Google Search Scraper — SERP, AI Overview, Ads

What does Google Search Scraper do?

Scrape Google Search (SERP) results at scale — organic listings, paid ads, AI Overview, knowledge graph, People Also Ask (with answers on request), local pack, news, shopping, perspectives, forums, short videos, related searches, inline videos — from Google. No CAPTCHAs, no TLS fingerprint tuning, no browser management.

Why use Google Search Scraper?

• Every SERP block. Organic, ads, AI Overview, knowledge graph, PAA, local pack, news, related, videos — one call returns them all.
• 200+ Google domains. google.com, google.co.uk, google.de, google.co.jp, etc. + gl / hl / device targeting.
• Fast and full modes. fast uses Google's lite endpoint (≈40% faster, organic-only). full returns the rich SERP.
• AI Overview chase. Optional follow-up fetch for Google's deferred AI Overview page_token.
• People Also Ask answers. Set expand_related_questions to open the first N questions during the search and get each answer with its sources.
• No anti-bot work. ScrapeBadger's residential proxy + browser-farm fallback handles SearchGuard, reCAPTCHA, consent walls.

What data can Google Search Scraper extract?

Field	Type	Description
organic_results	array	Per-result title, link, snippet, position, rich metadata
ai_overview	object	Google's AI-generated answer (optional chase)
knowledge_graph	object	Entity card: title, description, website, social links, images
related_questions	array	People Also Ask questions; with expand_related_questions, each answered one carries text_blocks and references
related_searches	array	Refined searches at the foot of the page
shopping_results	array	Products in the page's shopping carousels, with prices, store and product IDs
perspectives	array	"What people are saying" cards: articles, videos and social posts
discussions_and_forums	array	Forum threads (Reddit, Quora …) with answers where Google shows them
short_videos	array	Short-video clips (YouTube Shorts, TikTok …)
things_to_know	array	"Things to know" topics and their questions
ads	array	Paid ad tiles (sponsored links)
news_results	array	Top news section
local_results	array	Local pack (Google Maps inline results)
inline_videos	array	Video tiles inside the SERP
pagination	object	Next-page tokens for programmatic pagination

How to scrape Google Search

1. Click the Try for free button above.
2. Enter your search query in the Search query input (supports operators like site:, intitle:, "quoted", -exclude).
3. Optional: set country (gl), language (hl), domain, device, city-level location, time filter (tbs), safe search.
4. Pick mode: full for every SERP block (2 credits/page) or fast for organic-only (1 credit/page).
5. Set max_pages to paginate — the actor stops early when Google reports no more pages.
6. Click Start — results stream into the dataset tab, one record per SERP page.

How much will it cost?

$0.002 per SERP page fetched (mode=full), $0.001 per page (mode=fast). One dataset record per SERP page. max_pages: 3 → 3 records → $0.006 full / $0.003 fast. With expand_related_questions, each answered People Also Ask question is charged as one more full page: one page with 2 answered questions → $0.006. Apify compute units are extra; a typical 3-page run uses less than one CU.

Competitor benchmark

Actor	Author	Price	Notes
apify/google-search-scraper	Apify	~$3.50 / 1k results	Official Apify actor
scrapestorm/google-search-api-scraper	ScrapeStorm	~$2.00 / 1k results	Organic only, no AI Overview
compass/google-search-results	Compass	~$2.50 / 1k results	Per-result pricing
scrape-badger/google-search-scraper	ScrapeBadger	$2.00 / 1k full · $1.00 / 1k fast	All SERP blocks + AI Overview

Input

Configure the run in the Input tab above, or pass a JSON object matching the fields below when calling the Actor via the Apify API.

Field	Required	Description
q	✅	Search query — supports Google operators.
mode	—	full (default) or fast.
gl	—	Country code (default us).
hl	—	Language code (default en).
domain	—	Google domain (default google.com).
device	—	desktop or mobile.
location	—	City-level geo (e.g. New York, NY).
num	—	Results per page (1-100, default 10).
start	—	Page offset (0, 10, 20 …).
max_pages	—	How many pages to fetch (1-10, default 1).
safe	—	off / medium / high.
tbs	—	Time filter (qdr:d, qdr:w, qdr:m, qdr:y).
lr	—	Language restrict (e.g. lang_en).
nfpr	—	Disable auto-correction.
ai_overview	—	Chase deferred AI Overview (forces mode=full).
expand_related_questions	—	Open the first N People Also Ask questions and return their answers (forces mode=full; each answered question is charged as one more full page).

