Google Search Results Scraper (SERP) - No Login, Per Page avatar

Google Search Results Scraper (SERP) - No Login, Per Page

Pricing

from $7.77 / 1,000 results pages

Go to Apify Store
Google Search Results Scraper (SERP) - No Login, Per Page

Google Search Results Scraper (SERP) - No Login, Per Page

Scrape Google search results for your keywords: organic positions, titles, URLs, snippets, dates, sitelinks and related searches. Pick country and language, up to 10 pages per query. One row per result. No login.

Pricing

from $7.77 / 1,000 results pages

Rating

0.0

(0)

Developer

Cliqto Media

Cliqto Media

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

0

Monthly active users

an hour ago

Last modified

Categories

Features

Limited permissions

Share

Add search queries, pick a Google country and language, and get the public Google search results (SERP) as one row per result: organic results and related searches, each with its position. No Google login, no API key; you pay per loaded results page, about $0.015 for the first page of one query.

Use it for rank tracking, SEO research and competitor checks. The smallest run is the prefilled query apify web scraping with one page; it finishes in under a minute.

Google Search Results Scraper: queries in, organic results and related searches out, one row per result

Table of contents

What it does

The Actor opens the public Google results page for each query, in the country (gl) and language (hl) you choose, and reads up to 10 result pages per query. It saves:

  • organic results: position, title, URL, displayed URL, snippet, date and sitelinks (when Google shows them);
  • related searches: the "People also search for" list, once per query.

Optional filters work like Google's own search tools: Time range (past hour, day, week, month or year) and Results language (only pages in one language).

Text ads appear rarely (in recent test runs 2–3 ad rows per 1,000 rows); includeAds keeps or drops them. The pages show no shopping prices, so the field price stays in every row for compatibility (always null). Sitelinks (the short links under some results) are saved in the sitelinks list of that result and are not separate rows; the list is empty when Google shows none.

Every row has the query, page number and status, so you can filter and sort in the Dataset, in a spreadsheet or by API.

Good fit

  • Track where a site ranks for a list of keywords in one country.
  • Collect top 10–100 URLs for a topic for SEO or content research.
  • See which shops and brands rank for product queries.
  • Find related searches to grow a keyword list.

Not a fit / limits

  • The Actor reads Google's HTML results page. Ads are rare. It does not return shopping prices, People Also Ask, AI Overviews, featured snippets, maps or image packs.
  • Results are not personalised and not tied to a city. A Google Ads account, Search Console data or search volume are not included.
  • Google often answers automated traffic with a CAPTCHA. The Actor never solves CAPTCHAs: a refused page is loaded again through Apify Proxy GOOGLE_SERP, a proxy made for Google results pages (see How it works).

👉 Try it now: open the input form

How it works

  1. For each query and page, the Actor first asks Google for the simple results page through Apify datacenter proxy (or your proxy). This light route is fast.
  2. When Google refuses that page twice (CAPTCHA), the Actor loads it through Apify Proxy GOOGLE_SERP, a proxy made for Google results pages. It takes 10–20 seconds per page. How many pages go this way depends on Google and changes from hour to hour: in recent test runs it was from 2 of 10 pages to every page.
  3. After 3 such pages in a row, the whole run stays on GOOGLE_SERP and tries the light route again every 25 pages.
  4. If GOOGLE_SERP fails 3 times in a row too, the run sends no more requests through it: each query it stops gets a free error row, and later pages go over the light route if Google lets them (when you chose the GOOGLE_SERP group, the run stops paid requests).
  5. It saves one row per organic result and related search, plus a run report RUN_SUMMARY in the key-value store. RUN_SUMMARY.http shows how many pages went each way (cheapPages, serpProxyPages). The price per page is the same on both routes.

👉 Try it now: open the input form

Quick start

Input form: search queries (box 1), country and language (box 2), pages per query (box 3)

  1. Open the Input tab. In Search queries (queries), keep the prefilled apify web scraping or add your own queries, one per line (box 1).

  2. Keep Country us and Language en (defaults), or choose others (box 2).

  3. Keep Pages per query (maxPagesPerQuery) at 1 (prefill), or set up to 10 (box 3).

  4. Click Start, then open the Output tab (Dataset) and check the columns type, position, title and url (box 4).

    Output table of the prefill run: type, position, title and url columns (box 4)

Expected result: usually 7–10 organic rows and about 8 related searches for each query page.

