AI Citations Finder: LLM Mentions in ChatGPT & AI Overviews avatar

AI Citations Finder: LLM Mentions in ChatGPT & AI Overviews

Pricing

from $3.00 / 1,000 ai questions

Go to Apify Store
AI Citations Finder: LLM Mentions in ChatGPT & AI Overviews

AI Citations Finder: LLM Mentions in ChatGPT & AI Overviews

AI citations finder and MCP tool: real LLM mentions of your domain or brand in ChatGPT and Google AI Overviews, no prompts needed. One row per AI question: AI search volume, answer excerpt, cited sources, your source position. Fee per target searched plus per question.

Pricing

from $3.00 / 1,000 ai questions

Rating

0.0

(0)

Developer

PlainRows

PlainRows

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

6 days ago

Last modified

Share

An AI citations finder and MCP tool: give it a domain or a brand name and get the real questions people asked ChatGPT or Google AI Overviews where it is mentioned or cited, one row per question, with the AI search volume, the monthly trend, an excerpt of the AI answer, the sources the answer cites, whether your site is one of them and at which position, and the other domains cited next to you (your competitors in AI answers).

No prompts needed: you do not have to guess what people ask the AI. The questions come from a database of real ChatGPT and Google AI Overviews answers (LLM mentions), updated continuously by the data provider. One search per target, about 2 seconds.

Typical uses

  • AI visibility / GEO audit: which questions make ChatGPT or Google AI Overviews talk about your brand, and how often they are asked.
  • Citation tracking: targetCitedAsSource and targetSourcePosition tell you if the answer links to your site, and where in the list of sources.
  • Competitor research in AI answers: run it on a competitor's domain, or read otherCitedDomains to see who is cited next to you.
  • Content ideas: the questions with a high aiSearchVolume where you are mentioned but not cited are pages worth writing.
  • Reports: send the rows to Google Sheets, Looker Studio, Slack, n8n, Make or Zapier.

What you get for each question

FieldDescription
target, targetTypeThe domain or brand name searched; domain or keyword
platform, modelNamegoogle (AI Overviews, model google_ai_overview) or chat_gpt
questionThe real question asked to the AI
aiSearchVolume, monthlySearchesEstimated monthly AI searches for the question, and the monthly trend (month, searchVolume)
firstResponseAt, lastResponseAtFirst and last time this answer was seen
answerExcerpt, answerLengthThe first 300 characters of the answer as plain text (not the full answer), and the full length
targetMentionedInAnswerThe answer text names your target
targetCitedAsSource, targetSourcePosition, targetSourceUrlYour site is among the cited sources; its position (1 = first) and URL
sources, sourcesCountThe first 10 cited sources (position, domain, URL, title, source name) and how many there are in total
otherCitedDomainsThe other domains cited in the same answer (up to 20)
totalQuestionsAvailableHow many questions the database holds for this target (can be more than you asked for)
found, status, messageFOUND for a question row; rows without a question say why: NO_QUESTIONS, PROVIDER_ERROR, REJECTED, NOT_SEARCHED

Missing values are null, never guessed. Subdomains count as your site (www.pipedrive.com, support.pipedrive.com for pipedrive.com). Your position is computed over all the sources, even beyond the first 10 listed.

Example row (real data, pipedrive.com, Google AI Overviews, US, October 2026, shortened)

{
"target": "pipedrive.com",
"targetType": "domain",
"status": "FOUND",
"platform": "google",
"modelName": "google_ai_overview",
"question": "customer management software",
"aiSearchVolume": 165000,
"monthlySearches": [{ "month": "2026-06", "searchVolume": 246000 }, { "month": "2026-07", "searchVolume": 201000 }],
"firstResponseAt": "2025-10-31T07:27:07.000Z",
"lastResponseAt": "2026-09-02T09:55:57.000Z",
"answerExcerpt": "Customer management software, also known as a CRM (Customer Relationship Management) system, centralizes your customer data, tracks interactions, and automates sales and support tasks. Top CRM Software Options - Salesforce Sales Cloud : Best for large enterprises and fast-scaling teams needing…",
"targetMentionedInAnswer": true,
"targetCitedAsSource": true,
"targetSourcePosition": 12,
"sourcesCount": 12,
"sources": [
{ "position": 1, "domain": "salesforce.com", "url": "https://www.salesforce.com/crm/what-is-crm/", "title": "What Is CRM (Customer Relationship Management)? - Salesforce", "sourceName": "Salesforce" },
{ "position": 2, "domain": "hubspot.com", "url": "https://www.hubspot.com/products/crm", "title": "Streamline Your Entire Business With a Free CRM | HubSpot", "sourceName": "HubSpot" }
],
"otherCitedDomains": ["salesforce.com", "hubspot.com", "pcmag.com", "microsoft.com", "zoho.com", "oracle.com", "zapier.com", "zendesk.com", "uschamber.com", "appvizer.com"],
"totalQuestionsAvailable": 10519,
"country": "US",
"languageCode": "en"
}

In this real test, all 100 questions returned for pipedrive.com cited the site as a source, and 21 named it in the text of the answer (a citation link alone does not count); the database held 10,519 questions for it.

