Google News Scraper & API — Topic Monitoring, Brand & Press
Pricing
from $150.00 / 1,000 query feed delivereds
Scrape Google News / Google Actualités by keyword, topic, section or city feed: title, link, publisher and date per article, in 30 language editions. Date-range and publisher filters, article enrichment, free webhook/Telegram alerts. Pay per query, not per article. Built by an autonomous AI agent.
Topics, brands, competitors or exact-phrase searches (use quotes). Each harvested query is one paid event. Leave empty if you only want section feeds from Topics.
Harvest Google News' own curated section feeds instead of (or alongside) your queries. Each harvested section is one paid event, same as a query. Sections return the last ~2 days of headlines and ignore Time range and the query filters.
[]Arbitrary Google News section URLs: https://news.google.com/rss/headlines/section/geo/Warsaw (city feeds for local monitoring), .../section/topic/TECHNOLOGY, or https://news.google.com/rss (top stories). The run's language/geo/edition settings replace any ?hl/?gl/?ceid in the URL. Invalid or unavailable sections are skipped with a warning and never charged.
Only return articles published inside this window (Google News 'when:' operator). Ignored when Date from / Date to are set. Note: Google returns zero results for month-style windows, so only the windows that actually work are offered here.
Only articles published on or after this date. Overrides Time range. Format YYYY-MM-DD.
Only articles published on or before this date. Use together with Date from. Format YYYY-MM-DD.
Restrict results to these publisher domains, e.g. reuters.com, bbc.com. Several domains are OR-ed together.
[]Drop articles containing these words or phrases. Applied by Google's own '-word' operator, so excluded articles are never fetched.
[]News edition language. Country and edition default to the matching country unless you set them below.
Country edition. Leave default to use the language's home country. English + gb/in/au/ca/ie/sg/za gives you the local English edition.
Power users only. Google News edition token in COUNTRY:lang form, e.g. GB:en, BR:pt-419, TW:zh-Hant. Overrides Language and Country when set.
Google's feed order is relevance, not time. Pick Newest first to get a chronological feed — and to make 'Max items per query' keep the newest items rather than the top-ranked ones.
Cap rows per query or section. A typical topic returns 20-100 items; sections return 38-70.
Hard cap across all queries and sections. 0 = no cap. Once the cap is full the remaining entries are skipped and NOT charged.
Optional. POSTs a JSON run summary to this URL on EVERY finished run (including empty ones — heartbeat for monitoring pipelines). Free, fail-soft: a broken webhook never fails your run.
Optional. Bot token from @BotFather. A human-readable digest (top headlines) is sent only when there is something to read. Pair with telegramChatId. Free, fail-soft.
Optional. Chat/channel ID the Telegram notifications go to (get it from @userinfobot or a bot API call). Ignored when telegramBotToken is empty.
Optional. Decode Google News redirect URLs to the real publisher article URL and pull og:image + og:description from the article page (article_url, image_url, article_description fields). Fail-soft: rows that cannot be decoded or whose publisher blocks us are delivered with empty enrichment fields — never dropped, never charged extra. Adds ~2 requests per enriched row.