Google News Scraper
Pricing
from $0.01 / 1,000 articles
Scrape Google News articles by search, topic, place or URL in 82 editions (70 countries): title, publisher, date, related coverage, original URL and full text. More than 100 articles per search. Export to JSON, CSV or Excel.
news.google.com pages or RSS feeds, as copied from the browser: a search (https://news.google.com/search?q=tesla), a topic or section, a local news page, a Full coverage story, a publisher page or any news.google.com RSS feed. The country / language of the URL (hl, gl, ceid) is kept. Added to the searches, topics and places below. Max 1 000 URLs.
[]What to search, exactly as on Google News. Search operators work: "exact phrase", OR, -word, site:reuters.com, intitle:word (e.g. artificial intelligence -crypto).
Several searches in one run (one search per query and per edition below). Added to Search query; an article found by several searches is saved once.
[]Google News sections to read for each edition: top stories, world, business, technology… (about 40 articles for the front page, 60-75 for the other sections, per edition; most with their related coverage).
[]City whose local news to read (Google News Local section, e.g. Paris, New York, Munich), or a country in its own edition (Germany with DE:de). Google News has no local news for a region or a state (Bavaria, Texas: 0 articles). Read once per edition; not combined with the search queries.
Several places in one run: local news of each place. Added to Local news for a place.
[]Google News edition used for the searches, topics and places: it decides the country and the language of the articles.
Run every search, topic and place in several editions at once (e.g. the same query in France, Germany and Japan). Added to Country and language. A run reads at most 1 000 searches: each search, topic or place counts once per edition, each start URL once.
[]Maximum number of articles to save for the whole run (after deduplication and filters). 0 = no limit.
Cap for EACH search, topic, place or URL (per edition), so that the first one cannot use up the whole Max articles budget. 0 = no per-search cap.
Google News returns at most 100 articles per search. When a search reaches 100, the Actor splits its date range into smaller periods (down to one day) and reads each one, so a search can return thousands of articles. The range is Posted after → Posted before, or the last Look-back days when they are empty. No effect on topics and places.
How far back to split a search that reaches 100 articles, when Posted after is empty: that many days before Posted before, or before the run's start. Up to 3 650 days (10 years).
Order of the articles of each search, topic or place: as ranked by Google News (relevance), or newest first.
Replace the Google News redirect link by the real article URL on the publisher's site (articleUrl), resolved through Google News itself. Charged per article resolved (see Pricing).
Open each article on the publisher's site and add its full text, word count, reading time, main image, author(s), summary, keywords, section and last-update date. Includes the original article URL. Some publishers refuse robots or show a paywall: the article is then saved without these fields and this option is not charged for it.
Add the other articles that Google News groups with each article (same story, other publishers: title, publisher, link), free of charge.
Only articles from this recent period (Google News when:). For an exact range, use Posted after / Posted before instead.
Only articles published on or after this date: 2026-09-01, or a period before now such as 7 days, 2 weeks, 1 month (API: 24 hours and full ISO date-times work too). Reads the publication date given by Google News.
Only articles published on or before this date (the whole day is included), or older than a period such as 30 days.
Keep only articles from these sites (domain, e.g. reuters.com, bbc.co.uk). Added to every search as site:, and checked on topics and places.
[]Drop articles from these sites (domain, e.g. yahoo.com).
[]Match the search query in the article title only (Google News intitle:).
Drop the articles whose title contains one of these words or phrases, as whole words: war drops "War in…" but keeps "Warner" (case and accents ignored; in Chinese, Japanese, Korean, Thai… any part of the title counts).
[]Skip the articles that a previous run (same Memory key) already delivered: they are not saved and not charged, and their page is not even opened. First run = everything is new.
Name of the memory used by Only new articles. Give each schedule / task its own key (e.g. tesla-alerts) so that they do not share their memory. Letters, digits, - and _.
Forget everything remembered under this Memory key before the run: this run returns (and charges) every article again. Untick it afterwards.
Apify Proxy or your own proxies. Keep the default: it is included in the price. The residential Apify proxy is not available in this Actor.
{ "useApifyProxy": true}Maximum number of requests processed in parallel.
Most requests to Google News and to publisher sites in any 60 seconds, all of them counted. A budget, not an even pace: up to this many can leave at once when the minute starts (spread them with the minimum delay below). Lower it if the log shows HTTP 429.
Smallest gap between two requests, in milliseconds. Unlike the per-minute rate, this spreads the requests evenly instead of letting them go out in a burst. 0 = no gap.
Retries per request before it is marked as failed. Behind a proxy, a request the site turns away is also retried on a new proxy session up to 10 times without using up these retries.