Google News Scraper

Returns: one row per news article with title, url, source, publishedAt, googleNewsUrl, query, edition. Input: {"queries": ["nvidia earnings"], "editions": ["en-US"], "resolveUrls": true, "maxItems": 20} Best for: media monitoring, news sentiment datasets, competitor and brand mention tracking.

{
  "title": "À Val-de-Cognac, Garandeau transforme ses anciennes carrières en géant du solaire",
  "source": "Charente Libre",
  "publishedAt": "2026-10-12T18:28:56.000Z",
  "googleNewsUrl": "https://news.google.com/rss/articles/CBMi3gFBVV95cUxPYWZiX19IZnZISHpwYkxfOTJuUTcxYl9NckgxVzZMZEdDbUI4M1hUb3JWTkkzZGMza3BVbUlQZVRRbi1NbHREMy1RcGNpLUVhWlRLUW83QkV2OC1PeDRJdkdoVlhsNnBLRWU2TV8yOTJNYlRTWDhHRkc0eWxodEJJRjdDTlB0aHRRMl84bUI0WHI1OEp4Y2JwdGl5S290SHZMLWJvZWR1Q2IwTnk1RlVhQ3NUcUY2ekFrdG9rb0tkb19SdzBtRmZBNDR6Z09VQzdQN2k3TFB0YnBDUXFJYXc?oc=5",
  "query": "énergie solaire",
  "edition": "FR:fr"
}

What does Google News Scraper do?

Google News Scraper collects public article metadata from Google News keyword search feeds, news sections, and topic pages. It returns headlines, publisher names, publication times, Google News identifiers and links, and publisher article URLs when they are available. You can run one query, several queries, a named section such as TECHNOLOGY, or a Google News topic URL. The Actor works without a Google account or login.

The Actor uses Google News RSS feeds for keyword searches and the structured data in topic pages for sections and topics. Search feed links point to Google News article pages. With resolveUrls enabled, the Actor asks Google News for the publisher destination and stores it in url. Topic pages already include publisher destinations. The Actor does not fetch publisher websites or article bodies. Its output is suitable for headline monitoring, research queues, source comparison, and data pipelines that need links and dates.

Searches can include Google News operators such as quoted phrases, OR, site:, intitle:, after:, before:, and when:. You can also set a recent time range or explicit dates. Date ranges are split into daily search windows because one RSS search response has a limited number of items. Google News controls the order and coverage of results, so a date range does not guarantee every article published in that period.

What data can you get?

Each dataset row represents one Google News article. All rows contain a stable id and sourceUrl, the Google News feed or topic page used for that row. Optional fields use null or an empty array when the source does not provide a value. The Actor removes repeated Google News IDs across feeds and windows.

Field	Meaning
id	Stable identifier for this Google News article.
title	Headline, without the repeated publisher suffix used in some RSS titles.
url	Publisher article URL, if present on a topic page or resolved for a search result.
source	Publisher name shown by Google News, sometimes a bare domain such as investopedia.com.
sourceUrl	Google News RSS feed or topic page this row was read from. It is not the article; use url, or googleNewsUrl when url is null.
publishedAt	Publication time in ISO 8601 UTC when available. Date-only values appear as midnight US Pacific.
publishedAtPrecision	date when the source gave only a date, datetime when it gave a real time.
googleNewsId	Google News identifier used for deduplication.
sourceWebsite	Publisher home page from the feed's source element, or the origin of the article URL for topic rows. null when Google omits it.
domain	Hostname of the resolved publisher url without www.; null when url is null.
googleNewsUrl	Link to the article on Google News.
imageUrl	Image URL (topic and section rows only; null for keyword searches).
authors	Author names (topic rows only, when the page lists them; empty for keyword searches).
storyId	Story cluster identifier (topic rows only; null for keyword searches).
relatedArticles	Related articles when Google's RSS item lists a story cluster. Usually empty.
query	Search phrase that returned this article, or null in topic mode.
topic	Named section or topic identifier, or null in search mode.
inputType	search or topic.
edition	Country and language edition, such as US:en.
position	Item position within the source response.
scrapedAt	Time the Actor created the row.

