Google News Scraper: Real Article URLs (No Login)

Google News Scraper returns recent Google News results for your search queries and topic sections as JSON, CSV or Excel. Give it up to 10 queries or topics (Top stories, Business, Technology, …), an edition (country and language) and a period, and get back for each article the title, the outlet and its site, the outlet's own article URL (articleUrl) so a row links straight to the publisher instead of to Google, the publication time and the article's Google News link. For topic sections it also returns other outlets' coverage of the same story, as Google News groups it.

It needs no login and no API key, and it does not open the publishers' pages. The same article found by two queries is returned once, articles left out by a filter are not charged, and a scheduled run can return only the articles it has not returned before.

This Actor is an independent tool. It is not affiliated with, endorsed by or sponsored by Google. Try it with an empty input: it returns a small sample (20 articles about "artificial intelligence" from the last 24 hours, US English edition).

Use cases

• Brand and competitor monitoring. One query per brand, an hourly or daily schedule, and only the articles that appeared since the last run.
• Topic briefings. A daily pull of BUSINESS and TECHNOLOGY with the other outlets' coverage of each story, ready for a digest.
• Media analysis. Outlet mix, publication times and how widely a story was covered, as a table.
• Research and datasets. A news corpus for a period and an edition, de-duplicated across queries.
• Agent and MCP use. A short query with maxItems returns headlines, outlets and article URLs for an assistant to cite.

What data can you get?

Field	Description
id	Stable article ID (hash of the feed's article ID); used for de-duplication
source	Google News
query	Your query that found the article (the first one, when several did); null for topic rows
topic	The topic section that listed the article, for example TECHNOLOGY; null for query rows
edition	The edition searched, for example US:en
title	Article title, without the " - outlet" suffix
outlet	Outlet name, for example Reuters
outletUrl	Outlet's site, for example https://www.reuters.com
publishedAt	Publication time, ISO 8601 (UTC)
articleUrl	The outlet's own article URL, for example https://www.reuters.com/world/...; null when it could not be resolved
articleUrlReason	Fixed reason when articleUrl is null (not resolved, or not checked because resolveArticleUrls is off); null when the URL is there
googleNewsUrl	The article's Google News link
relatedCoverage	Topic rows: other outlets' coverage of the same story (title, outlet), as Google News lists it (usually up to 4); empty for query rows
isNew	With Only new articles: true; otherwise null
scrapedAt	When this run read the feed

Every row has every field; a value the feed does not provide is null. Article text and author names are not returned.

The outlet's own article URL is returned for every row by default (resolveArticleUrls: true): the run returns one row per article with articleUrl filled in, which is why it takes a little longer than a run that returns the Google News link alone. A link that cannot be resolved does not lose its row: articleUrl is null and articleUrlReason says why. Set resolveArticleUrls: false to get the Google News link only.

How to use Google News Scraper

1. Enter one or more Search queries (queries), or pick Topics (topics), or both.
2. Optionally pick the Edition (edition) and the Period (timespan), and narrow with Published after (publishedAfter), Only these outlets (includeOutlets) or Exclude outlets (excludeOutlets).
3. Set Maximum results (maxItems) and run the Actor. Resolve the outlet article URL (resolveArticleUrls) is on by default; turn it off for a faster run that returns only the Google News link.
4. Download the dataset as JSON, CSV or Excel, or read it through the API.

Queries

Queries use the Google News search syntax:

• Exact phrases go in double quotes: "interest rates".
• OR between terms: tesla OR byd OR rivian.
• A minus sign excludes a word: apple -fruit.
• site: keeps one outlet: chip exports site:reuters.com.
• The period is added as when:24h (from timespan). A query with its own when: keeps it.

Topics

topics reads the edition's topic sections, one request each: TOP (top stories), WORLD, NATION (the edition's country), BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE, HEALTH. A topic section lists the latest stories (Period does not apply; use publishedAfter), and each row carries relatedCoverage: the other outlets Google News groups under the same story, a quick signal of how widely a story is covered. When topics is set and queries is left at its default, only the topics are read.

Filters

• publishedAfter: keep articles published at or after an ISO 8601 date or date-time (UTC when no zone is given). For hourly schedules, set it to the previous run's start.
• includeOutlets / excludeOutlets: outlet domains (a domain also matches its subdomains: reuters.com matches uk.reuters.com) or exact outlet names, up to 50 each.
• Articles left out by a filter are not written and not charged.

Split by day (more than 100 articles per query)

splitByDay: true searches each query one day at a time (Google News' after: / before: operators), newest day first, so a query can return more than the feed's 100 articles per request. It is off by default.

