Google News Scraper

What does Google News Scraper do?

Google News Scraper turns Google News into a clean, deduplicated table of articles. Search by keyword, follow a topic section, watch a publisher, or read the front page of any of Google's language and country editions, and add your own RSS or Atom feeds on top. Every row has the title, the publisher's real article URL, the source, the published date and the summary.

No browser and no proxies. Google News publishes its results as feeds, and this actor reads them the way a feed reader does, which keeps runs fast and cheap.

The real URL, not the redirect

Google News links point to news.google.com and bounce to the article through a redirect that some tools return as the "URL". This actor decodes the link or follows the redirect and gives you the publisher's own address, with tracking parameters removed, so the same article found through two queries is one row, not two. The Google link is kept in googleUrl for reference.

Past the 100-item ceiling

Google News returns at most 100 items for any single query, whether you ask for one day or one year. That is the ceiling every other scraper in this category hits, and it is what buyers complain about in their reviews: they see hundreds of results on the Google News site and get a hundred rows back.

Give this actor a dateFrom and a dateTo and it splits each query into day-sized windows, runs them all and merges the results, deduplicated by the article URL. Measured on 3 September 2026 for the keyword "artificial intelligence" over one week:

Query	Articles returned
One query for the whole week	100
The same week split into days	632

New articles only

Turn on newOnly, give the run a watch name or save it as a task, and schedule it hourly or daily. Every run compares its results with the previous snapshot and returns only the articles that appeared since. You pay for those and nothing else. Brand monitoring, competitor news, a topic feed for a newsletter: one scheduled task each.

Input

{
  "queries": ["Apify", "\"web scraping\" when:7d", "site:reuters.com AI"],
  "topics": ["TECHNOLOGY"],
  "topStories": false,
  "feeds": ["https://news.ycombinator.com/rss"],
  "language": "en",
  "country": "US",
  "dateFrom": "2026-08-25",
  "dateTo": "2026-08-31",
  "maxItemsPerSource": 100,
  "resolveLinks": true,
  "newOnly": true,
  "watchKey": "brand"
}

Google's search operators work inside queries: quotes for exact phrases, site: for a publisher, -word to exclude, when:7d for a time window. Language and country pick the Google News edition (en/US, en/GB, de/DE, cs/CZ, pt/BR, ja/JP and every other one Google publishes).

Output

{
  "key": "https://www.reuters.com/technology/example-article-2026-09-02/",
  "title": "Example headline",
  "url": "https://www.reuters.com/technology/example-article-2026-09-02/",
  "googleUrl": "https://news.google.com/rss/articles/CBMi...",
  "sourceName": "Reuters",
  "sourceUrl": "https://www.reuters.com",
  "publishedAt": "2026-09-02T08:15:00.000Z",
  "summary": "Example headline - Reuters",
  "source": { "type": "query", "value": "site:reuters.com AI" },
  "language": "en",
  "country": "US"
}

Pricing

Event	Price
Article returned	$0.001 ($1 per 1,000 articles)
Duplicates, articles already seen in new-only mode, articles outside the time window	free

No start fee. A run that returns nothing costs nothing.

Frequently asked questions

How many articles does a query return? Google News caps a single query at 100 items. Set dateFrom and dateTo and the actor splits the range into day-sized windows, which returned 632 articles for one keyword over a week instead of 100.

Some URLs still point to news.google.com. Why? For a small share of links Google's redirect needs a browser to resolve. Those rows keep the Google link and are counted in the summary as unresolved, so you know exactly how many there were.

Can I read feeds that are not Google News? Yes. Put any RSS or Atom URL into feeds; they get the same output format and the same deduplication and new-only handling.

Which languages are supported? Every Google News edition. Pass the language and country codes of the edition you want.

How to use Google News Scraper

1. Open the actor and fill in queries in the input form, or paste the JSON input from the example above.
2. Click Start. The first rows appear in the Output tab within seconds; the run summary is in the Storage tab under the key SUMMARY.
3. Download the results as JSON, CSV or Excel from the Output tab, or read them through the API link shown there.
4. To run it on a schedule, click Save as a task, set the input once and add a schedule. Scheduled tasks keep their own change snapshot, so the new-only mode works without any extra setup.

Integrations

The actor is available in the Apify apps for Make, Zapier, n8n and Keboola, where you pick it by name and map the input fields. It is also exposed to AI agents through the Apify MCP server, and callable from any language through the Apify API; the API tab on this page has ready-made snippets for Node.js, Python and curl. Every run can post its results to a webhook when it finishes.

Support

Questions and problems go into the Issues tab of this actor. Each issue is answered by the maintainer, and fixes ship as new builds without any change on your side.

Built and operated by deriverge s.r.o. Same rules as our other actors: unified schema, honest billing, no charge for what you did not get.