Google Search Results Scraper - SERP, Ads & People Also Ask

Google search results scraper for your own searches, results pages back as rows: organic listings with their rank, the ads above them, the People Also Ask questions and the related searches at the bottom. One search or a list of up to 100, with the country, language, Google domain, dates and pages you choose.

The url on an organic row is the address of the page itself. Now and then Google won't give one up. That row still arrives, with url and domain set to null, and it is not billed.

Input	Google searches, typed or pasted as Google search addresses, operators like site:, quotes and OR included
Output	One row per organic result, ad, People Also Ask entry or related search, with rating, price range and the bold snippet words where Google prints them
Ceiling	100 rows per search, across up to 10 pages; 100 searches per run
Account needed	None, and no Google API key
Price	$1.00 per 1,000 organic results on the Free plan, a little less on paid plans. Ads, People Also Ask and related searches come free. The free plan's $5 a month covers about 5,000

🔍 What Google Search Results Scraper does

It runs each search against the Google domain you name, reads the results page, and turns it into rows.

Organic rows carry the title, the page's address, the displayed source, the snippet and the rank, plus the snippet's bold words and, on shop and review results, the star rating, review count and price range. Ad rows carry the same shape for the paid results Google marks as such, without the advertiser's address: Google prints only its own click link for an ad and we don't follow it, so url and domain are null there. People Also Ask rows carry the question. Related search rows carry one phrase each from the "People also search for" block.

Set maxPages above 1 and it follows Google's own Next link, page by page, stopping when your row limit is reached or Google runs out of pages. When Google runs out first, a free NO_MORE_PAGES row says how many pages it showed. Duplicate links are dropped across pages.

Filters ride on the same request: dates, results language, sites from one country, a city, and fields that write site:, intitle: or filetype: for you.

📋 What data you get from each Google search result

What you get	Field
Which kind of row it is	type
The listing: title, the page's address and its host, the source Google displays, the snippet	title, url, domain, displayedUrl, snippet
Position within its page, which page, and the next page Google offers	rank, page, nextPage
The snippet's bold words	snippetHighlights
Star rating, review count and price range, when the result shows them	rating, reviewCount, ratingSource, priceRange
Sitelinks under a result (with includeSitelinks)	sitelinks
People Also Ask: the question and the answer text	question, answer
Related search phrase	title on related_search rows
Product cards (with includeProducts)	title, price, oldPrice, seller, moreSellers, rating, reviewCount
Sites the AI Overview cites (with includeAiOverviewSources)	source, title, url, domain
Optional add-on: each result page's readable text	pageText
The search, its settings and the Google address read, on every row	query, countryCode, language, googleDomain, searchUrl
When the page was read	fetchedAt

▶️ How to scrape Google search results

1. Open Google Search Results Scraper and click Try for free.
2. Put your search in Search query, replacing the example, and any others in More searches.
3. Set Maximum results. Ten is a sensible first run.
4. Change Google domain, Country code and Language if you want another country's view, then click Start.
5. Download the dataset as JSON, CSV or Excel, or read it from the Apify API.

An empty query writes one labelled sample row and stops, so you can see the shape first.

💰 How much does it cost to scrape Google search results?

$1.00 per 1,000 organic results on the Free plan. Paid plans pay a little less: $0.99 on Starter, $0.98 on Scale and $0.97 on Business. On the free plan, the $5 Apify gives you each month covers about 5,000 organic results.

Only organic results are billed. Ads, People Also Ask entries, related searches, product cards and AI Overview sources arrive free beside them. maxResults caps each search, so it also caps what a search can bill: set it to 10 and ten organic results is the most one search can be charged for.

The sample row, every diagnostic row, an organic result that came without its address, a result left out for falling outside your dates or your site, and a search that finds nothing are not billed. Page text is an optional add-on, off unless you switch it on.

📥 What you give it

{
  "query": "best running shoes",
  "queries": ["trail running shoes", "running shoes for flat feet"],
  "countryCode": "US",
  "language": "en",
  "googleDomain": "google.com",
  "timeRange": "month",
  "maxResults": 20,
  "maxPages": 2
}

