GEO Radar — AI Brand Visibility Monitor

Does AI recommend your brand — or your competitor? GEO Radar is a generative engine optimization (GEO) monitor: it checks your customer prompts against Gemini, Perplexity, ChatGPT (web search) and Google AI Overviews, and tells you exactly where you stand in AI search — mentions, position, cited sources, and share of voice vs competitors.

AI assistants are the new search results page. If your brand isn't in the answer, you don't exist. Enterprise GEO platforms run $100–500/month on self-serve tiers (and $30k+/year on enterprise contracts) — GEO Radar gives you the same core data pay-per-check, no subscription: 50 prompts × 2 engines weekly is a few dollars a month.

What you get for every prompt × engine

• Mentioned? — is your brand in the AI's answer at all
• Position — are you named first, or after two competitors
• Cited? — does the AI cite YOUR domain as a source (and at which position)
• Competitor breakdown — same metrics for every competitor you list
• Cited sources — the exact pages the AI pulled from (your GEO target list)
• Run summary — mention rate, citation rate, average position and share of voice per engine

Engines and API keys

Engine	Key needed	Where to get it
Google Gemini	none — built-in shared key (fair-use limits)	optional override: aistudio.google.com
Perplexity	your key, required	perplexity.ai/settings/api
ChatGPT (web search)	your OpenAI key, required	platform.openai.com
Google AI Overviews	your SerpApi key, required	serpapi.com

Gemini out of the box: the built-in shared key answers from the model's knowledge (no live Google Search grounding — that quota requires a billed Google key). Such checks are flagged grounded: false, carry no citations, and are billed at the cheaper basic check rate. Plug in your own billing-enabled Gemini key to get fully grounded answers with cited sources.

Example output

{
  "engine": "perplexity",
  "prompt": "best crypto tax software europe",
  "brand_mentioned": true,
  "brand_position": 2,
  "brand_cited": true,
  "brand_citation_position": 3,
  "competitors": [
    {"name": "Koinly", "mentioned": true, "position": 1, "citation_position": 1}
  ],
  "cited_sources": [
    {"position": 1, "url": "https://koinly.io/blog/...", "domain": "koinly.io"}
  ]
}

Failed checks are recorded in the dataset too (with an error column), and the run summary — mention rate, citation rate, share of voice per engine — is saved to the run's OUTPUT record.

How to use

1. Enter your brand (name, aliases, domain — domain enables citation tracking).
2. List competitors you want share-of-voice against.
3. Add the prompts your buyers actually ask ("best X for Y", "X vs Y", "is X trustworthy") — up to 100 per run.
4. Pick engines and run — or put it on a weekly Schedule to build an AI-visibility trend line.

Who uses this

• SEO/GEO agencies — sell AI search optimization audits to clients without enterprise tooling
• SaaS founders — track whether AI recommends you in your category and which pages get cited
• Content teams — find which sources AI trusts in your niche, then get cited there
• Brand monitoring — catch when ChatGPT or Gemini starts recommending a competitor first