Google AI Overview Checker — Text, Bullets & Cited Sources avatar

Google AI Overview Checker — Text, Bullets & Cited Sources

Pricing

from $4.00 / 1,000 google ai overview queries

Go to Apify Store
Google AI Overview Checker — Text, Bullets & Cited Sources

Google AI Overview Checker — Text, Bullets & Cited Sources

Check which Google searches trigger an AI Overview and export its text, bullets and cited sources for each query. USD 5 per 1,000 queries.

Pricing

from $4.00 / 1,000 google ai overview queries

Rating

0.0

(0)

Developer

Muhammad Afzal

Muhammad Afzal

Maintained by Community

Actor stats

0

Bookmarked

1

Total users

1

Monthly active users

8 days ago

Last modified

Categories

Features

Limited permissions

Share

Google AI Overview Checker tells you whether a Google search shows an AI Overview and returns the overview's full text, bullet points, sections and cited sources for each query you enter. It costs USD 5 per 1,000 queries checked (USD 4 on the Gold plan), and it is built for SEO teams, content strategists and AI agents that track how Google's AI answers cite websites.

The Actor reports what Google showed for the query, country and language you choose. It never writes or rewrites an answer: when Google shows no AI Overview, the record says overviewPresent: false. Run it on a schedule to turn it into an AI Overview tracker for your keywords.

What data does the Google AI Overview Checker return?

It returns one dataset record per query with the AI Overview text, its list items, its headed sections and every page it cites.

FieldWhat it contains
queryThe Google search you asked about.
overviewPresenttrue when Google showed an AI Overview for the query, false when it did not.
overviewTextFull AI Overview text, sections separated by blank lines, list items on their own lines. null when there is no overview.
overviewBulletsEvery bulleted or numbered list item in the overview, as plain text.
overviewSectionsThe overview split at its headings: each section has the heading Google showed (or null for the opening text) and its text with list items.
sourcesUp to 20 cited pages in Google's order: title, url, domain, snippet (Google's preview of the page, when it shows one) and source (publisher name, such as "IBM").
statussuccess, or warning when there is no overview or part of it could not be read (see warnings).
warningsNotes such as "Google did not show an AI Overview for this query".
searchUrlThe Google search URL for the query, country and language, to review it in a browser.
countryCode, languageCodeThe Google country and language used.
extractorstructured when an overview was read, none when Google showed none.
googleTitle, errorKept for older integrations; both are null in current records.
scrapedAtWhen the query was checked (ISO 8601).

Each run also writes a SUMMARY record to its key-value store: run status, queries checked, AI Overviews found, any failed or skipped query with the reason, charged events and warnings.

How do I check Google AI Overviews for my keywords?

Enter your searches, one per line, and pick a country and language. Each query takes about a minute because Google builds the full AI Overview first; four queries run at a time, so one query finishes in about 1.5 minutes and 25 queries in about 10 minutes (the default run timeout is 15 minutes).

{
"queries": [
"what is retrieval augmented generation",
"how does a heat pump work"
],
"countryCode": "us",
"languageCode": "en"
}
  • queries: up to 25 searches per run, 2 to 300 characters each. Repeated queries are checked once.
  • countryCode: a two-letter country code such as us, gb, de, in or au (213 countries). Default us.
  • languageCode: a Google language code such as en, de, es, fr or pt-BR. Default en.

Call it from code with the Apify API client:

from apify_client import ApifyClient
client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("muhammadafzal/google-ai-overviews-scraper").call(
run_input={"queries": ["how does a heat pump work"], "countryCode": "us", "languageCode": "en"}
)
for item in client.dataset(run.default_dataset_id).iterate_items():
print(item["query"], item["overviewPresent"], len(item["sources"]))

What does the output look like?

