Google News Scraper — Real Publisher URLs

Scrape Google News by keyword, topic, language and country, and get back the real publisher URL for every article — not the opaque news.google.com redirect link that most scrapers return.

Fast, HTTP-only (no browser), typically under 10 seconds for a full query.

Google News scraping in practice: the real publisher URL, not the redirect

A media monitoring team needs the article link it can actually open, archive and cite — not the Google redirect that most scrapers return. This Actor resolves every result to the publisher URL. A real record:

{
    "article_uid": "CBMiwgFBVV95cUxP...",
    "title": "UN chief welcomes first global AI framework",
    "article_url": "https://news.un.org/en/story/2026/07/1165432",
    "google_rss_url": "https://news.google.com/rss/articles/CBMiwgFBVV95cUxP...",
    "decoded": true,
    "source_name": "UN News",
    "source_url": "https://news.un.org",
    "published_at": "2026-07-02T17:17:49.000Z",
    "snippet": "The UN Secretary-General welcomed the adoption of...",
    "query": "artificial intelligence",
    "feed_type": "search",
    "language": "en",
    "country": "US"
}

Every item always includes google_rss_url; article_url is filled whenever decoding succeeds (typically over 95%). Items are deduplicated across queries within a run.

Why this Google News scraper returns real publisher URLs

Most Google News scrapers hand you links like https://news.google.com/rss/articles/CBMi.... Those are useless if you want to fetch the article, feed an LLM pipeline, monitor brand mentions, or archive the source. This Actor decodes every link into the final publisher URL (for example https://www.reuters.com/technology/...):

	This Actor	Typical alternatives
Real article URLs	✅ decoded	❌ Google redirect links
Browser needed	❌ no (fast & cheap)	often Playwright/Chrome
Topic sections	✅ 9 sections	varies
Any language/country edition	✅	varies
Honest empty results	✅ 0 items = 0 charge	varies

Features: resolved publisher URLs, language and region control

• Decoded publisher URLs — every news.google.com redirect is resolved to the real article URL, ready to fetch or cite.
• Keyword and topic search — run any query (with Google News operators) or pull a topic section such as Business, Technology or Health.
• Every edition — pick any language and country pair, from en/US to pt-BR/BR.
• Date windows — restrict search results to the past hour, day, week, month or year.
• No browser — pure RSS over HTTP, so runs are fast and cheap.
• MCP-ready — call it from Claude, ChatGPT or any MCP client to give an AI agent fresh, citable news.

Use cases: media monitoring, adverse media screening and PR tracking

• News monitoring / brand tracking — track any keyword, company or person across every Google News edition.
• LLM & RAG pipelines — feed real article URLs to your ingestion pipeline; the decoded URL can be fetched directly.
• AI agents (MCP) — give your agent fresh news with citable publisher sources instead of dead redirect links.
• Market & competitor research — follow product launches, funding rounds and regulation news.
• Sentiment analysis — collect headlines and snippets per topic, language and country.

Input: query, language and region

Field	Type	Default	Notes
queries	array	["artificial intelligence"]	One search per item. Supports "exact phrase", site:reuters.com, intitle:x, OR.
topic	enum	—	Optional topic section: TOP, WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SCIENCE, SPORTS, HEALTH.
language	string	en	For example en, pt-BR, es, de, fr, ja.
country	string	US	For example US, BR, GB, DE, IN.
dateRange	enum	any	1h, 1d, 7d, 30d, 1y (search queries only).
maxItemsPerQuery	int	100	Google News RSS caps at ~100 items per feed.
decodeArticleUrls	bool	true	Turn off for maximum speed (you get only Google redirect URLs).
proxyConfiguration	object	Apify proxy	Datacenter proxy is enough.

Input example: a news query

{
    "queries": ["brazil economy", "\"central bank\" site:reuters.com"],
    "topic": "BUSINESS",
    "language": "en",
    "country": "US",
    "dateRange": "7d",
    "maxItemsPerQuery": 50,
    "decodeArticleUrls": true
}

Tips for cleaner news monitoring

• Widen coverage — Google News RSS returns up to ~100 items per feed, so split broad topics into several specific queries or per-day dateRange windows.
• Breaking news — set dateRange: "1h" and schedule the Actor to run on a short interval.
• Local editions — match language and country to the market you track, for example pt-BR + BR for Brazilian coverage or de + DE for Germany.
• Speed over URLs — set decodeArticleUrls: false when you only need headlines and can live with the redirect links.

Pricing: what Google News scraping costs

Pay per result: you are charged only for news items actually delivered, at US$ 2.00 per 1,000 items ($0.002 each). A run that finds nothing costs nothing, and decoding the real publisher URLs is included in the price. Example: a query that returns 80 articles costs US$ 0.16.

FAQ about Google News scraping and redirect URLs

Where does the data come from? Google News public RSS feeds — the same data you see on news.google.com, with no login and no personal data.

How fresh is it? RSS reflects Google News in near real time; use dateRange: "1h" for breaking-news monitoring.

Can I get full article text? Not from this Actor (by design, so it stays fast and cheap). Pipe article_url into any article-extraction Actor or your own fetcher.

How many results per query? Google News RSS returns up to ~100 items per feed. Use multiple, more specific queries (for example per-day dateRange windows) to widen coverage.

What if a URL cannot be decoded? The item is still returned with decoded: false and its google_rss_url, so you never lose an article. Decoding succeeds for the large majority of items.

MoreLock — the Brazil data suite

This Actor is part of MoreLock, a collection of Brazil-data Actors built to join on the CNPJ. People who run this one usually pair it with:

Actor	What it adds to your workflow
Brazil Due Diligence — CNPJ Registry + Reclame Aqui	Registry data and Reclame Aqui reputation joined into one record per CNPJ.
Brazil Company Risk Score — Registry, Reputation & Protests	A single risk grade per company, built from registry, reputation and protest signals.
Brazil Trademark Watch — INPI/RPI Monitor & Deadline Alerts	Deadline alerts on INPI/RPI publications for the trademarks you track.

See all 26 MoreLock Actors →

Changelog

• 0.1 — Initial release: keyword and topic search across every Google News edition, decoded publisher URLs, pay-per-result billing.

Support

Found a bug or need another field? Open an issue on the Actor's Issues tab or contact the developer through the Apify profile. Feature requests are welcome.