Google News Scraper – Full Article Text & Real Publisher URLs

Google News scraper for headlines, keyword search, topics and local news in any language and country. Get the real publisher article URL (decoded from news.google.com links), source, publish date and optional full article text, author and image. No login, no API key.

Price: $1.50 per 1,000 articles ($4 per 1,000 with full text). No start fee, no login, no API key.

What data do you get?

• 📰 Headline, source and publisher homepage – the title is cleaned (no trailing " - Reuters")
• 🔗 Real article URL (articleUrl) – the actual reuters.com/… link, not Google's redirect. Most Google News scrapers only return the news.google.com link, which you can't open in code, deduplicate or feed into other tools.
• 🕒 Publish date in ISO 8601 (UTC)
• 🧾 Full article text (optional) – main text with paragraph breaks extracted with Mozilla Readability (the engine behind Firefox Reader View), plus word count, author, description and main image
• 🚦 Honest status fields – decodeStatus and fullTextStatus (ok, paywalled, blocked, no-content, timeout…) tell you exactly what worked
• 🗞 Related coverage – for top stories, the other outlets covering the same story
• 🌍 Any edition – language + country (en-US, en-GB, de-DE, fr-FR, es-MX, pt-BR, tr-TR, ja-JP, zh-TW, …)

Use cases

• Media monitoring & PR – track mentions of your brand, competitors or executives every hour or day.
• Market & financial research – follow news about tickers, sectors or crypto and feed it to sentiment models.
• AI agents, RAG & LLM pipelines – fetch fresh news with clean full text in one call.
• News aggregation & newsletters – build topic digests with real publisher links.
• Academic & policy research – build news datasets by keyword, country and date range.
• SEO & content research – see which publishers rank in Google News for your keywords.

How to use it

1. Click Try for free / Start.
2. Enter one or more Search queries (Google operators like "exact phrase", OR, -word, site:reuters.com work) and/or pick Topics or Locations. Leave everything empty to get today's top stories.
3. Choose Language, Country and Time range, set Max articles per query / topic and decide whether you want full text.
4. Click Start and export the results as JSON, CSV, Excel or via API, or connect Make, Zapier, n8n, Google Sheets or webhooks.

Input

Field	Type	Description
queries	array of strings	Keywords to search. Google search operators work.
topics	array (enum)	TOP_STORIES, WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE, HEALTH
locations	array of strings	Local news, e.g. London, California
language	string	2-letter language code. Default en
country	string	2-letter country code. Default US
timeRange	enum	any (default), 1h, 1d, 7d, 30d, 1y
maxItems	integer	Max articles per query / topic / location. Default 20, up to 5000
decodeUrls	boolean	Decode real publisher URLs. Default true
fullText	boolean	Extract full article text, author, description, image. Default false
proxyConfiguration	object	Optional. Not needed for normal use

With {} (empty input) the Actor returns the 20 current US top stories.

Example input:

{
    "queries": ["artificial intelligence", "site:reuters.com bitcoin"],
    "topics": ["TECHNOLOGY"],
    "language": "en",
    "country": "US",
    "timeRange": "7d",
    "maxItems": 50,
    "decodeUrls": true,
    "fullText": true
}

More than 100 results: Google News returns at most 100 articles per feed. For search queries with maxItems above 100 the Actor automatically searches day by day (Google's after: / before: operators) and deduplicates, so you can collect thousands of articles for one keyword.

Output

One row per article (type: "article"). Real sample (full text shortened):

{
    "type": "article",
    "rank": 2,
    "title": "Jaguar reveals $130K electric vehicle two years after failed ‘woke’ rebrand video",
    "source": "New York Post",
    "sourceUrl": "https://nypost.com",
    "articleUrl": "https://nypost.com/2026/10/07/business/jaguar-reveals-130k-electric-vehicle-two-years-after-failed-woke-rebrand-video/",
    "articleDomain": "nypost.com",
    "googleNewsUrl": "https://news.google.com/rss/articles/CBMiugFBVV95cUxQTGU4d0ROampoQzV2c2t5...?oc=5",
    "decodeStatus": "decoded",
    "publishedAt": "2026-10-07T18:27:00.000Z",
    "description": "Jaguar is shifting gears — away from “woke” weirdos and back to classy babes after an attempted rebrand drew widespread derision.",
    "author": "Taylor Herzlich",
    "imageUrl": "https://nypost.com/wp-content/uploads/sites/2/2026/10/COMP-GettyImages-2188001841.jpg?quality=75&strip=all&w=1200",
    "fullText": "See more of our coverage in your search results.\n\nJaguar is shifting gears — away from “woke” weirdos and back to classy babes after an attempted rebrand drew widespread derision.\n\nThe British luxury automaker unveiled a $130,500 electric vehicle Tuesday night...",
    "wordCount": 739,
    "fullTextStatus": "ok",
    "fullTextError": null,
    "query": "electric vehicles",
    "topic": null,
    "location": null,
    "language": "en",
    "country": "US",
    "relatedArticles": [],
    "scrapedAt": "2026-10-07T21:07:23.703Z"
}