Field	Default	What it is
query	none, the box starts at best running shoes	Your search. site:, intitle:, quoted phrases, OR and minus terms all work. A site: search reads Google's plain web results, so it brings no People Also Ask or related searches, and an organic result from another site is left out and not billed.
queries	none	More searches, one per line. Each is read on its own, with its own pages and maxResults. 100 searches per run in all.
searchUrls	none	Google search addresses from your browser. The search and its tbs, lr, cr, uule, gl, hl and start are read from each. Only the All tab; only Google's own addresses.
googleDomain	google.com	The Google host, without https://. google.co.uk, google.ca and the rest are fine.
countryCode	US	Two-letter country, sent as Google's gl. A code Google does not know stops the run before any search, with a free BAD_INPUT row.
language	en	Interface language, sent as Google's hl, such as de or pt-BR. Checked the same way.
resultsLanguage	none	Only pages written in this language (Google's lr).
resultsCountry	none	Only sites Google places in this country, such as FR (Google's cr). Different from countryCode, which is where you search from.
location	none	Search as someone in this place: Google's name for it, like Chicago,Illinois,United States, or a uule code.
timeRange	none	hour, day, week, month or year.
dateFrom, dateTo	none	A date range, YYYY-MM-DD. Not together with timeRange. A result whose snippet date is clearly outside the dates is left out and not billed.
exactMatch	off	Wraps each plain search in quote marks. Searches with quotes, operators or OR are sent as they are.
site, wordsInTitle, wordsInText, wordsInUrl	none	Written into the search as site:, intitle:, intext: and inurl:.
fileTypes	none	Written in as filetype:pdf OR filetype:xlsx and so on.
maxResults	10	Rows per search, 1 to 100, counting organic results, ads and People Also Ask together; the related phrases take one place between them. It is also the most a search can bill you.
maxPages	1	Follow Google's pagination this far, 1 to 10.
startPage	1	The first page to read, 1 to 10. A run never reads past page 10.
includeAds	on	Return the paid results Google marks as ads.
includePeopleAlsoAsk	on	Return the People Also Ask questions, with answer text when the page carries it.
includeRelatedSearches	on	Return the related phrases from the foot of the page, one row each.
includeSitelinks	off	Fill sitelinks on organic rows. Their addresses are filled on every plan.
includeProducts	off	Free product rows for the shopping cards Google prints, text only, on top of maxResults.
includeAiOverviewSources	off	Free ai_overview_source rows for the sites the AI Overview cites, on top of maxResults. Their addresses are filled on every plan. The overview's own text is never returned.
includePageText	off	Optional add-on: each organic result's page read once, its text in pageText, up to 20,000 characters.
notionConnector, notionParentId	none	Write each row as a page into your Notion through an Apify connector. The dataset has them either way.
enableProxyFallback	on	On by default. Switched off, runs can come back with no results.
proxyConfiguration	leave it alone	Optional. Your own servers are used exactly as given.

📤 What you get back

A real organic row from a run on 11 October 2026:

{
  "ok": true,
  "type": "organic",
  "title": "Best of Road Running Shoes",
  "url": "https://believeintherun.com/shoe-reviews/type/best-of-road/",
  "domain": "believeintherun.com",
  "displayedUrl": "Believe in the Run",
  "snippet": "Independent, run-tested reviews of running shoes, Popular Brands New Balance Nike Asics Hoka Saucony Puma Adidas. These are the Best On Running Shoes Right Now ...",
  "rank": 4,
  "page": 1,
  "query": "best running shoes",
  "countryCode": "US",
  "language": "en",
  "googleDomain": "google.com",
  "fetchedAt": "2026-10-11T18:43:56.264Z"
}

Field	How to read it
url	The page's own address. null on a result Google gave no address for, and that row is not billed.
domain	The host of url, without www.. null when url is.
rank	Position within its own page, so page 2 starts counting again.
answer	People Also Ask rows only, and usually null: on our test runs none of the 12 questions came with answer text.
query	The search as sent to Google, with any operator fields written in. On a related_search row the phrase is in title.
snippet	Google's own description text, trailing Read more and all.
rating, reviewCount	Numbers only. ratingSource says when Google labels it, such as store rating. null on most results.
nextPage	The page number Google offers next, ready for startPage; null on the last page.
transport	first_try when Google answered the first request for the page, retry when a later try did. Rows and price are the same either way. Runs before 25 October 2026 wrote direct for a first try, and runs before 24 October wrote longer values in place of retry.

🧾 Reading the output

Row	How to spot it	Billed as a result
An organic result	ok: true, a type of organic and a url	yes
An organic result Google gave no address for	a type of organic and url set to null	no
An ad, People Also Ask entry, related search, product card or AI Overview source	ok: true and a type of ad, people_also_ask, related_search, product or ai_overview_source	no
The sample row	_sample: true	no
A diagnostic	ok: false and an errorCode	no

Filter on ok being true with _sample absent. For the billed results alone, keep the organic rows that have a url.