• At most 7 one-day windows per run in total: the number of days is the Period in days (hours count as one day), divided among the queries (for example 7 days for one query, 3 days each for two). Up to 7 queries.
• Windows are requested one at a time, 2 seconds apart. Neighbouring windows overlap by a day; an article found twice is returned and charged once.
• A query with its own when:, after: or before: is not split.

Editions

US:en, GB:en, CA:en, AU:en, IN:en, JP:ja, DE:de, FR:fr, ES:es, IT:it, BR:pt-419, MX:es-419, KR:ko, TW:zh-Hant, NL:nl. The edition decides which outlets and which language Google News favours.

Input example

{
    "queries": ["electric vehicles", "\"interest rates\""],
    "timespan": "7d",
    "edition": "GB:en",
    "maxItems": 50
}

Input

Field	Default	What it does
queries	["artificial intelligence"]	Search queries (up to 10)
topics	[]	Topic sections: TOP, WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE, HEALTH
timespan	24h	Period for queries: 1 to 48 hours or 1 to 30 days; does not apply to topics
edition	US:en	Country and language edition
publishedAfter	empty	Keep articles published at or after an ISO 8601 date or date-time
includeOutlets	[]	Only these outlet domains or exact names (up to 50)
excludeOutlets	[]	Drop these outlet domains or exact names (up to 50)
splitByDay	false	Search each query one day at a time, for more than 100 articles per query
onlyNewItems	false	Return only articles that earlier runs have not returned
maxItems	20	Maximum rows per run, across all queries (1 to 1,000)
resolveArticleUrls	true	Look up the outlet's own article URL for every returned row
proxyConfiguration	direct connection	Proxy settings (country, groups, custom proxy URLs)

Output

One row per article, every field present. Values below are illustrative.

Output example

{
    "id": "8d1f2a6c0b9e4f37",
    "source": "Google News",
    "query": "electric vehicles",
    "topic": null,
    "edition": "GB:en",
    "title": "EV sales rise for the third month in a row",
    "outlet": "Example News",
    "outletUrl": "https://www.example-news.com",
    "publishedAt": "2026-11-04T09:15:00.000Z",
    "googleNewsUrl": "https://news.google.com/rss/articles/CBMi...?oc=5",
    "articleUrl": "https://www.example-news.com/2026/11/04/ev-sales-rise",
    "articleUrlReason": null,
    "relatedCoverage": [],
    "isNew": null,
    "scrapedAt": "2026-11-04T10:00:00.000Z"
}

New articles only

Run the same input on a schedule and get only what is new:

• Only new articles (onlyNewItems: true): returns only articles that no earlier run with the same queries and edition returned. Each returned row has isNew: true. When nothing is new, the run ends successfully with an empty dataset and the status message No new items since the previous run (…).
• The Actor remembers what it returned in a named key-value store in your Apify account (state-google-news-…), one record per query, keyed by a hash of the queries and the edition. Changing those starts a new history; changing timespan or maxItems does not.
• An article counts as returned only after it was written to the dataset. Articles left out by maxItems or by your maximum cost come back as new on the next run.
• Articles not seen for 90 days are forgotten (they would count as new again).
• Do not run the same input twice at the same time: both runs read the same history, so both return (and charge) the same new articles.
• Runs paid for with a temporary account may not keep their history beyond that account's lifetime.

How much does it cost?

Google News Scraper uses pay-per-event pricing: you pay for the articles it returns, not for the time it runs. At the Free tier 1,000 articles cost $1.70; at Gold, $1.10.

Apify plan (discount tier)	Price per 1,000 articles	Price per article
Free	$1.70	$0.0017
Starter (Bronze)	$1.50	$0.0015
Scale (Silver)	$1.30	$0.0013
Business (Gold)	$1.10	$0.0011

Apify's Free plan includes USD 5 of monthly usage, which covers about 2,941 articles at the Free tier price, run start fees aside.

• Run start fee: the platform default per run (see the Pricing tab).
• Articles dropped as duplicates across queries or by the timespan filter are not charged.
• Resolving the publisher URL is included in the article price; it is not a second charge.
• With onlyNewItems, articles returned by earlier runs are not charged again.
• Maximum results (maxItems) caps the total rows written.

Use with AI agents (MCP)

Google News Scraper works as a tool for AI agents such as Claude, Cursor and n8n agents through the Apify MCP server. Load only this Actor with this server URL:

https://mcp.apify.com?tools=kizukidata/google-news-feed

• Claude Code: run claude mcp add apify "https://mcp.apify.com?tools=kizukidata/google-news-feed" -t http.
• Claude (web or desktop app): open Customize → Connectors, choose Add custom connector, name it Apify and paste the server URL above.
• Cursor: add the server to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):

{
    "mcpServers": {
        "apify": { "url": "https://mcp.apify.com?tools=kizukidata/google-news-feed" }
    }
}

• n8n: add the Apify node (on self-hosted n8n, install the community package @apify/n8n-nodes-apify) and choose Run an Actor and get dataset; for an n8n AI Agent, add Apify Tool as a tool.