The overview table shows the most useful columns for scanning results. Download the full dataset as JSON, CSV, Excel, or another format available in Apify. Image and publisher URLs are returned as data; this Actor does not visit those hosts. A publisher URL may be missing if Google News does not provide a usable destination or changes its link format.

How to use

1. Open the Actor input form and add at least one query, named section, or Google News topic URL.
2. Choose the editions you want to search. The default is en-US.
3. Set maxItems for the total number of records and maxItemsPerQuery for each search or topic in each edition.
4. Turn resolveUrls on when you need publisher article URLs from keyword searches. It is off by default, and resolved rows cost extra per result.
5. Run the Actor and open the Dataset tab to inspect or download the rows.

For a quick keyword test, use nvidia earnings with the prefilled values. To monitor a section, remove the query and add TECHNOLOGY or another section. To search an entity topic, supply its complete https://news.google.com/topics/... URL. You can combine these input types in one run. The total cap applies to the combined result set in input order, so put the most important source first when using a small cap.

The Actor checks input before making requests. Malformed date ranges, unknown editions, unsupported sections, and topic URLs outside Google News produce a clear error. Requests are paced at no more than two starts per second. Temporary Google News errors are retried with backoff. A consent page is not submitted or bypassed.

Input

Field	Type	Purpose
queries	String array	Keyword searches and Google News operators.
dateFrom, dateTo	Date strings	Inclusive daily search window range, both required together.
timeRange	Choice	Recent period: any, 1h, 1d, 7d, 30d, or 1y.
topics	String array	Uppercase named sections: TOP_STORIES, WORLD, NATION, BUSINESS, TECHNOLOGY, ENTERTAINMENT, SPORTS, SCIENCE, HEALTH.
topicIds	String array	Google News topic IDs beginning CAA; an alternative to full topic URLs. Default empty.
publishedAfter	ISO string	Keep publication timestamps strictly after an ISO date or timestamp. Default empty.
topicUrls	URL array	Google News topic pages for entities or other topics.
editions	String array	Editions as language-COUNTRY, for example en-US (list below).
resolveUrls	Boolean	Resolve search links to publisher URLs. Off by default; costs extra per result.
maxItemsPerQuery	Integer	Maximum items per query or topic and edition.
maxItems	Integer	Maximum items in the whole dataset.
dedupeBy	Choice	Deduplicate by url, googleNewsId, or none; IDs remain unique in all modes.
proxyConfiguration	Object	Apify Proxy settings for requests to Google News.

Example input for a recent keyword search and a section:

{
  "queries": ["climate change site:reuters.com"],
  "topics": ["SCIENCE"],
  "editions": ["en-US", "en-GB"],
  "timeRange": "7d",
  "resolveUrls": false,
  "maxItemsPerQuery": 50,
  "maxItems": 100
}

Supported editions are en-US, en-GB, en-IN, en-AU, de-DE, fr-FR, es-ES, it-IT, pt-BR, and ja-JP. Each edition is searched separately. The edition field in output uses Google News's country and language format, for example GB:en. If you set explicit dates, the Actor builds one search for each day and ignores timeRange. For large date ranges, the number of requests and run time increase with the number of days and editions. Google News may return fewer items than the cap, particularly for narrow searches.

maxItems is an exact upper bound: if at least that many distinct records exist in the selected sources, the dataset stops at that number. maxItemsPerQuery limits each input source in each edition before the total cap is reached. URL resolution adds one Google News article page request for each selected search result and groups up to ten resolution calls in one request. Topic rows already contain publisher URLs from their source page, so the setting does not change those URLs.

Use publishedAfter for scheduled monitoring: only strictly later publication timestamps are saved. Dates use midnight UTC; missing publication dates are excluded. Filtering happens before URL resolution. Repeated identical queries, sections, topic URLs and editions are fetched once. The Actor checks the remaining result budget before fetching another target and limits publisher URL resolution to rows that can still be charged. Retry responses are closed before the next attempt, freeing the HTTP connection. topicIds uses the same topic source as topicUrls. Neither option expands source coverage.