👉 Try it now: open the input form

Use cases

Buyer jobInput pathOutput entityKey fieldsNext integration
Rank trackingqueries, countryCode, maxPagesPerQueryorganic rowquery, position, url, pageSchedule + Google Sheets or API
SEO content researchqueries, languageCodeorganic rowtitle, description, displayedUrl, dateExport CSV / JSON
Shop and brand checkqueries, countryCodeorganic rowtitle, url, displayedUrl, descriptionExport to spreadsheet
Keyword ideasqueries, includeRelatedrelated rowtitle, urlFeed back into queries

👉 Try it now: open the input form

Input

Input fields

FieldPlain nameTypeRequiredDefault / prefillAllowed valuesEffectExample
queriesSearch queriesarray of textyesprefill ["apify web scraping"]1–500 queries, up to 256 characters eachOne Google search per query. Operators like site: and "exact phrase" work. Duplicates and extra spaces are removed.["best running shoes", "site:apify.com scraper"]
countryCodeCountrytextnodefault usGoogle country codes, e.g. us, gb, de, fr, jpCountry for ranking (gl)"de"
languageCodeLanguagetextnodefault enGoogle language codes, e.g. en, de, fr, pt-BRInterface language (hl)"de"
maxPagesPerQueryPages per queryintegernodefault 1, prefill 11–10How many result pages per query (usually 7–10 organic results each). Each loaded page is charged. About 500 pages fit in one run; split bigger jobs into several runs.3
includeAdsInclude adstrue/falsenodefault truetrue, falseSaves text ads as ad rows when Google shows them; this is rare.false
includeRelatedInclude related searchestrue/falsenodefault truetrue, falseAdds related search rows once per queryfalse
timeRangeTime rangetextnodefault anyany, hour, day, week, month, yearOnly results Google dates within this period, like Google's Tools → Any time menu (tbs=qdr:)"week"
searchLanguageResults languagetextnoempty (any language)Google results-language codes, e.g. en, de, pt, zh-CNOnly pages written in this language (lr). languageCode changes only Google's interface"de"
proxyConfigurationProxyobjectnoApify datacenter proxyApify proxy groups or your own URLsWhere requests come from on the light route. A page refused twice there is loaded through Apify Proxy GOOGLE_SERP instead. Choose the GOOGLE_SERP group to send every page there{"useApifyProxy": true}
deliveryConnectorSend run summary toMCP connectornoempty (no delivery)a Notion or Slack connector from your Apify accountAfter the rows are saved, posts a short summary of the run (counts, first rows, links to the Dataset and the run)pick it in the form
deliveryTargetNotion page or Slack channeltextnoemptyNotion page link or ID, Slack channel IDWhere the summary goes. Required for Notion: the summary is added at the top of that page"C0123456789"

An empty queries list stops the run with a clear input error and no page charge.

Row limit. The platform Max items setting (or maxItems in API input) caps the result rows the run saves. The run stops when the limit is reached, ends with status COMPLETE and RUN_SUMMARY.limitReached: "MAX_ITEMS", and charges only the pages it loaded. maxResults, limit and resultsLimit work the same way, and query, searchTerms or keywords are read as queries, so inputs written for other Google Search Actors run as they are. Unknown fields are ignored and listed in RUN_SUMMARY.ignoredInputFields.

Interacting settings

WhenInput AInput BEffective behaviorCost effect
Google has fewer pagesmaxPagesPerQuerythe queryThe Actor stops at the last page Google hasOnly loaded pages are charged
Ads on or offincludeAdsany queryOff drops the rare ad rowsFewer rows, same charge
Own proxy or RESIDENTIALproxyConfigurationcountryCodeYour proxy is the light route; refused pages still go through GOOGLE_SERPSame price per page; own proxy URLs are paid by you
GOOGLE_SERP groupproxyConfigurationany queryEvery page goes straight through GOOGLE_SERP, 10–20 s per pageSame price per page

Input examples

Track 3 keywords in Germany, 2 pages each

{
"queries": ["laufschuhe herren", "matratze test", "site:zalando.de sneaker"],
"countryCode": "de",
"languageCode": "de",
"maxPagesPerQuery": 2,
"includeRelated": false
}

Expected behavior: 6 result pages, organic rows only. Expected output: query, page, position, title, url. Cost note: 6 pages + 1 start = $0.071 on the Free plan.

👉 Try it now: open the input form

Output

One Dataset row per result. A query without results gets one not_found row; a query that could not be scraped gets one error row.

