Google AI Overviews Scraper

What is Google AI Overviews Scraper — AIO Citations & Brand Visibility?

Google AI Overviews Scraper is an Apify Actor for checking Google AI Overview results for one or many plain-text search queries. It returns one dataset record per query with the AI-generated answer text, cited source domains, cited URLs, citation titles, citation positions, citation counts, and optional brand-match detection for a target domain. It works with country, language, and device settings, and it is usable through Apify MCP for AI-agent workflows.

The key output fields are aioPresent, aioText, aioTextBlocks, sources, sourceCount, uniqueDomainCount, brandMatch, source, and serpUrl. Those fields give you a record of what Google showed, which citations appeared, and how the result was produced.

Best fit and connected workflows

This Actor fits workflows that start with a Google query and end with citation review or brand visibility analysis. It routes naturally into these use cases:

• SEO and AEO reporting for Google AI Overview citations and answer text
• Brand visibility checks for a known domain across Google AI Overview results
• GEO analysis where country, language, and device settings matter
• Agent workflows that need structured, one-record-per-query results through Apify MCP

For the next workflow step, these related Actors connect well:

• Use Google Maps Reviews Scraper when a cited entity or query leads into place-level research and reviews
• Use Google Ads Scraper - Live Search Sponsored Ads when the next step is sponsored-result analysis for the same search theme
• Use Google Ads Transparency Scraper - Creatives & Run Dates when ad creative and run-date research belongs in the follow-up workflow

Practical scenario

Maya is a search strategist reviewing the query suitable CRM for startups with targetDomain set to apify.com. She runs the Actor with countryCode set to us, languageCode set to en, and device set to desktop.

The returned record gives her:

• aioPresent, which shows whether Google rendered an AI Overview
• aioText, which contains the answer text
• sources, which lists citation position, domain, URL, and title
• brandMatch, which shows whether apify.com appeared in the citations
• sourceCount and uniqueDomainCount, which summarize citation volume and diversity

Her next action is to review the cited pages, compare citation position, and decide whether to update the target page, add supporting content, or report the visibility result to stakeholders.

Input

Use plain-text queries. One record is returned per query, and the Actor accepts up to 50 queries per run.

Field	Type	Description
queries	array[string]	Search queries to check for Google AI Overviews. One record per query is returned.
targetDomain	string	Optional brand domain or comma-separated domains to match in citations.
countryCode	string	ISO 3166-1 alpha-2 country code for Google SERP geo-targeting.
languageCode	string	ISO 639-1 language code for the Google interface.
device	string	desktop or mobile viewport emulation.
maxRetries	integer	Retries per query.
proxyConfiguration	object	Apify proxy configuration.

Valid input example

{
  "queries": [
    "suitable CRM for startups",
    "what is generative AI"
  ],
  "targetDomain": "apify.com",
  "countryCode": "us",
  "languageCode": "en",
  "device": "desktop",
  "maxRetries": 2,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Output

Each dataset item is one query result.

Field	Type	Description
query	string	Search query checked
countryCode	string	Country used
languageCode	string	Language used
device	string	Emulated viewport
serpUrl	string	Google SERP URL fetched
aioPresent	boolean	Whether an AI Overview rendered
aioText	string | null	Concatenated AI Overview answer text
aioTextBlocks	string[]	Text blocks in render order
sources	array	Citation records in appearance order
sourceCount	integer	Total cited sources
uniqueDomainCount	integer	Distinct cited domains
brandMatch	object | null	Brand-match result when targetDomain is set
source	string	The routing options listed in the input schema
scrapedAt	string	ISO 8601 timestamp
warnings	string[]	Non-fatal warnings

Illustrative output record

{
  "query": "suitable CRM for startups",
  "countryCode": "us",
  "languageCode": "en",
  "device": "desktop",
  "serpUrl": "https://www.google.com/search?q=suitable+CRM+for+startups&hl=en&gl=us&num=10",
  "aioPresent": true,
  "aioText": "AI Overview answer text...",
  "aioTextBlocks": [
    "AI Overview answer block one...",
    "AI Overview answer block two..."
  ],
  "sources": [
    {
      "position": 1,
      "domain": "example.com",
      "url": "https://example.com/page",
      "title": "Example source title"
    }
  ],
  "sourceCount": 1,
  "uniqueDomainCount": 1,
  "brandMatch": {
    "targetDomain": "apify.com",
    "matched": false,
    "position": null,
    "citedUrl": null,
    "matchCount": 0
  },
  "scrapedAt": "2026-05-29T14:32:11.000Z",
  "warnings": []
}

How it works

The Actor takes one or more Google search queries, builds the SERP URL from the query plus country and language settings, and fetches the page using an anti-detect browser path with Apify residential proxies by default. It parses the AI Overview container when it is present, extracts the answer text, individual text blocks, and cited sources, and can also check cited domains against targetDomain.

