Google News Scraper (bulk queries, lightweight)

News articles from Google News for up to 200 search terms per run: title, source, publication date, link and snippet. Works on search terms, on topic sections, and on a country's front page. You only pay for articles that are actually delivered.

Why this scraper

• Three kinds of input, one table. A search term like electric vehicles, a topic like technology, or the word headlines for the country's front page. Mix them in one run and every row tells you which query it came from.
• Properly region aware. Google News is strongly region bound. Set country and language and you get what a reader in that country sees, not a US default. A Dutch run returns NU, NRC and Volkskrant, not CNN.
• Clean dates. The publication date comes back as ISO 8601 so it sorts and filters straight away, with the original text alongside it.
• Only delivered articles are charged. A term without results comes back as a clear error record at no cost.
• A hard cost ceiling you control. maxArticlesPerQuery is the cap: one delivered article is one charged event.

Who this is for

Media and brand monitoring, topic tracking, competitor watching, and news feeds for dashboards or AI pipelines. Run the same list of terms on a schedule and you have a timeline of what was written about them, by whom and when.

Who this is not for

Read this before you buy. The url is a news.google.com link, not the publisher's own address. Google encrypts the target inside that link, so this actor cannot hand you nrc.nl/article/... directly. The link opens the correct article in a browser, and the publisher is returned separately in the source field.

That makes this actor a strong fit for monitoring, alerting and analysis, and a poor fit if you need to fetch the full article text automatically. If that is your use case, you need a scraper that visits each publisher, and this is not it. Better to know that now than after you have paid.

Input example

{
    "queries": [
        "artificial intelligence",
        "technology",
        "headlines"
    ],
    "maxArticlesPerQuery": 50,
    "country": "US",
    "language": "en"
}

Topic names that work: world, nation, business, technology, entertainment, sports, science, health. Anything else is treated as a search term.

Output example

{
    "query": "artificial intelligence",
    "country": "US",
    "language": "en",
    "position": 1,
    "title": "AI could kill all humans in next decade, warn experts",
    "source": "The Guardian",
    "publishedAt": "2026-09-10T01:31:00+00:00",
    "publishedText": "Thu, 10 Sep 2026 01:31:00 GMT",
    "url": "https://news.google.com/rss/articles/CBMirgFBVV95cUxNNUNtV20wYXJ",
    "snippet": "AI could kill all humans in next decade, warn experts",
    "status": "ok"
}

A term that cannot be delivered produces an error record instead, and is not charged:

{
    "input": "qzxwv nonsense 99999",
    "status": "error",
    "errorCode": "NO_RESULTS",
    "error": "No articles for 'qzxwv nonsense 99999' in US"
}

Error codes: INVALID_QUERY, NO_RESULTS, BLOCKED. INPUT_TRUNCATED appears once when your input is longer than this actor accepts.

Related actors from steadydata

• rss-feed-reader (publishing soon): the publishers' own feeds, for sources Google News omits
• google-trends-batch: whether interest in the topic is rising or falling

Pricing

Pay per event: one article-listed event per delivered article. No charge for terms that fail or return nothing, no separate platform-usage surcharge.

FAQ

How many articles do I get per term? Google returns up to about a hundred per feed, and fewer for narrow terms. You are only charged for what actually arrives.

How far back does it go? Google News is a moving window, weighted towards recent coverage. For older material a narrower search term helps, but there is no date filter in the feed itself.

Can I track a topic over time? Yes, and that is the main use: run the same term list daily and keep the rows with a timestamp. The publishedAt field makes deduplication across runs straightforward.

Why is the snippet almost the same as the title? Because that is what Google supplies in the feed. It is returned as given rather than padded out with invented text.

Is personal data collected? No. Public article metadata only: headline, outlet, date and link.

What happens when Google changes something? Feeds change from time to time; that is the nature of this work. The actor is monitored daily and fixed fast, and while it is broken you are not charged, because only delivered articles cost anything.