user Bearer API key. Create one from Settings → API keys; MCP access is available on the Pro plan and above.

Is there a free plan?

Yes — the free teaser runs an instant scan and shows your real visibility before you pay. No card required.

Can I cancel anytime?

Yes. Manage or cancel from the Stripe customer portal; access continues until the end of the billing period.

See how AI describes your brand — for free

Run an instant scan, get your real visibility, and connect an agent in minutes.