Google News Scraper: Headlines, Sources, Real Article URLs

This Google News scraper collects articles from Google News search results, topic sections, local news and top stories, for any country and language Google News publishes. Every row has the headline, publisher, publish time, rank and the real article URL on the publisher's site, decoded from the Google News redirect link.

• Search, topics, local news and top stories in one run, with Google's own operators like when:7d, site:, intitle: and "exact phrase".
• Real publisher URLs. Google News hides each article behind an encoded link. The scraper decodes it to the original URL, for example https://www.reuters.com/world/.... If one link cannot be decoded, the row is still saved with the reason.
• More than 100 results. A Google News feed stops at about 100 articles. Turn on Split the date window by day and the scraper searches each day separately.
• No duplicates. An article that appears under several queries is saved once.

What data you get

Field	What it holds
title	Headline, with the trailing " - Source" removed
sourceName, sourceUrl	Publisher name and home page
publishedAt	Publish time in ISO 8601 (UTC)
articleUrl, articleDomain	The article's own URL on the publisher's site, and its domain
articleUrlError	Why articleUrl is empty, when it is
rank	Position in Google's feed, starting at 1
query, topic, feedType	Which search, topic, location or top stories feed the row came from
country, language	The Google News edition that served the row
relatedArticles	Other outlets' coverage that Google groups with the story (topics and top stories)
googleNewsUrl, googleNewsId	The Google News link and its stable ID

How to use it

1. Add one or more Search queries, pick Topics, add Local news places, or turn on Top stories.
2. Set Country and Language, for example US and en, CA and fr, DE and de, BR and pt-419.
3. Optional: pick a Time range, or set Published from and Published to.
4. Set Max articles per query.
5. Run it and export to JSON, CSV or Excel, or call it from the API, Make, Zapier, n8n or an AI agent through the Apify MCP server.

Search operators

These go straight into a search query:

• when:1h, when:1d, when:7d: only recent articles
• site:reuters.com: one publisher only
• intitle:earnings: the word must be in the headline
• "interest rates": exact phrase
• tesla OR rivian, apple -iphone: either word, or exclude a word

Getting more than 100 articles

Google returns about 100 articles per feed. To go further, set Published from (and optionally Published to), turn on Split the date window by day, and raise Max articles per query. In our test, bitcoin over five days returned 473 unique articles instead of 100.

Example input

{
  "searchQueries": ["nvidia", "nvidia stock", "site:reuters.com tariffs", "intitle:earnings apple"],
  "includeTopStories": true,
  "country": "US",
  "language": "en",
  "maxArticlesPerQuery": 30
}

This run saved 150 articles, decoded all 150 publisher URLs and skipped 12 articles that nvidia and nvidia stock had in common.

Example output (one article)

{
  "feedType": "search",
  "query": "nvidia",
  "topic": null,
  "country": "US",
  "language": "en",
  "rank": 1,
  "title": "EXCLUSIVE: US DOJ is inquiring about Nvidia's purchase of Hugging Face, sources say",
  "sourceName": "Reuters",
  "sourceUrl": "https://www.reuters.com",
  "publishedAt": "2026-10-28T20:41:13.000Z",
  "articleUrl": "https://www.reuters.com/world/us-doj-is-inquiring-about-nvidias-purchase-hugging-face-sources-say-2026-10-28/",
  "articleDomain": "reuters.com",
  "articleUrlError": null,
  "googleNewsUrl": "https://news.google.com/rss/articles/CBMirgFBVV95cUxNQi1pQlpX...?oc=5",
  "googleNewsId": "CBMirgFBVV95cUxNQi1pQlpX...",
  "relatedArticles": [],
  "scrapedAt": "2026-10-28T22:05:02.642Z"
}

How much does it cost?

You pay per result: each article saved to the dataset is one result. A result costs $0.002 ($2.00 per 1,000), plus $0.0005 once per run, and the decoded publisher URL is included at no extra cost. A search that fills Google's 100-article feed costs about $0.20, and Apify's free plan credit of $5 a month covers about 2,500 articles.

Feeds that fail and searches with no matches add no rows, so they cost nothing. Failed feeds are listed in the PROBLEMS record of the run's key-value store.

Use cases

• Brand and competitor monitoring: run your company and product names every hour with when:1h on a schedule.
• Market and trading signals: collect headlines for tickers and sectors with exact publish times.
• Media research: see which outlets cover a story, using relatedArticles from top stories.
• AI and RAG pipelines: feed the decoded articleUrl to a content extractor to get the article text.

FAQ

Why decode the Google News links? A Google News link like news.google.com/rss/articles/CBMi... only redirects in a browser. Most tools that fetch it get a Google page instead of the article. The decoded articleUrl is the publisher's own address, which you can store, deduplicate by domain or pass to another scraper.

What if a link cannot be decoded? The article is still saved, with articleUrl set to null and the reason in articleUrlError. The run does not fail.

Can I make runs faster? Turn off Decode real article URLs if headlines and Google links are enough. Decoding takes one extra round trip to Google per article. You can also raise Decoding concurrency.

Do I need a proxy? No. Proxy is off by default because the feeds and the decoding work without one, and decoding is much faster direct. If Google starts refusing direct requests during a run, the scraper switches to Apify Proxy on its own. You can also turn a proxy on from the start in the input.

Why did a country fail? Google News only has certain country and language pairs. If you ask for one it does not have, Google quietly sends another country's news. The scraper catches this, skips the feed and explains it in PROBLEMS rather than giving you the wrong country's articles.

Does it get article text or images? No. It returns what Google News lists plus the decoded URL. Pass articleUrl to a content extractor if you need the full text.

Is this legal? It reads the same public feeds anyone can open. You are responsible for how you use the data and for respecting publishers' terms and copyright.

Other Actors by this developer

• Google Trends Scraper: interest over time, by region and related searches, built to survive Google's rate limits.
• Google Trending Searches Now: today's trending searches for any list of countries.