The server signs you in to Apify with OAuth. A client that cannot use OAuth can send your Apify API token in an Authorization: Bearer <APIFY_TOKEN> header instead.

What to send from an agent

Calling it with no arguments ({}) runs the small default sample. Pass queries (or topics) and maxItems; add edition, timespan or publishedAfter to narrow the results. The dataset has a compact view (publishedAt, title, outlet, articleUrl) for short answers. maxItems also bounds how many articles the run resolves a publisher URL for. When a run returns nothing, the dataset stays empty and the run's status message says why, and also names input fields it did not recognise, such as a mistyped key.

Integrations and API

Run the Actor from your own code or from an automation tool. Replace <APIFY_TOKEN> with your API token.

n8n, Make and Zapier

Use the Apify integration, choose Run Actor, and pass queries with onlyNewItems: true. Add Get dataset items and map title, outlet, publishedAt and articleUrl to a Slack message, a sheet, a digest or a model step. A daily schedule is usually enough for a news monitor; hourly works for a fast-moving topic.

cURL

curl -X POST "https://api.apify.com/v2/acts/kizukidata~google-news-feed/runs" \
  -H "Authorization: Bearer <APIFY_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"queries":["electric vehicles"],"edition":"GB:en","timespan":"7d","maxItems":50}'

Poll GET https://api.apify.com/v2/actor-runs/<RUN_ID> for the status, read defaultDatasetId, then fetch GET https://api.apify.com/v2/datasets/<DATASET_ID>/items?format=json.

Limitations

• No article text, no author names, no images. articleUrl is the article's address, not its content: the Actor does not open the outlet's page.
• articleUrl is looked up through Google News, so a publisher that Google links to in an unusual way, a lookup that Google refuses, or a lookup that runs out of time leaves that row's articleUrl empty (articleUrlReason says which).
• Titles are as Google News lists them, in the edition's language mix. The Actor does not translate them.
• The feed decides which articles it lists and in which order; it is not a complete archive of everything published, and it is by relevance, not strictly by date. Sort by publishedAt when you need the newest first.
• One request per query or topic (or per day with splitByDay). The feed lists at most 100 articles per request, so a run returns at most 100 × the number of requests.
• Each query gets a fair share of maxItems; a query with fewer articles passes the rest on to the queries after it.
• Maximum results (maxItems, default 20) caps each run, across all queries. Runs started by AI agents usually have no spending limit, so this is the main cost control.
• If Google News answers "too many requests" or shows a consent page, the run stops at once: the queries, topics or days not read yet are listed in OUTPUT uncheckedInputs, and the articles already read are kept. A query with no matching articles ends successfully with an empty dataset and says so; a run fails only when it could not return any article because of a problem on our side or the feed's.
• relatedCoverage lists what Google News groups under a story in topic feeds (titles and outlets only); search results carry none.
• splitByDay relies on Google News' after: / before: operators; Google decides what each day returns.

FAQ

Can I get only new articles since the last run?

Yes: set onlyNewItems to true and run the same input again (for example on a schedule). See New articles only.

Why did a query return nothing?

An accepted query with no articles in the period returns none. Try a longer timespan, fewer words, or another edition. Check the OUTPUT record for skipped queries.

Why is articleUrl empty on a row?

The lookup through Google News did not give that article's publisher URL (it changed shape, or Google refused a request). The row is still returned; articleUrlReason says which. The Google News link (googleNewsUrl) opens the article too. If every row is empty and the reason says resolveArticleUrls is off, turn that input on.

Why do I get fewer articles than I asked for?

The feed lists at most 100 articles per request, and a period or an outlet filter can leave fewer. With splitByDay one query can go past 100 by walking day by day; without it, widen the period or add another query.

Is it legal to use this data?

The Actor reads Google News search results and topic sections that anyone can open, without logging in, and does not open the outlets' pages. Whether you may collect and use this data depends on what you do with it and where you are: Google's terms, the publishers' copyright and the data-protection rules that apply to you are your responsibility. If you republish an article, follow the publisher's terms. This is not legal advice.

Related Kizuki Actors

Other Actors from Kizuki that work well with this one:

• Telegram Channel Scraper — channel posts on the same story, with views and reactions.
• YouTube Channel Scraper — the video coverage of the same story.

Your feedback

Bugs, a query that returns the wrong language, and feature requests go to the Actor's Issues tab in Apify Console. A publisher who wants its articles left out can open an Issue with the outlet's domain; the exclusion is applied in the next build, and a request from a third party is not enough.