If you provide a search API key, the Actor can retry queries through the BYOK fallback. You can also route every query directly through the managed search API.

Pricing

This Actor uses pay per event with Apify platform usage included. The primary charge is per Google AI Overview query checked. There is also a start event, and an additional brand-match event applies when a target domain matches citations. If the hybrid fallback uses a managed search API infrastructure, that path is charged separately. For the current event details and Apify platform usage context, open the live Pricing tab in Apify Console.

Example in words: if you run ten queries and one of them matches your brand domain, you are billed for ten query checks, one brand-match event, plus the Actor start event and any applicable Apify platform usage.

Free plan limit: Apify free-plan accounts can make 10 managed fallback search requests per rolling 30 days across this developer's search Actors. A fallback request is made only when the direct browser route is blocked, and runs that use your own key are not limited. This limit is set by the Actor developer, not Apify. When it is reached, the run stops making fallback requests, keeps every result already saved, finishes cleanly, and the status message shows when the next free request becomes available. Paid Apify plans have no such limit.

Use with AI agents (MCP)

This Actor is built for agent workflows through Apify MCP. It exposes a structured tool that accepts Google search queries and returns a dataset item per query with AIO answer text, citations, citation positions, and brand-match fields.

Tool description: check Google AI Overview presence for a query, extract the answer text, cited sources, citation positions, and optional brand-match information, then read the dataset item produced for that query.

Actor identity: khadinakbar/google-ai-overviews-scraper

Check the Google AI Overview for "suitable CRM for startups" in the US desktop SERP, return the cited sources, and tell me whether apify.com appears in the citations.

Output interpretation:

• aioPresent tells you whether an AIO was rendered
• sources gives citation order, domain, URL, and title
• brandMatch shows whether the requested domain was cited, plus the first-match position and matched URL
• source indicates whether the record came from the routing options listed in the input schema
• serpUrl preserves the exact Google page fetched for provenance

Provenance and scope:

• One dataset record is returned per query
• query, countryCode, languageCode, device, serpUrl, source, and scrapedAt provide the run context
• warnings can carry non-fatal parse notes when present
• brandMatch appears only when targetDomain is set

Pagination and cost guidance:

• Batch multiple queries when you want a compact report, up to the 50-query input cap
• Each query produces one dataset record
• Cost follows the event model, so the query list should match the reporting scope you need

API example

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({
  token: process.env.APIFY_TOKEN,
});

const run = await client.actor('khadinakbar/google-ai-overviews-scraper').call({
  queries: ['suitable CRM for startups', 'what is generative AI'],
  targetDomain: 'apify.com',
  countryCode: 'us',
  languageCode: 'en',
  device: 'desktop',
});

const dataset = await client.dataset(run.defaultDatasetId).listItems();
for (const item of dataset.items) {
  console.log(item.query, item.aioPresent, item.brandMatch);
}

Best results and outcome guidance

Use plain search queries rather than full Google URLs. Set countryCode, languageCode, and device to match the audience you are analyzing, since AIO content can vary by region and viewport. Add targetDomain when your workflow needs citation-based brand visibility. Review sourceCount, uniqueDomainCount, and sources together, since they describe both volume and diversity of citations.

Continue the workflow

• Then use AI Overviews Tracker — Brand Visibility in AI Answer Boxes to extend Google AI Overviews Scraper — AIO Citations & Brand Visibility research with a complementary monitor contract.
• Then use Google SERP All-in-One Scraper — AIO, PAA & Knowledge Panel to extend Google AI Overviews Scraper — AIO Citations & Brand Visibility with a neighboring search visibility work source when the brief calls for Google Serp data.

Design note

I found that the live output contract always includes query, aioPresent, source, and scrapedAt, which keeps every record useful even when aioText is empty.

FAQ

When should I use this Actor instead of a broader Google scraper?
Use it when your workflow is specifically about Google AI Overview presence, answer text, and citations.

How does targetDomain work?
It compares your domain or comma-separated domains against the cited source domains in each query result.

What if I need mobile and desktop reporting?
Run the same query set twice with device set to mobile and desktop to compare results.

Can I use this in an agent workflow?
Yes. The Actor is MCP-ready and designed to return structured records that agents can read and summarize.

Where do I inspect run-level usage details?
Check the dataset for records and the key-value store records named SUMMARY, OUTPUT, and RUN_SUMMARY.

How does this connect to related Google workflows?
Use the Google Ads or Google Maps related Actors when your next step is sponsored result analysis, ad transparency research, or place-level follow-up after a Google query check.

Responsible use

Use this Actor for public Google search result analysis and citation research. Review your use of Google search data against applicable terms, local laws, and your organization's policies. Keep query volume aligned with the reporting task, and choose the country, language, and device settings that reflect the audience you are measuring.