AI Brand Monitor — GEO / AI Search Visibility Tracker

    

Is your brand recommended by ChatGPT, Gemini, Perplexity and Claude — or are your competitors?

Buyers no longer "Google it." They ask an AI. When someone asks "what's the best project management software?", the AI names a handful of brands — and if yours isn't one of them, you lost that customer before you ever saw them. This is the new SEO, and it's called GEO (Generative Engine Optimization) / AEO (Answer Engine Optimization).

This Actor measures, with statistical rigor, exactly how visible your brand is inside AI answers — and tells you what to do about it.

---

Why this tracker is different

Most AI-visibility tools ask each question once and report the result as if it were the truth. But AI answers are non-deterministic — ask the same question three times and you can get three different brand lists. A single-shot check is noise dressed up as data.

	Typical trackers	AI Brand Monitor
Sampling	Ask once per prompt	Ask N times, report a mention RATE + stability
Reliability	A coin-flip snapshot	Statistically grounded, repeatable score
Cited sources	Opaque redirect URLs / titles	Real publisher domains (g2.com, wrike.com…)
Actionability	"Improve your visibility"	Citation-gap list: the exact domains to earn citations on next
Sentiment	Sometimes	Positive / neutral / critical, on the brand's actual context

We ran the same prompt 3× during testing and watched the mention rate move between runs. That variance is real — and it's exactly why one-shot tools mislead. We report the rate and the stability, so you know how much to trust each number.

---

What you get

For a brand + category (and optional competitors), the Actor sends realistic buyer-intent prompts to AI engines with live web search enabled, samples each prompt multiple times, and returns:

Per prompt (one row each):

• Mention rate % — in how many samples your brand appeared
• Stability % — how consistent that outcome was across samples
• Average rank — where your brand lands among the brands the AI lists
• Sentiment — how the AI describes your brand (positive / neutral / critical)
• Competitors mentioned — who you're up against in that answer
• Cited sources — the real domains the AI grounded its answer on
• Answer excerpt — the raw AI response, for evidence

Summary scorecard (one row):

• AI Visibility Index (0–100) — one headline score (mention rate 50% + rank 30% + citation rate 20%), with a band: strong (70+) / mid-tier / visibility gap
• Trend — Δ vs. your previous run for the same brand, so scheduled runs show whether you're gaining or losing ground
• AI visibility % — overall mention rate across all samples and engines
• Share of voice % — your mentions vs. each competitor
• Overall sentiment + breakdown
• Citation rate — how often AI engines cite your own domain as a source (add your brandDomain)
• Top cited sources — the domains shaping answers in your category
• Citation gap — domains that ground competitor-only answers where you're absent. This is your GEO to-do list: earn a presence on these and you move the needle.

Built for automation & AI agents

• Webhooks / alerts — set a webhookUrl and we POST the full scorecard to Slack, Make, Zapier or n8n the moment a run finishes.
• Schedule it — pair with Apify Scheduler for weekly tracking; the trend delta does the rest.
• Agent-friendly inputs — accepts the field names AI agents use interchangeably (brand/brandName/company, competitors/rivals, engines/llms, prompts/questions).

Use it as an MCP tool (inside any AI agent — no install)

Call this Actor directly from Claude, Cursor, ChatGPT agents, LangChain, Make, Zapier or n8n via the Model Context Protocol:

Hosted (zero setup) — point your MCP client at:

https://mcp.apify.com?tools=foxlabs/ai-brand-monitor

Self-hosted (stdio clients — Claude Desktop, Cursor, Cline, Continue):

$npx @apify/actors-mcp-server --tools foxlabs/ai-brand-monitor

{
  "mcpServers": {
    "ai-brand-monitor": {
      "command": "npx",
      "args": ["-y", "@apify/actors-mcp-server", "--tools", "foxlabs/ai-brand-monitor"],
      "env": { "APIFY_TOKEN": "YOUR_APIFY_TOKEN" }
    }
  }
}

Your agent can then ask "How visible is Notion in AI search?" and get the scorecard back inline.

---

Quick start

1. Enter your brand (e.g. Notion) and category (e.g. project management software).
2. (Optional) Add competitors to benchmark against.
3. Run it.

That's it — no API keys, no setup. Sensible defaults are pre-filled.

{
  "brand": "Notion",
  "category": "project management software",
  "competitors": ["Asana", "Trello", "ClickUp", "Monday.com"],
  "engines": ["gemini", "openai", "anthropic"],
  "maxPromptsPerEngine": 5,
  "samplesPerPrompt": 3
}

---

Engines

Engine	Status
Gemini (Google) — live Google Search grounding	✅ Supported out of the box, no setup
ChatGPT (OpenAI) — live web search	✅ Supported out of the box, no setup
Claude (Anthropic) — live web search	✅ Supported out of the box, no setup
Perplexity	🔑 Bring your own API key

Three engines run with zero setup — Gemini, ChatGPT and Claude, each with real-time web search enabled. To also query Perplexity, paste your own Perplexity API key in the Advanced inputs; it runs on your account.

Engines disagree more than you'd think. In our own testing, Gemini and ChatGPT both recommended Notion for "project management software" — but Claude didn't mention it at all. Tracking multiple engines is the whole point: your visibility isn't one number, it's different on every surface.

---

Input reference

Field	What it does
brand	The brand to track in AI answers
category	Product category — used to auto-generate realistic buyer prompts
competitors	Brands to benchmark share-of-voice against
engines	Which AI engines to query (Gemini supported by default)
maxPromptsPerEngine	How many buyer-intent prompts (1–25). More = broader coverage
samplesPerPrompt	How many times to repeat each prompt (1–5). More = more reliable
prompts	(Advanced) Your own exact prompts, overriding auto-generation
*ApiKey	(Advanced) Bring your own key per engine

---

How it works (transparent by design)

1. Prompt generation — realistic buyer-intent questions are built from your category (or you supply your own).
2. Multi-sample querying — each prompt is sent to each engine samplesPerPrompt times, with live web search on.
3. Deterministic analysis — brand and competitor mentions are detected by exact token matching (no fuzzy guessing); rank is the order brands appear in; sentiment is classified from the sentences that mention your brand.
4. Aggregation — mention rate, stability, share-of-voice, sentiment, cited sources and the citation gap are computed across all samples.

No scraping, no cookie tricks — only official AI provider APIs.

A note on sentiment (v1): sentiment is classified with a transparent, deterministic keyword model over the brand's mention context. It's a fast, explainable signal — not a black-box LLM judgment.

---

Who it's for

• SEO / GEO agencies — sell evidence-backed AI-visibility reports to clients (recurring).
• Brand & marketing teams — track your standing vs. competitors over time.
• Founders — find out, today, whether AI is recommending you or your rivals.

Schedule it weekly to watch your AI visibility trend — because the answers change, and so should your strategy.

---

Built by foXLabs.