Output fields

FieldMeaningTypeCan be empty?Example / enum
queryYour search querytextno"apify web scraping"
pageGoogle result page (1 = first)numbernull on status rows1
typeRow type (ad rows are rare)textnull on status rowsorganic, related, ad
positionPlace among rows of the same type for this query; organic positions continue across pages; a URL that Google repeats on a later page is saved oncenumbernull on status rows1
titleResult title or related search texttextyes"Apify"
urlResult URL; for related, a Google search URLtextyes, when Google hid the link"https://apify.com/"
displayedUrlSite as Google shows it (domain and breadcrumb)textyes"www.apify.com › store"
descriptionSnippet texttextyes
dateDate shown before the snippet, as Google prints ittextyes"Oct 9, 2026"
sitelinksLinks shown under a result, as {title, url}; an empty list when Google shows none. They are not separate rows and do not change positionarrayno[{"title": "Python Introduction", "url": "https://www.tutorialspoint.com/python/index.htm#introduction"}]
priceAlways null: the Google pages show no pricestextalways null
countryCode, languageCodeCountry and language usedtextno"us", "en"
searchUrlGoogle page that was readtextno
statuscomplete, not_found or errortextnocomplete
statusReason, errorCodeReason for not_found/error rows or a note on a resulttextyesSOURCE_BLOCKED
fetchedAtWhen the page was read (ISO 8601)textno

null means Google did not show this value. It is not zero and not an error.

Output entities and surfaces

SituationWhereShapeWhat to do
Normal resultDatasetone row per result, status: completeUse type, position, url
No results for a queryDatasetone not_found rowCheck the query spelling or operators
Some queries failedDataset + RUN_SUMMARYerror rows, run status PARTIALRun the failed queries again
Query stopped after some pagesDataset + RUN_SUMMARYrows of the loaded pages, then one error row; query status partialRun this query again later
Run stopped by time or charge limitRUN_SUMMARYstatus CAPPED, rows so far are keptRaise the limit or split the queries

RUN_SUMMARY in the Key-value store has the run status (COMPLETE, PARTIAL, NO_RESULTS, CAPPED, ERROR), the stop reason, totals (pages, charged pages, rows, organic, ads, related, errors) and, per query, its status (results, partial, not_found, error, not_processed), pages and organic, ad and related counts.

The Apify run itself shows SUCCEEDED for COMPLETE, PARTIAL, NO_RESULTS and CAPPED. It shows FAILED when no query could be loaded or Google kept refusing; rows already saved stay in the Dataset.

Output example

Trimmed first row of a run for coffee machine (us/en, 1 page):

{
"query": "coffee machine",
"page": 1,
"type": "organic",
"position": 1,
"title": "Coffee & Espresso Makers | Macy's",
"url": "https://www.macys.com/shop/home/kitchen-dining/kitchen/coffee-espresso-makers?id=24732",
"displayedUrl": "www.macys.com › Home › Kitchen & Dining › All Kitchen",
"description": "Shop coffee makers at Macy's with stylish, high-quality ...",
"date": "Oct 9, 2026",
"sitelinks": [],
"price": null,
"countryCode": "us",
"languageCode": "en",
"searchUrl": "https://www.google.com/search?q=coffee+machine&hl=en&gl=us",
"status": "complete",
"statusReason": null,
"errorCode": null,
"fetchedAt": "2026-10-22T04:06:02.872Z"
}

A page usually returns 7–10 organic rows and about 8 related searches.

Moving from a one-row-per-page format

Some SERP tools save one row per page with nested lists. Here each result is its own row:

One-row-per-page fieldThis Actor
searchQuery.termquery
searchQuery.pagepage
organicResults[].title / url / displayedUrl / description / daterows with type: "organic", same field names
organicResults[].siteLinkssitelinks (when Google shows them)
paidResults[], paidProducts[]rare ad rows (text ads only); no product ads
relatedQueries[]rows with type: "related"

👉 Try it now: open the input form

Pricing

Pricing unit

You pay per loaded Google results page (serp-page) plus a small start fee per run. Organic results and related searches on that page are included. Platform usage (proxy, compute) is included in the price.

EventFreeBronzeSilverGold+
Run start (apify-actor-start)$0.004$0.004$0.004$0.004
Results page (serp-page)$0.0111$0.00999$0.00888$0.00777
  • A query with no results costs one page: Google was asked and answered.
  • error rows (Google refused the page) are free.
  • The start fee is charged per GB of run memory (at least one). This Actor always runs with 256 MB, so the start fee is $0.004.
  • Why this price: Google often blocks light requests, and then the page goes through Apify's GOOGLE_SERP proxy, which costs about $0.0025 per page before the Store fee. The price stays the same even when every page needs it.
  • If you set a maximum charge for the run, the Actor stops before it, keeps the rows and ends with status CAPPED. Queries it did not reach get a free error row with CHARGE_LIMIT, so you can see what is missing.