One record per query (text shortened here):

{
"query": "what is retrieval augmented generation",
"status": "success",
"overviewPresent": true,
"overviewText": "Retrieval-Augmented Generation (RAG) is an AI technique that connects Large Language Models (LLMs) to external, authoritative knowledge bases...\n\nHow RAG Works\nThe RAG workflow typically follows four main steps:\n1. Ingestion & Indexing: ...",
"overviewBullets": [
"Ingestion & Indexing: Reference documents are split into smaller chunks, converted into vector embeddings, and stored in a vector database.",
"Retrieval: When a user asks a question, the query is matched against the stored document vectors using semantic search."
],
"overviewSections": [
{ "title": null, "text": "Retrieval-Augmented Generation (RAG) is an AI technique that connects Large Language Models (LLMs) to external, authoritative knowledge bases..." },
{ "title": "How RAG Works", "text": "The RAG workflow typically follows four main steps:\n1. Ingestion & Indexing: ...\n2. Retrieval: ..." }
],
"sources": [
{
"title": "What is RAG? - Retrieval-Augmented Generation AI Explained",
"url": "https://aws.amazon.com/what-is/retrieval-augmented-generation/",
"domain": "aws.amazon.com",
"snippet": "Retrieval-Augmented Generation (RAG) is the process of optimizing the output of a large language model...",
"source": "Amazon Web Services (AWS)"
}
],
"googleTitle": null,
"countryCode": "us",
"languageCode": "en",
"extractor": "structured",
"warnings": [],
"error": null,
"searchUrl": "https://www.google.com/search?q=what+is+retrieval+augmented+generation&hl=en&gl=us",
"scrapedAt": "2026-11-08T06:26:37.492Z"
}

A query without an AI Overview is still a valid answer: overviewPresent: false, overviewText: null, empty lists and the warning "Google did not show an AI Overview for this query."

How much does the Google AI Overview Checker cost?

You pay per query checked, including queries where Google shows no AI Overview, plus a small start fee per run; queries that could not be checked are not charged.

EventFREEBRONZESILVERGOLD and above
Google AI Overview query (per query checked)$0.005001$0.004875$0.00475$0.004
Actor start (per run, per started GB of memory; the default 256 MB is one start)$0.006251$0.006094$0.005938$0.005

Discounts from the FREE price: 2.5% on BRONZE, 5% on SILVER, 20% on GOLD and above. Apify platform usage is included in these prices.

Worked examples on the FREE plan at the default memory:

  • 1 query: $0.006251 + $0.005001 = $0.011252
  • 25 queries: $0.006251 + 25 × $0.005001 = $0.131276

Set a maximum charge for the run in Apify Console or the API; the Actor stops checking queries when the limit is reached and never charges past it.

When should I use this Actor, and when not?

Use it to see whether your keywords trigger a Google AI Overview, which pages Google cites in it, and how both change over time or between countries and languages.

Use something else when:

Use it from an AI assistant

Google AI Overview Checker works from Claude, ChatGPT, Cursor or any other MCP client through Apify's MCP server. Add https://mcp.apify.com/?tools=muhammadafzal/google-ai-overviews-scraper as an MCP server, then ask, for example: "Check whether 'best crm for small business' and 'what is a heat pump' show a Google AI Overview in the US and list the sites each one cites."

FAQ

Does every Google search show an AI Overview?

No. Google decides per query, country, language and moment; informational questions show one most often. When Google shows none, the record says overviewPresent: false, and that check is still charged because it is a real answer.

Is the overview text generated or rewritten by the Actor?

No. The text, list items and sources are what Google showed for the query; nothing is summarized, translated or filled in.

Why do two runs of the same query sometimes differ?

Google rebuilds AI Overviews often and varies them by location and time, so wording and cited sources can change between runs. Schedule the Actor to track those changes.

Which countries and languages can I check?

213 Google Search countries by two-letter code and Google's interface languages such as en, de, es, fr, ja or pt-BR. An unsupported code stops the run before any query is checked or charged. If Google results for a supported country cannot be read at the moment, those queries are listed in SUMMARY and not charged.

What happens if a query cannot be checked?

It is skipped without a charge and listed in the run's SUMMARY record with the reason; the other queries still run. If no query can be checked, the run fails with a message and charges only the start fee.

How do I track AI Overviews over time?

Save your queries in a task, schedule it daily or weekly in Apify Console, and compare overviewPresent and sources between runs.

Yes. Google sometimes wraps a cited link in its own redirect; the Actor reads where it points and stores the real page URL. If Google does not reveal the target, the Google link is kept and the record carries a warning.

Limits and responsible use

  • Up to 25 queries per run; split longer lists into several runs.
  • About a minute per query, four at a time. If you lower the run timeout, queries that no longer fit are skipped without a charge and listed in SUMMARY.
  • Desktop results only, one results page per query.
  • The Actor reads publicly shown Google results. Do not submit personal or confidential queries, and verify important claims at the cited sources; an AI Overview is not professional advice.

Last verified: 2026-11-08