Google News API: search, topics, local news and original article URLs

Get Google News articles as clean JSON for any search query, topic (Business, Technology, Sports...), city or region, or country edition. Every article comes with its headline, publisher, publish time, the Google News link and, by default, the original article URL on the publisher's site.

Built on Google News' public RSS feeds: no browser, no login, no proxy. Fast (100 articles in about 3 seconds) and cheap: $1.50 per 1,000 articles.

What you can use it for

• Media monitoring and brand alerts: track mentions of a company, product or person every hour or day.
• Market and investment research: news flow for tickers, sectors and competitors.
• AI and RAG pipelines: feed fresh headlines and source URLs to an LLM, then fetch the full article from articleUrl.
• Content and SEO research: what is trending in a topic or country right now, and who is covering it.
• Local news: headlines for a city, state or country.
• Datasets: daily snapshots over a date range for research or model training (check the terms first, see the legal note).

Why this Actor

• Original article URLs. Google News links point to news.google.com redirects. This Actor resolves them to the publisher's real URL (100% resolved in our tests) and still returns the Google News link.
• Searches, topics, locations, top stories and raw feed URLs in one run, across any number of editions (country and language).
• More than 100 results per search. Google caps each search at about 100 articles. Set a date range and turn on Split date range by day to run one search per day (up to 92 days).
• Exact date filtering in UTC, with plain dates (2026-10-01) or relative spans (24 hours, 7 days).
• Deduplication across feeds and editions, publisher exclusion, relevance or newest-first ordering.
• Related coverage: topic and top story feeds group several outlets per story; you get them in relatedArticles.
• Clean schema validated before saving, and a per-feed run summary so you can see exactly what happened.

Input example

{
    "queries": ["nvidia earnings", "\"electric vehicles\" site:reuters.com"],
    "topics": ["TECHNOLOGY"],
    "locations": ["Boston"],
    "editions": ["US:en", "GB:en"],
    "dateFrom": "7 days",
    "maxItems": 200,
    "sortBy": "newest"
}

Each query, topic and location runs once per edition. In the example: (2 queries + 1 topic + 1 location) x 2 editions = 8 feeds.

Inputs

Field	What it does
queries	Google News searches. Operators work: "exact phrase", OR, -word, site:, intitle:, when:24h.
topics	WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE, HEALTH, a topic ID (CAAq...) or a news.google.com/topics/... URL.
locations	Local headlines for a place name, for example New York or Texas.
topStories	Also fetch Top stories for each edition.
feedUrls	Any Google News search, topic, location or story URL (RSS or web form). Keeps its own edition.
editions	COUNTRY:language, for example US:en, GB:en, IN:en, DE:de, FR:fr, ES:es, MX:es-419, BR:pt-419, JP:ja, TW:zh-Hant. Default US:en.
dateFrom, dateTo	Date range in UTC: YYYY-MM-DD, an ISO date-time, or 24 hours / 7 days / 2 weeks.
splitByDay	One search per day of the range, to get past the 100 per search limit.
maxItems	Total articles to save (default 100, 0 = no limit).
maxItemsPerFeed	Cap per feed.
sortBy	relevance (Google order, feeds interleaved fairly) or newest.
dedupe	Remove the same article seen in several feeds or editions (default on).
excludeSources	Drop publishers whose name or homepage contains these strings.
resolveUrls	Resolve original article URLs (default on). Turn off for the fastest runs.

Output example

One item per article:

{
    "id": "CBMiggFBVV95cUxPVjFqcXcxNFVD...",
    "title": "Fed holds rates steady despite concerns over inflation",
    "source": "CNBC",
    "sourceUrl": "https://www.cnbc.com",
    "googleNewsUrl": "https://news.google.com/rss/articles/CBMiggFBVV95cUxPVjFqcXcxNFVD...?oc=5",
    "articleUrl": "https://www.cnbc.com/2026/10/28/fed-interest-rate-decision-october-2026.html",
    "urlResolved": true,
    "publishedAt": "2026-10-28T18:00:18.000Z",
    "description": null,
    "relatedArticles": [
        { "title": "Fed leaves rates steady as governor Lisa Cook dissents", "source": "Axios", "googleNewsUrl": "https://news.google.com/rss/articles/CBMib0FV..." }
    ],
    "feedType": "top",
    "feedValue": "Top stories",
    "query": null,
    "language": "en",
    "country": "US",
    "edition": "US:en",
    "rank": 1,
    "feedUrl": "https://news.google.com/rss?hl=en-US&gl=US&ceid=US%3Aen",
    "scrapedAt": "2026-10-28T19:21:03.512Z"
}