Cost examples (Free plan)

ScenarioInput scopePagesCharge
Prefill1 query, 1 page1$0.015
Small keyword list10 queries, 1 page10$0.115
Top 100 for 10 keywords10 queries, 10 pages100$1.114
Large batch (one run)125 queries, 4 pages500$5.554
Query without results1 query1$0.015

👉 Try it now: open the input form

Limits

  • Up to 500 queries and 10 pages per query in one input, but about 500 pages fit in one run: up to 3 pages go through GOOGLE_SERP at a time, 10–20 seconds each. After 3,600 seconds the run stops with status CAPPED and keeps all rows; split bigger lists across runs.
  • A page on the light route takes about a second; the prefill finishes in under a minute either way.
  • Google decides how many pages a query has; some queries end before page 10.
  • Max items (or maxItems) stops the run early with status COMPLETE; pages already loaded are charged, nothing else.

👉 Try it now: open the input form

API, export and integrations

  • API: start the Actor with the same JSON input and read the Dataset items. See the API tab for ready code.
  • Export: download the Dataset as JSON, CSV, Excel or HTML from the Output tab.
  • Scheduling: create a Schedule with your keyword list to track rankings every day or week.

Send a summary to Notion or Slack

Pick a Notion or Slack connector in Send run summary to and, for Notion, a page in Notion page or Slack channel. After the rows are saved, the Actor posts the result count, the first rows and links to the Dataset and the run. In Notion each run adds its summary at the top of the same page. Delivery never changes the run status or the charge; its result is in RUN_SUMMARY.delivery. Without a connector nothing is sent.

Use with AI agents (MCP)

  • MCP server with only this Actor: https://mcp.apify.com?tools=cliqtomedia/google-search-scraper
  • Smallest input:
{ "queries": ["apify web scraping"], "maxPagesPerQuery": 1 }
  • Key output fields: type (organic or related), position, title, url, description.
  • One query page takes 10–20 seconds and costs about $0.015 with the start fee. Keep maxPagesPerQuery low for quick answers.

👉 Try it now: open the input form

Troubleshooting

Some queries have error rows with SOURCE_BLOCKED

Symptom: the run is PARTIAL, some queries have one error row. Likely reason: Google refused the page on the light route and through GOOGLE_SERP too, or GOOGLE_SERP was not reachable. Check: statusReason and RUN_SUMMARY. Action: run only the failed queries again later. Error rows are not charged.

Fewer pages than maxPagesPerQuery

Symptom: a query has 3 pages, you asked for 10. Likely reason: Google has no next page for this query. Action: nothing to fix; only loaded pages are charged.

No ad rows

Symptom: includeAds is on, but there are no ad rows. Likely reason: the results pages the Actor gets from Google rarely carry ads. Action: use the organic rows; ad rows appear only when Google shows ads on that page. Empty sitelinks on a result means Google showed no sitelinks for it.

Results look different from my browser

Likely reason: your browser shows personalised results for your location and account. The Actor gets anonymous results for the chosen country.

👉 Try it now: open the input form

FAQ

Why does url show only the site's home page for some results?

On some pages Google hides the real link behind an encrypted redirect. The Actor opens the redirect to get the real URL. If Google does not reveal it, url is the site's home page, displayedUrl still shows the path, and statusReason says so. Video and forum cards show no address, only a source such as "YouTube · channel": then url is that site's home page (YouTube, Reddit, Medium and other big sites) or empty, and statusReason names the source. Results Google groups under one site with no address of their own get that site's home page.

Do I need a Google account or API key?

No. The Actor reads public results pages without login or cookies of any account.

How many results do I get per page?

Usually 7–10 organic results and 6–10 related searches, depending on the query. A URL that Google repeats on a later page of the same query is saved once.

Is the order the same as on google.com?

position follows the order on the Google results page for the chosen country and language. It can differ a little from your personalised page.

What does null mean?

Google did not show that value, for example no date. It is not zero and not an error.

👉 Try it now: open the input form

👉 Try it now: open the input form

Support

For bugs, questions or feature requests, open the Issues tab of this Actor. Add the run ID, your input (queries can be shortened) and the statusReason text. Never send tokens, cookies or passwords.

⭐ Found it useful? Rate this Actor. Something wrong? Report a problem — we fix reported problems.

The run's final status message ends with the same two links.

👉 Try it now: open the input form