AI Visibility Tracker - Brand & Citation GEO

What is AI Visibility Tracker - Brand & Citation GEO?

AI Visibility Tracker - Brand & Citation GEO asks AI answer engines the questions your buyers ask and records whether your brand shows up. It sends buyer-intent prompts to Perplexity, ChatGPT and Gemini, with Claude as an option, and returns one dataset row per prompt and platform: whether the brand was mentioned and how early, mention count, share of voice against your competitors, sentiment, whether your domain was cited, every cited URL and source domain, and an answer excerpt. SEO and GEO agencies, brand and content teams, and AI agents use it to measure and report brand visibility in AI search.

Why GEO and brand teams use it

• Mentions and citations together. brandMentioned, brandPosition and brandMentionCount show presence in the answer; domainCited and citedUrls show whether your site is a source.
• Share of voice. List rivals in competitors and brandShareOfVoice compares your mentions with theirs.
• The sources AI engines trust. sourceDomains lists every domain an answer cites, a direct shortlist for digital PR and content placement.
• Ready-made buyer prompts. queryTemplates builds recommendation, alternatives and category-leader questions from your brand and industry; add your own in queries.
• Run-level GEO summary. RUN_SUMMARY adds a GEO score, platform scores, top source domains, top competitors and recommendations.
• Bounded cost. maxQueries times the number of platforms sets the exact number of checks.

Best fit for this Actor

• Best fit for tracking one brand across AI answer engines with a fixed set of buyer questions.
• Use targetDomain whenever you want citation tracking, and competitors whenever you report share of voice.
• Designed for answers to brand and category prompts; site crawlability audits belong to a technical SEO tool.
• When the job is domain citations for topic keywords with the brand in the background, use AI Search Visibility Tracker instead.

Example workflow: a monthly AI visibility report for a client

A GEO agency starts with a client in the CRM software market. The strategist sets brandName, targetDomain, industry: "CRM software", three competitors and five custom queries taken from sales calls, with the three default platforms. The dataset returns fifteen rows, one per prompt and platform, with brandMentioned, brandPosition, brandShareOfVoice, domainCited and sourceDomains. She compares this month's rows with last month's, then exports the top cited sourceDomains the client is missing as a content and PR target list. She schedules the same input monthly so the trend line builds itself.

Input

{
  "brandName": "Apify",
  "targetDomain": "apify.com",
  "industry": "web scraping tools",
  "competitors": ["Zyte", "Octoparse"],
  "queryTemplates": ["recommendations", "alternatives", "category_leaders"],
  "platforms": ["perplexity", "chatgpt", "gemini"],
  "maxQueries": 3,
  "responseFormat": "concise"
}

Field	Type	What it controls
brandName	string	The brand, company or product to track, written as customers write it. Required.
targetDomain	string	Your domain, such as apify.com, for citation tracking.
industry	string	Market category used in built-in prompts. Default software tools.
competitors	array of strings	Rival brands or domains for share of voice.
queries	array of strings	Your own prompts, one per line.
queryTemplates	array	Built-in prompt types. Default recommendations, alternatives and category_leaders.
platforms	array	perplexity, chatgpt, gemini and optional claude. Default the first three.
maxQueries	integer	Unique prompts per run before multiplying by platforms. Default 3.
responseFormat	string	concise or detailed answer excerpts. Default concise.
webhookUrl	string	HTTPS endpoint that receives the run summary.
demoMode	boolean	A no-charge connectivity check that writes OUTPUT and RUN_SUMMARY only.

Output

One row per prompt per platform. A real Perplexity row (identifier and cited URL list shortened):

{
  "runId": "exampleRunId01",
  "platform": "perplexity",
  "query": "Which web scraping tools should I consider, and is Apify a good option?",
  "queryCategory": "recommendations",
  "brandName": "Apify",
  "targetDomain": "apify.com",
  "brandMentioned": true,
  "brandMentionCount": 4,
  "brandPosition": 2,
  "brandShareOfVoice": 1,
  "sentiment": "neutral",
  "domainCited": true,
  "citedUrl": "https://blog.apify.com/best-web-scraping-tools/",
  "citedUrls": [
    "https://www.octoparse.com/",
    "https://scrapingbee.com/blog/ai/best-ai-scraping-tools",
    "https://blog.apify.com/best-web-scraping-tools/",
    "https://www.zyte.com/learn/best-web-scraping-tools/"
  ],
  "sourceDomains": ["octoparse.com", "scrapingbee.com", "blog.apify.com", "zyte.com"],
  "competitorMentions": [],
  "competitorMentionCount": 0,
  "aiAnswerExcerpt": "The best web scraping tool depends on your use case: no-code tools are best if you want speed and simplicity, while frameworks/APIs are better if you need scale... Apify is a solid option, especially if you want a versatile cloud platform...",
  "modelUsed": "perplexity/sonar",
  "checkedAt": "2026-10-01T01:36:29.239Z"
}

