Google AI Overview Scraper

Try for free

A robust, high-performance utility designed for developer automation, data integration, and AI training. Features built-in captcha bypass, headful/headless browser execution, and proxy support to scrape Google data seamlessly, reliably, and at scale.

Pricing

from $3.90 / 1,000 results

Rating

0.0

(0)

Developer

Coding Frontned

Actor stats

Bookmarked

151

Total users

Monthly active users

7 days ago

Last modified

Reliability contract

Uses one bounded public Google result request per query; no browser fingerprint spoofing or hidden browser state.
Writes a dataset row only when a recognizable AI Overview contains at least 50 characters of answer text.
Distinguishes found, not_present, explicit_unavailable, blocked, unrecognized, and transport-failure boundaries.
Treats normal no-overview and Google’s explicit “AI Overview is not available” response as verified query outcomes.
Holds all rows until every requested query has a verified outcome; challenge, malformed, or transport failures fail atomically.
Never stores raw HTML. Responses are capped at 6 MB and public answer text is capped at 20,000 characters.

The default GOOGLE_SERP proxy is recommended for public Google result pages. AI Overview availability varies by query, locale, time, and Google’s serving decisions.

Input

Field	Type	Default	Description
`queries`	string[]	required	One to 20 unique queries, up to 300 characters each.
`maxItems`	integer	`10`	Process at most this many input queries.
`gl`	string	`us`	Two-letter Google country code.
`hl`	string	`en`	Language code, optionally with country suffix.
`maxRequestRetries`	integer	`1`	Zero to two retries per query.
`requestTimeoutSecs`	integer	`60`	Hard 15–90 second request timeout.
`proxyConfiguration`	object	Google SERP proxy	Standard Apify proxy configuration.

{
  "queries": ["benefits of green tea"],
  "maxItems": 1,
  "gl": "us",
  "hl": "en",
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["GOOGLE_SERP"]
  }
}

Output

The dataset contains only genuine overview records:

{
  "query": "example query",
  "aiOverviewText": "The source-backed overview text returned by Google...",
  "sources": [{ "title": "Example source", "url": "https://example.org/article" }],
  "relatedQuestions": ["A related question?"],
  "searchUrl": "https://www.google.com/search?q=example+query&gl=us&hl=en",
  "scrapedAt": "2026-08-31T00:00:00.000Z"
}

OUTPUT records counts and a per-query status even when the dataset is validly empty. An empty dataset therefore means Google did not serve an overview for the tested query—not that a placeholder row was lost.