Google Search API

Get Google search results as clean JSON, from SerpData, from $0.17 per 1,000 result pages, more than 10 times cheaper than the most popular Google search scraper on Apify, on every plan. Most searches return in well under a second.

Search any query from any country or city, in any language, over any date range. Get the organic results, or the full results page with People Also Ask, top stories, the local pack, shopping, videos and related searches at the same price.

Why use this Google search scraper?

• Lowest price on Apify. From $0.17 per 1,000 result pages, charged only for pages returned. A search that fails is free.
• Rich results at no extra cost. The full results page costs the same as the organic results.
• Fast. Most searches return in well under a second.
• Any location and language. Search as if from a country or a city, keep only pages from one country or in one language, and set Google's interface language.
• Date ranges and filters. The past hour, day, week, month or year, or dates you choose; file types such as PDF; no spelling correction; SafeSearch.
• Batch or live. Run it in bulk from the Apify Console, or call it like an API.

What data can I extract from Google Search?

Simple mode (organic results)	Rich mode adds
Title	People Also Ask
Link	Top stories
Snippet	The local pack
Position	Shopping
Sitelinks, when the page has them	Videos
Date and rating, when Google shows them	Related searches

Each result page also records the search parameters it ran with.

How to scrape Google search results

1. Click Try for free.
2. Enter one search per line, such as "roman aqueducts".
3. Choose organic results or the full results page, how many pages you want per search, and any location, language or date settings.
4. Click Start, then download the results as JSON, CSV, Excel, XML or HTML.

Input

{
  "queries": ["roman aqueducts", "pont du gard"],
  "mode": "rich",
  "maxPagesPerQuery": 2,
  "gl": "fr",
  "quickDateRange": "m6"
}

Field	What it does
queries	One Google search per line
mode	simple for the organic results, rich for the full results page
maxPagesPerQuery	Pages of ten results per search, up to 10
gl	Two-letter country code, such as gb. Required when uule is set
uule	A city, as a canonical-name Google UULE starting with w+. Set the matching gl too
restrictToCountry	Only pages from this country, such as de. Sent to Google as cr=countryDE
searchLanguage	Only pages in this language, such as de. Sent as lr=lang_de
languageCode	Google's interface language, such as de or pt-BR. Sent as hl
quickDateRange	The past hour, day, week, month or year (h, d, w, m, y), with an optional count: d10 is the past 10 days
afterDate, beforeDate	A date such as 2026-01-31, or a time ago such as 3 months, in UTC. Use these or quickDateRange, not both
fileTypes	File types such as pdf and docx. Each search gets filetype:pdf OR filetype:docx added
includeUnfilteredResults	Also return the similar results Google normally hides (filter=0)
disableAutocorrect	Search for each query as typed, even if Google would correct it (nfpr=1)
safeSearch	Hide explicit results (safe=active)

Output

One dataset item per result page:

{
  "page": 1,
  "searchParameters": { "q": "pont du gard", "type": "search", "engine": "google" },
  "organic": [
    {
      "title": "Pont du Gard",
      "link": "https://en.wikipedia.org/wiki/Pont_du_Gard",
      "snippet": "The Pont du Gard is an ancient Roman aqueduct bridge built in the first century AD to carry water over 50 km (31 mi) to the Roman colony of Nemausus (Nîmes).",
      "position": 1
    }
  ],
  "latencyMs": 369
}

In rich mode, a page also carries peopleAlsoAsk, topStories, places, shopping, videos and relatedSearches when Google shows them. searchParameters lists every location, language and date setting the search used.

How much does it cost to scrape Google search results?

From $0.17 per 1,000 result pages, in either mode, with no start fee and no monthly fee. A page is one Google results page of up to 10 results. You are charged only for pages returned; a search that fails is not charged. The price drops on bigger Apify plans:

Apify plan	Price per 1,000 result pages
Free	$0.20
Starter	$0.19
Scale	$0.18
Business	$0.17

• The free Apify plan includes $5 of usage a month, enough for about 25,000 result pages.
• The Starter plan ($19 a month) covers about 100,000 result pages a month.

Use cases

• SEO rank tracking: see where your pages and your competitors' pages rank, in any country or city.
• Keyword research: collect People Also Ask questions and related searches.
• Competitor monitoring: watch who ranks for the searches you care about, and how that changes.
• Brand and news monitoring: find new pages mentioning a name, with a date range.
• AI and RAG: ground language models in fresh search results.
• Market research: see which businesses, products and sources show up for a topic.

Live HTTP API

This Actor also runs in Standby mode, so you can call it like any API and get the result in the response:

curl "https://serpdata--google-search-api.apify.actor/search?q=roman+aqueducts&mode=rich" \
  -H "Authorization: Bearer $APIFY_TOKEN"

Query parameters:

Parameter	Description
q	The search query. Required.
mode	simple (default) or rich.
start	Offset into the results: 0, 10, 20 … up to 90.
gl	Two-letter country code. Required when uule is set.
uule	Canonical-name Google UULE starting with w+.
hl	Google's interface language, such as de or pt-BR.
lr	Only pages in these languages, such as lang_de. Join up to ten with |.
tbs	Date filter: qdr:d for the past day (h, d, w, m, y, with an optional count such as qdr:m6), or cdr:1,cd_min:1/31/2026,cd_max:3/1/2026 for a range in month/day/year.
filter	0 also returns the similar results Google normally hides.
cr	Only pages from these countries, such as countryDE. Join up to ten with |.
safe	active turns on SafeSearch; off turns it off.
nfpr	1 searches for the query as typed, without spelling correction.

Responses have the same shape as dataset items, without page. Errors return JSON with a message and are never charged. 400 means a parameter is invalid, 402 means the run has reached its maximum cost, and 429 means too many requests at once, so slow down and retry. Any 5xx status means the search could not be completed; it is safe to send the same request again.

Integrations and API access

• No-code: connect runs to Make, Zapier, Google Sheets, Google Drive, Slack, Amazon S3 and others, or trigger anything with a webhook.
• Scheduling: run daily or weekly to track rankings over time.
• Code: use the Apify API, or the apify-client package for Node.js or Python.
• AI agents: call this Actor through the Apify MCP server.

Frequently asked questions

Is it legal to scrape Google search results?

Google search results are public and can be read without an account. Whether you can collect and use them depends on where you are and what you do with the data, so check Google's terms of service and get legal advice for your use case.

What is the difference between simple and rich mode?

Simple mode returns the organic results. Rich mode reads the full results page and adds People Also Ask, top stories, the local pack, shopping, videos and related searches. Both cost the same.

How many results can I get for one search?

Up to 10 pages of ten results, so 100 results per search. Set Pages per query.

Can I search as if from another country or city?

Yes. Set Country to a two-letter code such as gb. For a city, also set City (UULE) to a canonical-name Google UULE.

Can I get only recent results?

Yes. Set Quick date range, such as w for the past week, or set After date and Before date.

What export formats are supported?

JSON, CSV, Excel, XML and HTML, from the dataset of any run.

Current limits

• Up to 100 results (10 pages) per search.
• The knowledge panel is not included.
• Verbatim mode is not supported.

Related Actors

• Google Maps Scraper: places, details and photos from $0.034 per 1,000 places.
• Google Maps Reviews Scraper: every review for any place, $0.01 per 1,000 reviews.
• Indeed Scraper: Indeed jobs from 62 countries, $0.03 per 1,000 jobs.