Google News Scraper

Google News Scraper pulls articles from Google News search results. Give it one keyword or a list of keywords and it returns titles, publisher names, direct article URLs, publish dates, and thumbnails for every result.

Supports locale and country targeting, so you can pull regional news feeds in any language Google News covers. Run multiple queries in a single actor run and all results land in one dataset, tagged by query.

What it collects

For each article:

• articleTitle - the headline as it appears on Google News
• sourceName - publisher or outlet name (e.g. Reuters, BBC, TechCrunch)
• articleUrl - direct link to the original article on the publisher's site, not a Google redirect
• googleNewsUrl - the full Google News read link
• publishedAt - ISO 8601 datetime in UTC (e.g. 2026-05-18T01:14:00Z)
• publishedText - the relative label shown on the page (e.g. "4 hours ago", "Yesterday")
• thumbnailUrl - article image from Google News CDN
• searchQuery - which query produced this result (useful when running multiple queries)
• scrapedAt - ISO 8601 timestamp of when the article was collected
• error - null on success, error message if extraction failed for that item

How to use

1. Open the actor and click Try for free
2. Enter one search query in the Search query field, or add multiple terms in the Search queries list
3. Set language and country for regional results (optional, defaults to US English)
4. Set max items per query and click Start

Results land in the Storage tab as JSON, one item per article.

Single query

{
  "query": "Tesla USA",
  "language": "en-US",
  "country": "US",
  "maxItems": 50
}

Multiple queries in one run

Use the queries field to run several searches at once. Each result includes a searchQuery field so you can filter or group by keyword downstream.

{
  "queries": ["Tesla USA", "Apple earnings", "AI regulation EU", "Bitcoin price"],
  "language": "en-US",
  "country": "US",
  "maxItems": 50
}

This collects up to 50 articles per query, so the above would return up to 200 total results in one run.

You can also combine both fields. If you set query and queries, the actor merges them and deduplicates:

{
  "query": "Tesla USA",
  "queries": ["Apple earnings", "Tesla USA"],
  "maxItems": 30
}

"Tesla USA" appears in both but runs only once.

All inputs

Field	Required	Default	Description
query	No	-	Single search term (e.g. "Tesla USA")
queries	No	-	List of search terms for multi-query runs
language	No	en-US	Google News language code (e.g. en-US, en-GB, de, fr, ja, es, pt-BR)
country	No	US	Two-letter country code (e.g. US, GB, DE, FR, JP, BR)
ceid	No	auto	Edition string like "US:en". Built automatically from country + language if left blank
maxItems	No	100	Max articles per query, up to 1000
requestTimeoutSecs	No	30	Per-request timeout in seconds (5-120)

At least one of query or queries must be provided.

Example output

{
  "articleTitle": "Tesla Raises U.S. Model Y Prices. Its Stock Just Forged A Lower Buy Point.",
  "sourceName": "Investor's Business Daily",
  "articleUrl": "https://www.investors.com/news/tesla-raises-u-s-model-y-prices-lower-buy-point/",
  "googleNewsUrl": "https://news.google.com/read/CBMihgFBVV95cUxN...",
  "publishedAt": "2026-05-18T01:14:00Z",
  "publishedText": "4 hours ago",
  "thumbnailUrl": "https://news.google.com/api/attachments/CC8iK0Nn...",
  "searchQuery": "Tesla USA",
  "scrapedAt": "2026-05-18T05:30:00.000000+00:00",
  "error": null
}

Regional targeting

Google News results vary by language and country. Set both fields to get the right regional feed.

UK tech news:

{
  "query": "AI regulation",
  "language": "en-GB",
  "country": "GB"
}

German business coverage:

{
  "query": "Volkswagen",
  "language": "de",
  "country": "DE"
}

Brazilian Portuguese news:

{
  "query": "Petrobras",
  "language": "pt-BR",
  "country": "BR"
}

Japanese tech news:

{
  "query": "Sony",
  "language": "ja",
  "country": "JP"
}

Leave ceid blank in all cases. The actor builds it from country and language automatically.

Common language and country codes

Region	language	country
US English	en-US	US
UK English	en-GB	GB
German	de	DE
French	fr	FR
Spanish (Spain)	es	ES
Spanish (Mexico)	es-419	MX
Portuguese (Brazil)	pt-BR	BR
Japanese	ja	JP
Korean	ko	KR
Chinese (Simplified)	zh-CN	CN
Italian	it	IT
Dutch	nl	NL
Polish	pl	PL
Russian	ru	RU

What people use it for

Brand monitoring - run your company name, product names, and key competitors as a query list. Schedule the actor daily or weekly to track coverage over time.

SEO and content research - see which publishers dominate news results for a keyword. Useful for identifying where to pitch stories or which outlets to monitor.

Market intelligence - collect news around earnings, product launches, regulatory changes, or industry events. Pipe results into a spreadsheet or database for analysis.

Competitor tracking - run competitor brand names as queries. The searchQuery field lets you group results by competitor in your downstream tool.

Journalist research - quickly gather source material, check who has covered a story, and see the publication timeline across outlets.

News datasets - build labeled training data, news corpora, or topic-specific archives. Each article includes the source name and publish datetime for filtering.

Alerts and notifications - schedule the actor on a cron and connect it to a Zapier or Make webhook to push new articles into Slack, email, or any other channel.

Scheduling

You can schedule this actor to run automatically from the Apify console:

1. Open the actor and click Schedule
2. Set the interval (e.g. every 6 hours, once a day)
3. The dataset accumulates results across runs

To get only new articles since the last run, filter by publishedAt in your downstream tool using the timestamp from the previous run.

Output fields reference

Field	Type	Description
articleTitle	string	Article headline
sourceName	string	Publisher name
articleUrl	string	Direct URL to the original article
googleNewsUrl	string	Full Google News read URL
publishedAt	string	ISO 8601 publish datetime (UTC)
publishedText	string	Relative time label (e.g. "Yesterday", "3 hours ago")
thumbnailUrl	string	Thumbnail image URL from Google News CDN
searchQuery	string	The query that returned this article
scrapedAt	string	ISO 8601 timestamp of when the article was scraped (UTC)
error	string	Error message if extraction failed for this item, otherwise null

Performance

A single query typically returns up to 100 articles in under 30 seconds. Google News shows up to 100 results per search page.

For multi-query runs, the actor processes each query sequentially. A 5-query run with 100 articles each completes in roughly 2-3 minutes.

Set maxItems lower if you only need the most recent headlines and want faster runs.

Frequently asked questions

Can I scrape Google News without a search query? The actor requires at least one search term. Google News does have topic feeds and top stories sections, but those require different URL patterns. This actor is focused on keyword search results.

Are the article URLs direct links to the publisher? Yes. The actor extracts the real article URL from each result, not the Google News redirect URL. The articleUrl field points straight to the publisher's page.

How many results does Google News return per query? Typically around 100 results per search page. The actor collects whatever Google News shows for your query, up to the maxItems limit you set.

What happens if one query fails? The actor logs the error and continues to the next query. Results for successful queries are still saved to the dataset. No partial run crashes the whole job.

Can I run the same query for multiple countries at once? Not in a single run - country and language apply to all queries in a run. To get US and UK results for the same keyword, run the actor twice with different country settings, or use the Apify scheduler with different input configurations.

Does it collect the full article text? No. The actor collects the data visible on the Google News search results page: title, source, URL, publish date, and thumbnail. To get full article text you would need to follow the articleUrl and scrape the publisher's page separately.