Top stories also include relatedArticles (title, source, googleNewsUrl) – other outlets covering the same story.

description, author, imageUrl, fullText and wordCount are filled only when fullText is on (Google News RSS itself does not contain them).

fullTextStatus values: ok, paywalled (only the free part, if any), blocked (site refuses bots, e.g. HTTP 403 or captcha), no-content, not-html, timeout, failed, not-requested.

Queries without results and invalid topics produce a free type: "error" row (errorCode: NO_RESULTS, INVALID_INPUT, SCRAPE_FAILED), so the run still succeeds.

The dataset has ready-made views in Apify Console: Overview, Full text and Errors & status.

Pricing

Pay-per-event: you only pay for articles you get. Platform usage is included.

Event	Price	When it is charged
article	$0.0015 ($1.50 / 1,000)	Each article saved, with decoded URL
article-full-text	$0.004 ($4 / 1,000)	Instead of article, when full text was successfully extracted
Errors / no results	Free

Each article is charged once: either article or article-full-text. If you turn on full text but the page is paywalled or blocked, you pay only the normal article price.

Job	Cost
20 top stories (default input)	$0.03
100 articles for 1 keyword, real URLs	$0.15
100 articles with full text (≈80 extracted)	about $0.35
1,000 articles per day for a month	about $45

Tip: set Max cost per run in the run options. The Actor stops cleanly when your budget is reached.

Use with AI agents (MCP)

Add the Actor to the Apify MCP server (https://mcp.apify.com?tools=openharvest/google-news-scraper) or let agents find it with search-actors ("google news"). The input is small and every field has a default, so agents can call it with only a query.

Example call-actor input for "What's the latest news about Nvidia this week? Summarize the articles.":

{
    "queries": ["Nvidia"],
    "timeRange": "7d",
    "maxItems": 10,
    "fullText": true
}

Example for "Give me today's top business headlines in Germany":

{ "topics": ["BUSINESS"], "language": "de", "country": "DE", "timeRange": "1d", "maxItems": 15 }

FAQ

How are the real article URLs decoded? Google News links (news.google.com/rss/articles/CBMi…) are encrypted redirect ids. The Actor reads each article's signature from Google News and asks Google's own web endpoint (the same one the Google News website uses) to resolve the ids in batches. Old-format ids are decoded locally. In our tests more than 99% of articles were decoded; if one fails you still get the googleNewsUrl and decodeStatus: "failed".

Why is the full text missing for some articles? Some publishers block automated access (blocked) or put articles behind a paywall (paywalled). Those rows are charged as a normal article, not as full text. The Actor never bypasses paywalls or logins.

How many articles can I get? Up to 100 per topic / location feed (Google's limit) and up to 5,000 per search query thanks to the day-by-day deep search.

Are duplicates removed? Yes. If the same article appears in several of your queries or topics, it is saved and charged only once.

Which languages and countries work? Every Google News edition. Use the 2-letter codes, e.g. en/GB, fr/FR, es/MX, pt/BR, hi/IN, ja/JP, tr/TR.

Can I run it on a schedule? Yes. Use Apify Schedules with timeRange: "1h" or "1d" to collect only fresh articles.

Is it legal to scrape Google News?

The Actor reads only publicly available Google News feeds and public article pages, without logging in, and it does not circumvent paywalls. Headlines and article texts are copyrighted by their publishers: use full texts for analysis, research, monitoring or internal AI processing, and check the publisher's terms and your local copyright law before republishing content. Google's feed terms limit some commercial uses. If you are unsure, consult a lawyer.

Related actors

• RSS Feed Reader & Discovery – Parse Any RSS/Atom/JSON Feed, Find Feeds on Websites, Get Full Text (coming soon)
• Hacker News Scraper – Stories, Comments, Search & Who Is Hiring (coming soon)
• Google Trends Scraper (coming soon)
• Telegram Channel Scraper – Public Posts, Views & Media (coming soon)

Support & feedback

Found a bug or a missing field? Open an issue in the Issues tab. We usually reply within a day.