Field reference

Field	Description
id	Google News article ID, stable across runs. Use it to dedupe scheduled runs.
title	Headline without the " - Publisher" suffix Google adds.
source, sourceUrl	Publisher name and homepage.
googleNewsUrl	Google News link (redirects to the article).
articleUrl, urlResolved	Publisher's original URL, when resolved.
publishedAt	Publish time, ISO 8601 UTC.
description	RSS description text when it adds something beyond the headline (often null).
relatedArticles	Other outlets covering the same story (topic and top story feeds).
feedType, feedValue, query	Which feed produced the article: search, topic, location, top or custom.
language, country, edition	Edition the article came from.
rank	Position in its feed (1 = top).
feedUrl, scrapedAt	The RSS URL fetched and when.

Run summary (key-value store OUTPUT)

Articles saved and matched, feeds succeeded and failed (with the error per feed), editions Google actually served, duplicates removed, articles outside the date range, URL resolution stats and HTTP stats.

Pricing

Pay per event: $0.0015 per article saved ($1.50 per 1,000). You are not charged for articles removed by the date range, deduplication, publisher exclusion or limits, or for feeds that fail. The only other charge is the tiny platform run-start fee that every pay-per-event Actor has.

Run	Articles	Price
Brand alert, 3 queries, last 24 hours, every hour	about 30 per run	about $0.05 per run
Daily briefing, 8 topics, 1 edition	about 550	about $0.83
1 query, 30 days split by day	up to about 3,000	up to about $4.50

Set Maximum cost per run in the run options to cap spending. The Actor saves exactly as many articles as the cap allows (and only resolves URLs for those), stops cleanly and still writes the run summary.

Limits and good to know

• Google returns about 100 articles per search feed and about 40 to 70 per topic or location feed. Use more queries, editions or Split date range by day to get more.
• Searches use Google's relevance order, not strictly newest first. Use sortBy: newest or a date range when recency matters.
• Unknown editions (for example XX:zz) are silently served as US English by Google. The run summary flags this with a warning per feed.
• URL resolution uses two light requests per 10 articles plus one per article. If Google starts rate limiting, resolution pauses, the remaining articles keep their Google News link and the run still succeeds. Run summary field urlResolution shows the counts.
• Articles have headline, publisher and time only; the RSS feeds do not include full article text. Fetch articleUrl yourself if you need the body, respecting the publisher's terms.
• Default memory is 256 MB. Peak memory in our tests was about 145 MB for 1,000 articles with URL resolution. There is no reason to raise it.

FAQ

Do I need an API key or a proxy? No. The Actor reads Google News' public RSS feeds. A proxy option exists only for unusual cases.

How fresh is the data? Live. Google News feeds update every few minutes.

How do I get only new articles on a schedule? Schedule the Actor (for example hourly) with dateFrom: "2 hours" and dedupe on id in your destination.

Why is articleUrl null for some articles? Google rate limited URL resolution for that run, or the link could not be decoded. googleNewsUrl always works and redirects to the article.

Can I search in other languages? Yes. Add editions such as DE:de or JP:ja; queries are sent as you type them.

Legal note

This Actor reads public Google News RSS feeds and saves headline-level metadata (headline, publisher, time and links). It does not log in, bypass access controls or break captchas, and it does not collect personal data. Google's feeds carry a notice that they are provided for personal, non-commercial feed reader use, and article content belongs to the publishers. You are responsible for making sure your use complies with Google's terms, publishers' rights and applicable law (for example copyright). Link back to the original article when you republish anything. Google News is a trademark of Google LLC. This Actor is not affiliated with or endorsed by Google.

Support

Something not working, or want another field? Open an issue on the Actor's Issues tab with your input. Contact: joesimo@mail.grokbot.com