News Datasets.
Made Easy.

Datanews helps you to easily collect news datasets of any size for any industry or use case you have. Millions of articles with full content and metadata delivered directly to you.

Get Free Price Estimate

Everything you need to collect
complete news dataset.

Datanews has all necessary features needed to collect a dataset that will include full coverage on the researched topic.


Historical Data

We have an archive of historical data for the last 5 years, with ability to extract full site history for some news sites.


Keywords and Sources

We can collect data by keywords, sources, or both. You can send a list of tickers, company names, or any other keywords. We will identify and extract full coverage about them.


Full Content

Each article will include all important information, like url, title, full content, etc.



Our scale allows to extract millions, tens of millions or even more news articles. You may be sure that we have enough data to fit your ML model.

Use cases

Datasets for all use cases
with one simple service.

We can help you collect a news dataset that will cover all your use cases. Take advantage of our infrastructure and focus on delivering value, not on dataset collection.

Get Free Price Estimate

Finance news.

Collect relevant financial news from key news outlets.

Company news.

Collect company news coverage by company name, tickers, or other company associated keywords.

Historical news.

Up to 5 years of historical news coverage for any keyword. Even more for specific sources (e.g. extract all articles from techcrunch).

News content for Machine Learning.

Full article content is provided for majority of publications. Perfect input data for training machine learning models.

Any keywords, any sources, any data.

Search news by any keywords, any news sources, or a combination of both.

Article example.

Example of a single article. Dataset can include millions of these.

News article
    "url": "https://www.businessinsider.com/facebook-amazon-google-apple-ceos-testify-antitrust-hearings-2020-7",
    "source": "businessinsider.com",
    "authors": [
      "Isobel Asher Hamilton"
    "title": "Tim Cook, Mark Zuckerberg, Jeff Bezos, and Sundar Pichai will all testify before Congress in an antitrust hearing",
    "pubDate": "2020-07-02T10:25:16+00:00",
    "country": "us",
    "language": "en",
    "description": "Facebook, Amazon, and Google had all reportedly been confirmed for the antitrust hearing by mid-June, but Apple had stayed quiet.",
    "imageUrl": "https://i.insider.com/5cf657ec11e20538196144f4?width=1200&format=jpeg",
    "content": "..."

Fair, simple pricing for all.

All types of businesses need access to high-quality data, so we give you the option to decide how much you need to use.

$ 0.001 / article

* for datasets less than 250k articles
* 50$ minimum charge

Historic data

Finance / Company / Industry news

Multi-language support

Full content

Extraction by Keywords & Sources

Suitable for research / ML training

Get Free Price Estimate


We offer variable pricing with discounts for larger organizations. Get in touch with us and we’ll figure out something that works for everyone.

Contact us

Do I have access to article's content?

Yes, full content is included for articles where available. Note: content for paywalled sites is not included.


What languages do you support?

Our API supports 10+ languages, including, but not limited to english, german, french, italian, spanish and russian.


How the work process looks like?

You send us a list of keywords / sources, we send an estimated cost for the whole dataset. If you agree, then we extract the data, and upload it onto Google Drive.


How to contact you?

You can contact us by sending an email to support@datanews.io. We will promptly respond to you as soon as possible.


Do you have historical data?

Yes, we have an archive with 5 years of historical data, and full history for some specific sites.


Can I extract all articles from selected news site?

It depends. Please contact us, and we can see whether this is possible.

Get started

Start with Datanews at no cost.

Datanews News Collection is the best and easiest way to collect news datasets for your ML models. So, what are you waiting for?

Get dataset price estimate for free