News monitoring gets expensive fast when every article costs a full-fat scrape. Most monitoring jobs do not need full articles; they need headlines, sources, and links for a hundred keywords, every morning. Google News already aggregates that, so I built the Google News Lite API on Apify to read it cheaply: send search terms, get one JSON row per matching article.
Disclosure: the Apify links in this post are affiliate links. If you run the Actor, I may earn a referral commission at no extra cost to you.
Does Google News have an API?
No. Google retired its official News API years ago and never replaced it, so the question "how do I get a Google News API" keeps getting asked and keeps having the same answer. RSS feeds exist, but they were built for feed readers, not for monitoring: filtering is limited and running a hundred keyword feeds gets unwieldy. The working answer is a scraper you call like an API: search terms in, structured headline rows out.
What the Google News Lite API returns
The Google News Lite API returns one row per article as structured JSON: title, link, source, snippet, date, and image, plus the rank, the matching query term, and your targeting metadata.
| Field | Example | Notes |
|---|---|---|
title |
"Apple Vision Pro gets enterprise push" | The headline |
link |
article URL | Direct to the publisher |
source |
"The Verge" | Publisher name |
snippet |
short excerpt | Enough for triage and sentiment |
date |
"3 hours ago" | Relative publication date |
image |
thumbnail URL | When available |
Rows are de-duplicated by URL per query, and each run collects up to about 100 articles per search term.
Who this is for
PR and comms teams watching brand and executive coverage, analysts tracking markets or technologies for breaking developments, and developers feeding fresh headlines into agent workflows, alerting, or NLP pipelines.
The manual way, and where it breaks
Scraping Google News directly starts with consent pages and ends with parser archaeology. The markup rotates, dates arrive as relative strings in whatever language the region serves, and the same story appears five times under five URLs. By the time you have handled localization, de-duplication, and blocking, your "quick headline fetch" has a backlog. I keep a graveyard folder of half-finished news parsers; it is not small.
The faster way: run the Google News Lite API
Apify Console
- Open the Google News Lite API and click Try for free.
- Add your
searchTermsand pick atimeRangelikeday. - Run it and download the dataset as JSON, CSV, or Excel.
REST
curl -X POST "https://api.apify.com/v2/acts/johnvc~google-news-lite-api/runs?token=YOUR_APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{ "searchTerms": ["Apple Vision Pro", "OpenAI"], "timeRange": "day", "maxResultsPerSearch": 25 }'
Run endpoint reference: the Apify API docs.
Monitor headlines in Python
from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("johnvc/google-news-lite-api").call(
run_input={
"searchTerms": ["Apple Vision Pro", "OpenAI"],
"timeRange": "day",
"maxResultsPerSearch": 25,
}
)
for article in client.dataset(run["defaultDatasetId"]).iterate_items():
print(article["title"], "|", article["source"], "|", article["link"])
Each term is searched separately, so the output is already grouped for per-keyword reporting.
Build a daily news digest feed
The digest is the classic setup: a fixed term list, timeRange set to day, one run every morning. The task Build a daily news digest feed by API is that configuration ready to clone.
Bulk monitor 100 keywords at once
Per-article billing is what makes wide keyword nets affordable, since a quiet keyword costs almost nothing. Bulk monitor news for 100 keywords at once shows the wide-net pattern.
It works in other languages too
Country and language are input fields, not an afterthought. Two Chinese-language tasks prove the point: a daily Chinese news digest and bulk brand monitoring in Chinese. Swap country and language and the same recipes run for any market.
Use it from Claude and other MCP clients
Because the Actor is MCP-ready, Claude, Claude Code, and Cursor can pull fresh headlines mid-conversation: "what happened around our brand in the last 24 hours" becomes a tool call with real citations. The task Feed Google News headlines to Claude via MCP has the setup, and you can read more about Claude and Claude Code at claude.ai.
FAQ about scraping Google News
Is this Google News scraper free?
It bills per article returned, with no setup fee, and maxResultsPerSearch caps each term at up to 100 articles so spend is bounded before the run starts. New Apify accounts include free platform credit, which covers a lot of headlines.
How is the Lite scraper different from the full Google News API?
Lite is the cheap, headline-level variant: title, link, source, snippet, date, and image, capped at 100 articles per term, with country and language targeting. The full Google News API adds location-based searches, safe search, and pagination control. If you need those, use the full version; if you are monitoring keywords at volume, Lite is the better price.
Does the scraper return full article text?
No, and that is a deliberate trade. You get the headline, snippet, source, and a direct link; if a story matters, fetch the article from the link. Keeping rows headline-level is what keeps 100-keyword monitoring cheap.
What does a media monitoring scraper actually collect?
Coverage over time: who wrote about a term, when, where, and with what framing in the headline and snippet. Stack daily runs and you have share-of-voice, source breakdowns, and a timeline of how a story developed.
Can Claude call the news scraper through MCP?
Yes. Connect the Apify MCP server and the Actor appears as a callable tool, so agents and chat sessions can ground themselves in the day's coverage instead of last year's training data.
Can I schedule the scraper for an automatic daily digest?
Yes, that is its natural habitat. Save your term list as a task, attach an Apify schedule (a cron line like 0 7 * * *), and read the fresh dataset each morning. Start from the Google News Lite API.
More from Truffle Pig Data
Related Actors for adjacent feeds: the full Google News API when you need deeper search control, the Google Finance API for the market-data side of a monitoring stack, and the Google Forums Search API for what discussion boards say before the press does.
Wrapping up
Headline monitoring should be boring and cheap, not a scraping project. The Google News Lite API keeps it that way: search terms in, one row per article out, billed per article.
Top comments (0)