Code	What it means
BAD_INPUT	A setting could not be used: a country or language code Google does not know, a malformed date, an address that is not a Google search. problems lists them. Nothing was searched.
NO_RESULTS	The page loaded and nothing usable was on it. A query with no matches lands here.
NOT_SERVED	This search page was not served this run. Re-run it later. Runs before 13 October 2026 wrote BLOCKED or JS_REQUIRED here.
RATE_LIMITED	Google answered with a rate limit. Wait and re-run.
NETWORK	Google was unreachable or answered badly. A googleDomain that is not a real Google host also lands here rather than saying the value is wrong.
TIME_LIMIT	The run got close to its time limit and stopped on its own. The results delivered before that are complete; searchesNotRead names any search it did not reach. Give the run more time, or ask for fewer pages, to get the rest.
INCOMPLETE	Google did not return a later page (failedPage says which). The results of the pages before it are delivered and billed as usual; nothing is billed for the missing pages.
NO_MORE_PAGES	You asked for more pages than Google showed (pagesRead and pagesAsked give both numbers). Everything read is delivered and billed as usual; this row itself is free.
OFF_QUERY_PAGE	Google answered a later page with results for a different search, and again when asked a second time. That page is left out and not billed, and the run carries on with the next one. offQueryPages lists every page left out (offQueryPage is the first). Two such pages in a row end the run there.
OFF_SITE_RESULTS	A site: search came back with organic results from other sites mixed in. Those were left out and not billed; the results on the site are delivered as usual. offSiteResults is how many, offSitePages on which pages.
OUTSIDE_DATES	Google returned results dated outside the dates you asked for. Those were left out and not billed. outsideDates is how many.
PARTIAL	An extra was not read: further pages of a search whose results kept falling outside your dates, links of free rows, or page text. It says which; nothing was billed for them.

A diagnostic never fails the run, TIME_LIMIT included: a run the time limit stopped before any result came back still ends as succeeded, with an amber status message.

The Overview table hides the newer columns and the diagnostic fields. Switch the dataset to JSON to read them.

💡 What people use it for

• Tracking where a page ranks for a list of keywords, week by week, with rank and page already split out.
• Pulling the People Also Ask questions and related searches for a topic to see what the audience actually asks.
• Checking which advertisers sit above the organic results for a commercial term in one country.
• Running the same search across google.com, google.co.uk and google.de, or from one city, to compare.

From a seed word to who ranks for it, in three runs:

1. Run Keyword Autocomplete Scraper on a seed with expandQuestions on, and keep the suggestion values you care about.
2. Put the best few into Google Trends Scraper's searchTerms with compareKeywords on, to see which one is growing.
3. Paste the winners here into queries to see who ranks for them today.

🚧 What it does not do

• With a small maxResults, ads are taken first. Cap a run at 5 on an advert-heavy term and ads can use most of the budget. Switch includeAds off if you only want organic rows.
• No local pack, news box or knowledge panel, and none of the AI Overview's own text: only the sites it cites.
• Filters are Google's own. It can ignore one; dates are checked against the snippet, but language, country and city are taken as Google answers them.
• Ads and People Also Ask are spotted from how Google marks them up, so a layout change on Google's side can leave some of them out until the reading catches up.
• Rows are a snapshot of that page at that moment. Rankings move daily.
• It is not an official Google tool. Dami's Studio is independent and is not affiliated with or endorsed by Google, or by any other company named on this page.

🧭 Which Google scraper do you need?

If you want	Use
The web results page for a keyword: organic, ads, People Also Ask	This one
What a search box suggests as people type	Keyword Autocomplete Scraper
Interest in a keyword over time and by region	Google Trends Scraper
News articles with the publisher's real link	Google News Scraper
Image results and the pages they sit on	Google Images Scraper
The ads a brand runs on Google	Google Ads Transparency Scraper
The same job on Yandex, by region	Yandex Search Engine Scraper
Businesses and places	Google Maps Scraper

❓ Questions people ask

Do I need a Google API key to scrape Google search results?

No. Nothing to apply for, no quota to watch.

Can I search more than one keyword at a time?

Yes. Put up to 100 searches in queries, or paste Google search addresses into searchUrls. Each one gets its own pages and its own maxResults, and the rows say which search they came from.

Why don't the rankings match what I see on Google?

Google tailors results to where you are, your language and your history. The run sees what a logged-out visitor sees for the googleDomain, countryCode and language you set. To compare, search in a private browser window with the same Google domain.

Why did I get fewer rows than maxResults?

Google decides how much it prints. A narrow query, a date filter, or site: on a small domain often runs out before your limit does. If you asked for more pages than it showed, a free NO_MORE_PAGES row says so.

Can I call it from code or connect it to an AI assistant?

The API tab has code for Python, JavaScript and the CLI. MCP clients like Claude and ChatGPT use https://mcp.apify.com/?tools=fetch-actor-details,dami_studio/google-search-results-scraper; coding assistants can load our skill file. All three bill your Apify account at the same price.

Is scraping Google search results legal?

These are public search results. Apify's write-up on the legality of web scraping is a good starting point, and we are not lawyers.

🆘 If something breaks

Open the Issues tab on the actor page. Send the query and the run ID. The errorCode on the diagnostic row usually names the problem on its own.