Output

Every successful run streams records into the run's dataset. Download as JSON, CSV, XML, Excel, or HTML from the Dataset tab; consume programmatically via the Apify API or webhooks.

Example record:

{
  "search_parameters": {
    "q": "best coffee shops near me",
    "gl": "us"
  },
  "organic_results": [
    {
      "position": 1,
      "title": "Top 10 coffee shops in NYC",
      "link": "https://\u2026",
      "snippet": "\u2026"
    },
    {
      "position": 2,
      "title": "Local coffee near you",
      "link": "https://\u2026",
      "snippet": "\u2026"
    }
  ],
  "ai_overview": {
    "text_blocks": [
      {
        "type": "paragraph",
        "snippet": "\u2026"
      }
    ],
    "references": []
  },
  "pagination": {
    "next_page_token": "\u2026"
  }
}

Tips / Advanced options

• Use mode=fast when you only need organic results. Half the price, 40% faster. Only upgrade to full when you need AI Overview, knowledge graph, local pack, or news.
• Batch with the scheduler. Apify's scheduler lets you run one query at a time on a cron. For bulk, start multiple runs in parallel.
• Filter by time. tbs: qdr:d for past 24h — perfect for brand monitoring.
• Combine with google-autocomplete-scraper. Expand seeds into long-tail keywords first, then SERP each one.

FAQ, Disclaimers, Support

Does this actor bypass Google CAPTCHAs?

Yes. ScrapeBadger's residential proxy + browser-farm fallback handles SearchGuard, reCAPTCHA, and consent walls transparently. You'll never see a CAPTCHA in the output.

Can I target a specific city?

Yes — set location to a city string like New York, NY. For more precision use a Google UULE-encoded value in uule (SDK-level).

What's the max num per page?

Google enforces 1-100. The actor clamps into that range.

Why do I get fewer results than max_pages suggests?

The actor stops early when Google reports no next page, saving credits. This is intentional.

Can I scrape Google Jobs / Maps / News via this actor?

No — those have dedicated actors: google-jobs-scraper, google-maps-scraper, google-news-scraper. This one is pure SERP.

Disclaimer

This Actor scrapes public Google data only. You're responsible for compliance with Google's Terms of Service and any applicable data-protection laws (GDPR, CCPA, etc.) in your jurisdiction. ScrapeBadger does not store the scraped results — they are delivered directly to your Apify dataset.

Support

Something not working? Open a ticket in the Issues tab above — we triage within one business day. Full API reference: docs.scrapebadger.com.

Related Actors

• google-maps-scraper — Places, reviews, photos
• google-news-scraper — Search + topic + trending
• google-autocomplete-scraper — SEO seed expansion

Powered by

ScrapeBadger — Google-optimised residential proxy pool + browser-farm fallback, 99.7% uptime, unmetered bandwidth. No CAPTCHAs reach you.

Targeting and advanced filters

Use location: "New York,New York,United States" for a canonical Google location. The API resolves it to UULE; do not combine it with uule or lat/lon. Coordinates must be supplied together. radius is accuracy in meters, with a maximum of 199 for desktop and 1000 for other devices.

The Input tab also supports google_domain, color_scheme, no_cache, udm, tbm, and Google's as_* advanced query controls. no_cache: true requests a fresh fetch and is billed normally. JSON results include search_parameters with the actual Google query parameters and search_metadata with safe fetch diagnostics. Requested location is not proof of Google's observed location.

output: "html" emits a record containing html and stops after one page. Full JSON supports web results and udm: "18" forums; other vertical selectors require HTML output or the dedicated Google actors. Fast mode supports web only.

Google request parameters

The optional oq, sourceid, and ie inputs override the original query, source identifier, and input encoding hint sent to Google. The defaults are the current query, chrome, and UTF-8. Pass an empty string to omit one of these fields.