Recipes

Track brands across all AI search surfaces

Use mode:all_live to capture brand mentions across all 6 live AI search surfaces in a single call. Returns per-surface rank, citations, and fan-out queries — the full GEO picture for the price of a coffee.

Why it matters

The AI search universe is now 6+ surfaces — Perplexity, ChatGPT, Gemini, Google AI Overview, Google AI Mode, Bing Copilot — and each one has its own ranking, citation graph, and fan-out behavior. Calling them one at a time is slow and wasteful; mode:all_live fans them out in parallel, prorates the price on partial failure, and returns a single uniform envelope.

Cost: $0.50 per call when all 6 succeed; partial-success refund based on (k / 6) × 50¢ when fewer surfaces return. Latency is parallelized — p95 tracks max(per-surface p95) ≈ 30s, not the sum.

Code

bash
curl https://api.mentionsapi.com/v1/check \
  -H "Authorization: Bearer lvk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "all_live",
    "query": "best CRM for startups",
    "brand": "HubSpot"
  }' | jq '.providers | to_entries | map({surface: .key, mentioned: .value.mentioned, rank: .value.rank})'

Sample output

bash
--- best CRM for startups ---
{ surface: 'perplexity',   mentioned: true,  rank: 6,    citations: 10 }
{ surface: 'chatgpt',      mentioned: true,  rank: 4,    citations: 4  }
{ surface: 'gemini',       mentioned: true,  rank: 5,    citations: 6  }
{ surface: 'ai_overview',  mentioned: true,  rank: 3,    citations: 8  }
{ surface: 'ai_mode',      mentioned: true,  rank: 2,    citations: 5  }
{ surface: 'bing_copilot', mentioned: false, rank: null, citations: 7  }

HubSpot is mentioned on 5 of 6 surfaces for "best CRM for startups", with rank 2-6. Bing Copilot didn't render the brand — a useful gap signal: time to invest content for the Bing-driven half of the AI search universe.

Next steps