Output

The following record is from a local TECHNOLOGY section run. The particular headline and link are live data, so later runs will return different articles.

{
  "id": "CBMihgFBVV95cUxPRGYyaUdWZ2luQW5WZUZ4UkRyemNncUhiUHlNU2g0SjI3TDRENjdyNlczajY5QzVhVTNmalg1WTJMTHRmUTItRWpvaWVZeURZQVVvZ2MzcmZYVjJBZS1IMEtkVDRNZ3BaZUNNU0ltcFhqZFdiNWlvRmVDRUpmamdBaUdlYXo3Zw",
  "title": "Firefox just got a redesign with round tabs, new themes, and Compact Mode",
  "url": "https://www.theverge.com/tech/1001867/mozilla-firefox-157-redesign-compact-mode",
  "source": "theverge.com",
  "sourceUrl": "https://news.google.com/topics/CAAqKggKIiRDQkFTRlFvSUwyMHZNRGRqTVhZU0JXVnVMVlZUR2dKVlV5Z0FQAQ?hl=en-US&gl=US&ceid=US%3Aen",
  "publishedAt": "2026-09-29T13:00:00.000Z",
  "googleNewsId": "CBMihgFBVV95cUxPRGYyaUdWZ2luQW5WZUZ4UkRyemNncUhiUHlNU2g0SjI3TDRENjdyNlczajY5QzVhVTNmalg1WTJMTHRmUTItRWpvaWVZeURZQVVvZ2MzcmZYVjJBZS1IMEtkVDRNZ3BaZUNNU0ltcFhqZFdiNWlvRmVDRUpmamdBaUdlYXo3Zw",
  "googleNewsUrl": "https://news.google.com/articles/CBMihgFBVV95cUxPRGYyaUdWZ2luQW5WZUZ4UkRyemNncUhiUHlNU2g0SjI3TDRENjdyNlczajY5QzVhVTNmalg1WTJMTHRmUTItRWpvaWVZeURZQVVvZ2MzcmZYVjJBZS1IMEtkVDRNZ3BaZUNNU0ltcFhqZFdiNWlvRmVDRUpmamdBaUdlYXo3Zw?hl=en-US&gl=US&ceid=US%3Aen",
  "imageUrl": "https://platform.theverge.com/wp-content/uploads/sites/2/2026/09/firefox-project-nova-redesign-promo.webp?quality=90&strip=all&crop=0,0,84.3,100",
  "authors": [
    "Stevie Bonifield"
  ],
  "storyId": "channel_story_360:CAAqNggKIjBDQklTSGpvSmMzUnZjbmt0TXpZd1NoRUtEd2lJeTZ1SEVoSGNVaXI2UG1tN2VDZ0FQAQ",
  "relatedArticles": [],
  "publishedAtPrecision": "datetime",
  "query": null,
  "topic": "TECHNOLOGY",
  "inputType": "topic",
  "edition": "US:en",
  "position": 1,
  "scrapedAt": "2026-09-29T21:52:31.463Z"
}

How much does it cost?

The Actor is charged per result: each article row pushed to the dataset is one result. Rows where resolveUrls successfully found the publisher URL are billed at a higher per-result rate than other rows; it is never billed twice for the same row. Topic and section rows already contain publisher URLs, so they are billed as ordinary results. If resolution fails, the row keeps googleNewsUrl and url is null. The Pricing tab shows the current rates; refer to it before running a large collection. This README does not fix a price because pricing can change. A small maxItems value is the simplest way to estimate cost for your own searches.

Publisher URL resolution adds requests and typically makes search runs longer. A plain RSS search can return many headlines in one request, while URL resolution fetches a Google News article page for each selected result and sends small batches to Google News for the final links. Topic pages often provide publisher links in their embedded data and do not need that extra step. Wide date windows and multiple editions also raise request counts. Use the smallest input that answers your question, then expand it once you have inspected the dataset.

