Google News Monitor - Full Text & Real URLs
Pricing
Pay per event
Monitor Google News for keywords, site: queries and topics in any country; decoded publisher URLs, optional full text, only-new mode.
Google News searches, one per line. Google News operators work: "exact phrase", OR, -exclude, site:reuters.com, intitle:word, when:1d, after:2026-09-01 and before:2026-09-30. Each query is read in every edition below. An article found by several queries or editions is returned once, with all of them in matchedQueries and matchedEditions. Leave empty when you only monitor topics.
[]Optional Google News sections to monitor in every edition: TOP (top stories), WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE or HEALTH. Rows from a section have the query topic:NAME, for example topic:TECHNOLOGY. Unknown names are skipped and listed in the run status.
[]Google News editions to read, as COUNTRY:language codes: US:en, GB:en, IN:en, AU:en, CA:en, CA:fr, DE:de, FR:fr, ES:es, IT:it, NL:nl, PL:pl, JP:ja, BR:pt-419, MX:es-419 and other editions Google News offers. en-US style codes are accepted and UK is read as GB. Each edition is one feed per query, so 3 queries x 2 editions read 6 feeds. Google serves another edition for a pair it does not offer (for example PL:en returns US:en); such an edition is skipped, returns no articles, is not charged and is named in the run status. If none of the editions exists, the run fails.
[ "US:en"]Keep only articles published within this period before the run. The Actor adds the matching when: operator to every query that has no when:, after: or before: of its own, and then checks the publication date of every article again against the time range and the query's own when:, after: and before: (the last two with one day of tolerance), because Google News sometimes returns older articles (on 30 September 2026, 8 of 100 results of site:bbc.co.uk when:1d were from 2011 to 2025). Topic sections are filtered by the same date check.
Most articles to return in one run, newest first, across all queries, topics and editions. One Google News feed holds about 100 articles at most, so this is the practical maximum per query and edition.
Return only articles that an earlier run with the same queries, topics, editions and time range has not returned yet. The memory is kept per combination of those settings in your account, so two monitors with different queries never hide each other's articles, and only articles that were actually returned are remembered. Use it with a schedule to get a stream of fresh news.
Google News links point to news.google.com, not to the publisher. When on, the Actor decodes every returned article to its original publisher URL (originalUrl) and checks site: queries against it too. This takes one request per article plus one per 10 articles to Google News (about 1.1 per article), so a run of 100 articles takes one to two minutes longer. Always on when Full text is on.
Open each article on the publisher's site and extract the article text, author, main image and summary (snippet). Plain HTTP, no browser: paywalled and script-rendered pages return partial or no text, which fullTextStatus reports per article. Adds roughly one second per article.