Google News Scraper - Headlines, Sources, Dates

Get Google News search results as clean structured data: headline, publisher, publisher domain, publish date (UTC), Google News link and the related coverage cluster for every story. Search any keyword or Google News operator, pick a country edition, a language and a time window, and export to JSON, CSV, Excel or the Apify API.

No API key, no login, no browser. Pay only for delivered articles: $1.00 per 1,000 articles, with an almost-zero start fee ($0.00005), so small scheduled monitoring runs stay cheap.

Why use it

Manually	With this actor
Open Google News, search, copy headlines one by one	One run returns up to 100 articles per query
Dates shown as "3 hours ago"	ISO 8601 UTC publishedAt
Same story repeated across queries	Deduplicated across queries (not charged twice)
Hard to track 20 brands/competitors	20 queries in one scheduled run, sent to Sheets/Slack/webhook

Output example (real row)

{
  "title": "OpenAI connects the Dots",
  "source": "Platformer",
  "sourceUrl": "https://www.platformer.news",
  "sourceDomain": "platformer.news",
  "publishedAt": "2026-09-30T02:25:34+00:00",
  "googleNewsUrl": "https://news.google.com/rss/articles/CBMia0FVX3lx...?oc=5",
  "articleId": "CBMia0FVX3lx...",
  "position": 13,
  "relatedArticles": [
    {"title": "OpenAI launches Dots, its Muse competitor", "source": "The Verge", "googleNewsUrl": "https://news.google.com/rss/articles/CBMimgFB...?oc=5"}
  ],
  "relatedCount": 2,
  "query": "openai",
  "language": "en-US",
  "country": "US",
  "timeRange": "any",
  "scrapedAt": "2026-09-30T10:00:00+00:00"
}

Output fields

Field	Type	Description
title	string	Headline (publisher suffix removed)
source	string	Publisher name
sourceUrl / sourceDomain	string	Publisher homepage and domain
publishedAt	string	Publish date, ISO 8601 UTC
googleNewsUrl	string	Google News article link (redirects to the publisher)
articleId	string	Stable Google News article ID (use it for dedup in your pipeline)
position	integer	Rank in the Google News result list
relatedArticles / relatedCount	array / integer	Other outlets covering the same story (story cluster)
query, language, country, timeRange	string	The search that produced the row
scrapedAt	string	Extraction timestamp

How to use

1. Add one or more Search queries (keywords, brand names, "exact phrase", site:reuters.com, -exclude).
2. Choose Time range, Country and Language (defaults: any time, US, en-US).
3. Click Start.
4. Download results as CSV, JSON, Excel, or read them via the API / integrations.

Input

Field	Default	Description
queries	demo: artificial intelligence, openai	Keywords or Google News operators
maxItemsPerQuery	100	1-100 (Google News RSS returns up to ~100 per query)
timeRange	any	any, 1h, 1d, 7d, 30d, 1y
language	en-US	e.g. en-GB, es, de, fr, pt-BR
country	US	Two-letter edition, e.g. GB, ES, DE, IN
deduplicateAcrossQueries	true	Each article delivered and charged once per run

Example: daily brand monitoring

{"queries": ["\"Acme Corp\"", "\"Acme\" lawsuit", "Acme competitor"], "timeRange": "1d"}

Example: Spanish market news

{"queries": ["inteligencia artificial", "Banco Santander"], "language": "es", "country": "ES", "timeRange": "7d"}

Use cases

• Brand, competitor and executive media monitoring
• PR and share-of-voice reporting by publisher
• News datasets for LLM / RAG pipelines and sentiment analysis
• Market and investment research alerts (ticker, company, sector)
• Lead signals: funding, expansion or hiring news about target accounts

Integrations

Schedule the actor daily or hourly and send results to Google Sheets, Slack, Zapier, Make, n8n, Airbyte, LangChain or any webhook. Use articleId to keep only new stories between runs.

Pricing

Pay per event:

Event	Price
Actor start	$0.00005
Article delivered	$0.001 ($1.00 / 1,000)

Duplicates, invalid queries and queries with no results are not charged. If you set a maximum cost per run, the actor stops delivering exactly at that limit.

FAQ

Does it need a Google API key or proxy? No. It reads Google News public RSS search feeds.

How many articles per query? Google News returns up to about 100 per query. For more, split into narrower queries or use time ranges.

Do I get the full article text? No. You get headline, publisher, date, link and related coverage. The link redirects to the publisher's page.

Can I use search operators? Yes: "exact phrase", site:, -word, OR work as in Google News.

What if a query has no results? It is listed in noResults in the run OUTPUT summary and nothing is charged. Queries are never replaced with other data.

Are results the same as the Google News website? They come from Google's own feed for the chosen edition; order can differ slightly from the web UI.

Can I monitor news continuously? Yes, schedule it and deduplicate by articleId.

Troubleshooting

Problem	Solution
Fewer than 100 articles	Google has fewer matches; widen the time range or query
Run failed with "Invalid input"	Check language (e.g. en-US), country (two letters) and timeRange values
Query in failed in OUTPUT	Temporary Google error after retries; re-run. The run fails only if every query failed
Wrong language results	Set both language and country to the same market