AI Search Rank Tracker

Know what ChatGPT, Claude, Gemini, and Perplexity say when your customers ask about your market.

AI assistants are the new search results page. When someone asks ChatGPT "what's the best CRM for a small agency?", the brands in that answer win the customer — and the brands that aren't mentioned don't even know they lost. This actor measures your brand's AI visibility the way rank trackers measure Google rankings, and alerts you when it changes.

What it does

For every prompt you define (the questions your customers actually ask), the actor queries up to four AI assistants — ChatGPT, Claude, Gemini, and Perplexity — with web-grounded answers where available, then scores each response:

• Brand mentioned? How early in the answer? How many times?
• Recommended — is your brand endorsed, or just name-dropped?
• Negative context detection — flagged when your brand appears near criticism
• Share of voice — your mentions vs. each competitor you track
• Visibility score (0–100) per answer, per provider, and overall
• Cited sources — which domains the AI assistants cite when answering (these are the pages to influence)

Every run produces a clean dataset (one row per prompt × provider, full answer text included) plus an aggregated summary report, and stores history so runs are compared run-over-run — with webhook alerts when your visibility moves.

Why teams run it on a schedule

AI answers drift constantly: models update, grounding sources change, competitors publish. A single snapshot is trivia; a weekly tracked trend is strategy. Schedule this actor weekly (Apify Schedules, two clicks) and you get:

• A time series of your AI visibility per provider
• Alerts the week a competitor displaces you in ChatGPT's recommendations
• The exact cited domains to target with content/PR to improve your AI presence

Pricing

You pay per event — no subscription: one event per AI query (reduced rate with your own API keys), plus one run-report event per run. A weekly tracking run with 10 prompts × 4 providers is 40 queries + 1 report.

FAQ

Which questions should I use as prompts? The ones your buyers ask: "best X for Y", "X vs Y", "how do I solve Z". Start with 5–10; refine based on which show competitor mentions.

Do I need my own OpenAI/Anthropic keys? No — the default mode uses built-in access. Optionally supply your own keys for the reduced BYOK rate.

Is this SEO? It's GEO — Generative Engine Optimization. The citedDomains field tells you exactly which pages the AIs trust; earning mentions on those pages is how you move these scores.

Can other tools consume this? Yes — standard Apify dataset (JSON/CSV/Excel export, API access), webhook alerts for Slack/Zapier/n8n, and the actor works via MCP for AI-agent workflows.

---

Built and maintained by signalgrove — small data tools, clean output, honest pricing. Found an issue or want a feature? Open an issue on this actor's page; we respond fast.

---

Also by signalgrove: ATS Jobs API — job postings from Greenhouse, Lever, Ashby, Workday and 4 more ATSs, with new-job delta mode and webhook alerts.