Use with AI agents (MCP)

An AI agent can discover and call this Actor through Apify MCP. Configure the MCP server URL with the Actor's account and name:

https://mcp.apify.com/?tools=axiomworks/google-news-scraper

Your agent may need its normal Apify authorization to run Actors. The input and output schemas tell the agent what arguments to send and which fields it can expect. The Actor itself does not require a Google News login.

Example prompts for an agent:

• “Find up to 20 Google News articles about battery recycling in the US edition and return the titles, sources, dates, and publisher URLs.”
• “Collect the latest TECHNOLOGY section headlines for the UK and Germany and group them by publisher.”
• “Search Google News for site:example.com robotics between two dates and give me the Google News links for each distinct result.”

Agents should treat returned headlines and linked pages as source data. They may summarize or compare the metadata, but the Actor does not retrieve full publisher article text. Use sourceUrl to identify the Google News response behind each row. If a publisher URL is absent, an agent can still cite the Google News link and should not infer a destination.

Integrations and API

You can run the Actor on a schedule, export its dataset to supported Apify integrations, or call it from your own application. The examples below show the same small input. Provide your Apify token through your normal secret management. Do not put tokens in public code or exported datasets.

Python with the Apify client:

from apify_client import ApifyClient
import os

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("axiomworks/google-news-scraper").call(
    run_input={"queries": ["battery recycling"], "maxItems": 20}
)
for article in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(article["title"], article.get("url"))

JavaScript with the Apify client:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('axiomworks/google-news-scraper').call({
  queries: ['battery recycling'], maxItems: 20,
});
for await (const article of client.dataset(run.defaultDatasetId).iterateItems()) {
  process.stdout.write(`${article.title} ${article.url ?? article.googleNewsUrl}\n`);
}

cURL through the Apify API:

curl -X POST \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"queries":["battery recycling"],"maxItems":20}' \
  'https://api.apify.com/v2/acts/axiomworks~google-news-scraper/run-sync-get-dataset-items'

The synchronous cURL endpoint is useful for short jobs. Larger date ranges and URL-resolved searches may take longer; for those, start a run and fetch its dataset when it finishes. A scheduled run can refresh the same search repeatedly. Google News ordering and publisher links may change between runs, so use the stable article ID when joining datasets over time.

FAQ

Why did I get fewer items than maxItems? Google News returned fewer distinct articles for your inputs, or some responses repeated the same articles. The cap limits output; it does not make Google News supply a minimum.

Why is url empty? A search record needs a successful Google News URL resolution. If it fails, use googleNewsUrl. Topic records usually have publisher URLs already.

Does the Actor download article bodies? No. It collects headline metadata and links from Google News. It does not request publisher hosts, paywalled pages, or article text.

Can I use a topic URL? Yes, if it is an HTTPS URL under news.google.com/topics/. The Actor applies each selected edition to the topic path.

How do dates work? Provide both dateFrom and dateTo in YYYY-MM-DD form. The Actor searches daily windows from the first date through the last. Google News determines which results match those operators.

Does the Actor need a Google login? No. It uses publicly accessible Google News responses. Access can still vary by region, network, and Google News availability.

Is it legal to scrape Google News?

The Actor gathers public headline metadata and links. Google News and publisher content may be protected by terms, copyright, database rights, or local law. Check the rules that apply to your use and location, especially if you plan to republish results or run at scale. This Actor is not affiliated with or endorsed by Google News, Google, or any publisher. It does not provide legal advice. Its output does not grant rights to the underlying articles. Use modest limits and respect publishers when following links outside this Actor.

Feedback

If a field disappears or publisher URL resolution stops working, report the affected input, edition, and a Google News link in the Actor's Issues tab. Google News can change its RSS format, topic data, and link resolution protocol. A short reproducible example helps identify whether the source changed or a specific query has no results. For questions about run settings, include the input JSON without credentials. Feedback on missing sections, ambiguous dates, or output fields is also welcome through the same tab.