Field	What it contains
platform, modelUsed	The AI engine and model that answered
query, queryCategory	The exact prompt and its template type
brandName, targetDomain	What was tracked
brandMentioned, brandMentionCount, brandPosition	Presence, count and how early the brand appears, from 1 early to 10 late
brandShareOfVoice	Brand mentions divided by brand plus competitor mentions
sentiment	Positive, neutral or negative tone around the mention
domainCited, citedUrl, citedUrls, sourceDomains	Whether your domain was a source, plus every cited URL and domain
competitorMentions, competitorMentionCount	Which of your competitors appeared
aiAnswerExcerpt	Answer text for review
runId, checkedAt	Identifier of the producing run, and the check time

Pricing

This Actor uses Pay per event with Apify platform usage included in the event price. You pay one AI visibility check event for each prompt answered on one platform and saved, plus one Actor Start event per run. demoMode runs add no visibility check events. The live Pricing tab on this page shows the current event prices.

Worked example in words: three prompts on the three default platforms make nine checks, so the run costs nine AI visibility check events plus one start event. Five prompts on four platforms including Claude make twenty checks. The run logs its maximum event cost before it sends the first prompt.

Use with AI agents (MCP)

Connect the Actor to an MCP client through Apify MCP and ask:

Check whether Apify is mentioned and cited by Perplexity, ChatGPT and Gemini for web scraping tool questions, compare its share of voice with a managed unblocker and Octoparse, and list the domains these answers cite most.

Agent checklist:

1. Set brandName, targetDomain, industry and competitors; keep maxQueries small for a first pass.
2. Read rows with get-dataset-items; group by platform and count brandMentioned and domainCited.
3. Read RUN_SUMMARY for the GEO score and top source domains, and cite checkedAt and modelUsed as provenance.
4. Cost scope: checks equal prompts times platforms, so maxQueries and platforms set the spend.

API example

curl -X POST "https://api.apify.com/v2/acts/khadinakbar~ai-visibility-tracker/run-sync-get-dataset-items?token=YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"brandName": "Apify", "targetDomain": "apify.com", "industry": "web scraping tools", "competitors": ["Zyte", "Octoparse"], "maxQueries": 3}'

Related AI search tools

Your next step	Use	Why
When you track domain citations for topic keywords	AI Search Visibility Tracker	Citation tracking where the domain leads
When you monitor Google AI Overviews specifically	AI Overviews Tracker	Google AI Overview visibility

Best results

Starting situation	Helpful action	Expected outcome
A first baseline	Defaults with maxQueries: 3	Nine checks across three engines
Real buyer language	Add sales-call questions to queries	Prompts that match how customers ask
Share-of-voice reporting	List rivals in competitors	brandShareOfVoice and competitor mentions per row
Citation tracking	Set targetDomain	domainCited and your cited pages
Month-over-month trends	Schedule the same input	Comparable rows with checkedAt
Testing a connection	demoMode: true	OUTPUT and RUN_SUMMARY with no checks billed

AI answers vary from run to run, so read each row as a dated sample and compare trends across repeated runs with the same prompts.

Builder's note

I designed each row as one prompt on one engine because that is the unit an agency reports on and the unit a client can verify by asking the same question. I learned that the cited source domains are often the most actionable part of the result, since they show exactly which publications shape AI answers in a category, so every row keeps them alongside the mention data.

FAQ

Do I need my own AI provider keys?

No. The Actor uses owner-managed access to each engine.

Can I add Claude?

Yes. Add claude to platforms.

Where is the GEO score?

In RUN_SUMMARY in the default key-value store, together with platform scores and recommendations.

Responsible use

This Actor sends prompts to AI providers and records their answers, citations and derived metrics. Keep prompts free of private customer data and secrets, review provider terms before publishing results, and present AI answers as samples taken at a point in time.