Google News Keyword & Full-Text Article Scraper

Search Google News for any keyword and get back real, resolved article links — not Google's redirect URLs — along with headline, source, and publish date. Turn on full-text and image extraction when a snippet isn't enough and you need the actual article body.

What this actor does

• Searches Google News for one or more keywords in a single run and deduplicates articles that match more than one query
• Resolves each result to the publisher's real article URL instead of a Google redirect link whenever it can
• Filters results by recency, a custom date range, language, country, a specific publisher domain, or excluded words
• Can optionally fetch each article's page to pull the full body text and the main article image
• fullText, imageUrl, and author are always present on every record but come back null when the corresponding option is off or nothing was found; publishedAt, snippet, source and domain are empty strings when Google News doesn't supply them

Output per article

• query — the search keyword that returned this article
• title — the article headline
• url — the publisher's actual article URL, resolved from Google News (if resolution fails, the Google News link is kept)
• source — the publisher name, e.g. BBC, Reuters
• domain — the publisher's domain, e.g. bbc.com (from the resolved URL, or from Google News' source link when unresolved)
• publishedAt — the article's publication date and time in ISO 8601, as listed by Google News
• snippet — the short summary Google News shows for the article
• fullText — the full article body text; null unless extractFullText is enabled
• imageUrl — the article's main image URL; null unless includeImages is enabled
• author — the article's byline; null unless extractFullText or includeImages is enabled (either one opens the article page) and the page publishes one
• language — the language code used for the search
• scrapedAt — when the actor started on that query

Input

Field	Type	Default	Description
queries	array of strings	–	Search keywords; each is searched separately and results are deduplicated across all of them. Required.
maxResultsPerQuery	integer	20	Maximum articles to return per query (1–100).
language	string	en	Language code for the search, e.g. en, de, fr, es.
country	string	US	Country code for regional news, e.g. US, GB, DE, IN.
dateRange	string	any	Recency filter: 1h, 6h, 1d, 7d, 1m, 1y, or any. Ignored when dateFrom/dateTo are set.
dateFrom	string	–	Only articles published on or after this date (YYYY-MM-DD); overrides dateRange.
dateTo	string	–	Only articles published on or before this date (YYYY-MM-DD); use together with dateFrom.
siteFilter	string	–	Restrict results to one publisher domain, e.g. reuters.com.
excludeWords	array of strings	–	Words or phrases added to the search as exclusions, so Google News leaves out matching articles.
extractFullText	boolean	false	Fetch each article's page and extract the full body text and author byline.
includeImages	boolean	false	Fetch each article's page and extract its main image URL (the author byline is picked up too when present).
maxConcurrency	integer	5	How many articles to process in parallel when extractFullText/includeImages is on (1–20).

Example: quick scan of today's headlines

{
  "queries": ["artificial intelligence", "climate change"],
  "maxResultsPerQuery": 20,
  "dateRange": "1d"
}

Example: full article text and images

{
  "queries": ["renewable energy"],
  "maxResultsPerQuery": 10,
  "extractFullText": true,
  "includeImages": true
}

Example: one publisher, custom date range, with exclusions

{
  "queries": ["elections"],
  "siteFilter": "bbc.com",
  "dateFrom": "2026-01-01",
  "dateTo": "2026-02-28",
  "excludeWords": ["opinion", "editorial"],
  "maxResultsPerQuery": 50
}

Example: non-English regional news

{
  "queries": ["Nachrichten"],
  "language": "de",
  "country": "DE",
  "maxResultsPerQuery": 10
}

Use cases

• Media monitoring — track coverage of your brand, competitors, or industry across the publishers Google News indexes
• Market research — pull news mentioning a company, product, or market trend for analysis
• Newsletter curation — build a themed news digest from one or more topics
• Sentiment analysis — feed full article text into an NLP pipeline to gauge coverage tone on a topic
• Academic or policy research — gather a dated, filtered news dataset for a study
• Dataset building — collect full article text at scale for downstream text-analysis work

FAQ

Does this return the real article link or a Google redirect? The real, resolved publisher URL, not the news.google.com redirect link Google News shows in its own results. In the rare case a link can't be resolved, the Google News link is returned in url instead.

Why are fullText, imageUrl, or author null in my results? All three fields are always present on the record, but they only get a real value when you turn on extractFullText and/or includeImages and the article page actually has that data; otherwise they come back as null. Pages that fail to load also leave them null. By default (both options off) the actor returns headlines and snippets only, which is faster.

Can full-text extraction read paywalled articles? No — sites behind a paywall (e.g. subscription-only news sites) won't return usable full text, since the actor only reads what's publicly visible on the page.

Can I search multiple keywords at once? Yes, add as many entries as you like to queries; results across all of them are deduplicated by their Google News link, so an article found by an earlier query is not repeated for a later one.

What's the difference between dateRange and dateFrom/dateTo? dateRange is a quick recency filter (e.g. "past week"). Setting dateFrom and/or dateTo gives you a custom date window instead and makes the actor ignore dateRange. Both are passed to Google News as search conditions, so the dates Google applies are what count.

Can I restrict results to one news source? Yes, set siteFilter to that publisher's domain, e.g. reuters.com.

How many articles can I get per query? Up to maxResultsPerQuery, capped at 100, limited by how many matching articles Google News returns for that query and filter combination in a single result list (there is no paging beyond it).

Is this affiliated with Google? No. This is an independent, third-party actor that reads Google News' publicly available search results; it is not affiliated with or endorsed by Google.