Input

{
"targets": ["pipedrive.com", "Pipedrive"],
"platform": "google",
"country": "US",
"maxQuestionsPerTarget": 100
}
  • targets (required): up to 100 per run. A value with a dot and no space (pipedrive.com, https://www.pipedrive.com/en) is searched as a domain (where your site is cited or named); anything else (pipedrive, pipedrive crm) as a brand name or keyword mentioned in the questions and answers.
  • platform: google (Google AI Overviews, default) or chat_gpt (ChatGPT).
  • country: for Google AI Overviews, one of US, GB, CA, AU, IE, IN, DE, AT, CH, FR, BE, ES, MX, IT, NL, PT, BR, PL, SE, JP (default US). ChatGPT data exists for the United States only: another country with chat_gpt is refused at once, before anything is searched (only the $0.002 run start fee is charged). languageCode defaults to the country's main language.
  • maxQuestionsPerTarget (default 100): 1 to 1,000 questions per target.

Pricing

Pay per event, no subscription (see the Pricing tab for the current prices):

  • Target searched: a fee once per domain or brand name searched, including a search that finds 0 questions. The data source bills every search, empty or not, so an empty search is not free: its row says "0 question found; the search itself is charged".
  • AI question: plus a fee per question row returned.
  • A small Actor start fee per run.

For example, one domain with 100 questions = 1 target fee + 100 question fees. Targets the data source failed or refused, and targets not reached (budget or timeout), get a free row with found: false and the reason: no fee at all. A target starts only when the run's maximum cost still covers its fee and at least one question; the questions are cut to what the budget allows.

Free Apify plan: not available. Every search costs the data source $0.10, so on a free Apify plan the run searches nothing and charges only the $0.002 run start fee (one free row per target says so). Any paid Apify plan works.

Good to know

  • The data is a database of AI answers collected by the data provider, not a live question to the AI: an answer can be a few weeks old (lastResponseAt says when it was seen). For live answers to your own prompts, use AI Search Visibility Tracker.
  • Small or new brands can have few or no questions yet: try a smaller maxQuestionsPerTarget first.
  • A brand-name target counts a cited domain as yours when its name is the brand, or the brand is one hyphen-separated part of it (pipedrive matches pipedrive.com and blog.pipedrive.com, not mypipedriveguide.com). Use the domain for an exact match.
  • Not affiliated with OpenAI or Google.

Run summary

Each run stores a summary in the key-value store (OUTPUT): targets searched, targets with no question, questions billed, and anything skipped because of the free-plan or budget limit.

Use it from code, automations or AI agents

API (one call, results returned directly):

curl -X POST "https://api.apify.com/v2/acts/plainrows~ai-citations-finder/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{"targets": ["pipedrive.com"], "maxQuestionsPerTarget": 100}'

n8n, Make, Zapier: use the Apify integration, pick plainrows/ai-citations-finder, schedule it monthly and send the rows to Google Sheets or Slack.

AI agents (MCP tool): add https://mcp.apify.com?tools=plainrows/ai-citations-finder to Claude, Cursor or any MCP client, then ask "Which ChatGPT questions mention pipedrive.com?".

FAQ

What are LLM mentions? The times a large language model (ChatGPT, Google's AI Overviews) names a brand or cites a website in its answer. This Actor lists them question by question.

Do I need to write prompts? No. The questions come from a database of real AI questions and answers: you only give your domain or brand name.

Why am I charged for a target with 0 questions? The data source bills each search, even when it finds nothing. The "Target searched" fee covers that search; the question fees apply only to questions returned.

Can I get ChatGPT data for another country? Not yet: the source has ChatGPT answers for the United States only. Google AI Overviews works for every country in the list.

Mentioned or cited, what is the difference? targetMentionedInAnswer = the answer text names you; targetCitedAsSource = the answer links to your site as a source. Mentioned but not cited is a content opportunity.

More PlainRows tools

Same pay-per-result pricing, same clean JSON, built to chain together in n8n, Make, Zapier or AI agents.

Support

Missing a field or found a bug? Open an issue on this page. Issues are read and answered.