Rankings

Measure how a set of brands rank in AI answers (mention rate, average position, and a per-model breakdown) from your agent.

Overview

Give your agent a query and a set of brands; it returns how often each model names each brand, in what position, and the per-model split: the raw signal for tracking AI visibility.

Example

Ask your agent in plain English and it calls the tool for you:

# you say
How do Attio, HubSpot and Folk rank in ChatGPT and Claude for “best CRM for SaaS”?

The tool returns structured JSON to your agent (you never see this directly):

{
  "query": "best CRM for SaaS",
  "rankings": [
    { "competitor": "hubspot", "mentionRate": 0.92, "avgPosition": 1.4 },
    { "competitor": "attio", "mentionRate": 0.71, "avgPosition": 2.1 }
  ],
  "billing": { "cost": { "usd": "0.14" } }
}

Your agent reads that JSON and answers you in plain English. That reply is written by your own AI; the JSON is just what it reasons over:

# your AI replies
HubSpot dominates: named in 92% of answers, almost always first (avg. position 1.4).
Attio shows up in 71%, usually 2nd. Folk barely registers. To move up, focus on the prompts where Attio is mentioned but ranked below HubSpot.

Full input and output schema: POST /v1/rankings. Pass an optional models list to narrow which models run (see per-model pricing).