Event Registry logo
Media Journalism · Market Research

Event Registry

News intelligence platform from Slovenia that collects and enriches content from over 150,000 publishers worldwide, clusters articles into world events, and delivers them through three web products and a documented news API with a free tier.

Active GDPR compliant Free plan · Free trial Freemium API available Verified by Guidaio
Overview

What is Event Registry?

Event Registry is a news intelligence platform operated by Event Registry d.o.o., a company incorporated and based in Ljubljana, Slovenia. It continuously collects what the world's news and blog publishers put online, runs it through natural language processing, and hands it back as structured data rather than raw text. The company states coverage of more than 150,000 sources and over a million articles a day, with an archive reaching back to 2014.

The distinctive step is enrichment at ingestion. Every article arrives with the people, places and organisations it mentions recognised and disambiguated, with categories assigned, with a sentiment score, and with duplicates flagged. Articles describing the same real-world development are then clustered into a single event that spans languages and publishers, which is what turns a firehose into something an analyst can actually read.

Four web products sit on top of that pipeline. ER Monitor watches mentions of a brand, a person or an organisation and sends alerts and digests. ER Intelligence handles search, trend discovery, interactive visualisations, sentiment analysis and exportable reports. ER Compare puts brands, competitors and topics side by side and looks at how media and demographics perceive them. A separate ESG Tool detects sustainability signals at sentence level against the SASB and SDG frameworks and exports to JSON, CSV, TXT or XML.

The same engine is sold to developers under a second brand, NewsAPI.ai, which is the same company under the same terms: a REST API, official Python and Node.js SDKs, a code-generating sandbox, a separate text analytics API for annotating your own text, and an official MCP server that lets AI assistants query live news from a prompt.

The site advertises 800+ client companies and institutions, a 99.99% uptime SLA, SOC 2 compliance and GDPR readiness. Those last two are badges rather than documented certifications, and the logos shown on the homepage are claimed customers, not facts about the vendor. Note also that the site quotes 50+ languages on its API pages and 60+ on its homepage.

What it does

  • Search articles and world events across more than 150,000 global news and blog sources
  • Filter by keyword, concept, entity, source, category, location, sentiment, author and source rank
  • Read articles already enriched with entities, categories, sentiment and duplicate detection
  • Group coverage of the same story into a single multilingual event instead of a flat article list
  • Query the archive back to 2014 for historical research and dataset building
  • Receive real-time email alerts and digests when relevant news breaks
  • Plug live news into an AI assistant through the official MCP server
Audience

When to use Event Registry / When not to

A quick filter to help you decide if Event Registry is the right fit.

When to use Event Registry

  • Developers and data teams who want news delivered already structured, through a REST API, a Python or Node.js SDK, or an MCP server
  • Risk, geopolitical and supply-chain analysts who need to see a story before it reaches mainstream coverage
  • Investment and market analysts looking for event-driven signals and sentiment across global media
  • Communications and brand teams tracking mentions, perception and competitors in real time
  • Academic researchers, who can register with an institutional address and request a discount for strictly academic use

When not to use Event Registry

  • Anyone wanting to build a commercial product on the free tier: free access is explicitly limited to evaluation and testing
  • Users without a company or academic email address, since personal and disposable providers are refused at signup
  • Teams planning to resell, republish or redistribute the data, or to build a competing media intelligence service
  • Buyers who need a published postal address, a named data protection officer or a signed data processing agreement before contracting
  • Organisations that need the web platform at a known price, since only the API rates are public and everything else is quoted per company
  • Mobile-first users, as there is no iOS or Android application and no browser extension
Get started

How to use Event Registry

A typical end-to-end flow, from setup to results.

  1. Register at eventregistry.org/register using a company or academic email address, as personal and disposable providers are refused
  2. Confirm your account through the mandatory activation email
  3. Collect the API key issued with the account, which comes with 2,000 free tokens and requires no credit card
  4. Choose your entry point: the web interface for ER Monitor, ER Intelligence and ER Compare, or the API for programmatic access
  5. Build a first query in the sandbox, inspect the response, and copy the generated code straight into your application
  6. Install the Python or Node.js SDK from the EventRegistry GitHub organisation, or call the REST endpoints directly
  7. Optionally install the official MCP server with npx and set your key in the NEWSAPI_KEY variable to query news from an AI assistant
  8. Set up topics, filters and email alerts for the subjects you need to follow
  9. Watch token consumption: recent article searches cost one token, archive searches cost more per year queried, and unused tokens do not carry over
  10. Subscribe to a paid plan when the free tokens run out, or contact sales through Calendly for platform and enterprise pricing
Quick read

Pros & Cons

Pros

  • Data arrives already structured with entities, categories, sentiment and event grouping, which removes a whole post-processing layer
  • Unusually deep archive for the price, reaching back to 2014
  • API pricing is public and itemised down to the token cost of each call, which is rare in this market
  • Permanent free tier of 2,000 tokens with no credit card required
  • Strong developer surface: official Python and Node.js SDKs on GitHub, a code-generating sandbox, and an official MCP server for AI assistants
  • EU-established vendor under Slovenian law, with the public site resolving to a Slovenian network
  • A stated discount for strictly academic use, requested from an academic email address

Cons

  • No postal address anywhere on the site: there is no legal notice page, no contact page and no address in the footer
  • A single published email address handles legal, GDPR, billing and academic requests at once, and no support address is given
  • The footer link to Clients Support no longer resolves, and the Contact link opens a Calendly booking form rather than a page
  • Web platform pricing is quote-only, so only the API rates can be compared before talking to sales
  • The site contradicts itself on coverage, quoting 50+ languages on the API pages against 60+ on the homepage, and over 30,000 publishers in the homepage social metadata against 150,000+ in its body
  • SOC 2 and GDPR readiness are displayed as badges with no certificate number, no report and no trust centre, and the privacy policy still cites the repealed 1995 directive
  • No published data processing agreement, no named data protection officer and no statement of where customer data is hosted
Pricing

Pricing & Plans

There is a permanent free plan: creating an account yields a free API key with 2,000 searches, no credit card required, limited to content from the last 30 days and to non-commercial evaluation. The cheapest public paid tier is the 5K Plan at 90.00 USD per month, which adds the historical archive from 2014 and custom sources on demand. All prices are stated in US dollars, subscriptions are billed monthly and renew automatically, fees are non-refundable, and unused tokens do not carry over to the following month. Access to the web platform — ER Monitor, ER Intelligence and ER Compare — is priced individually per company and is only available on request.

Plan 1
  • Free — 2
  • 000 tokens
  • no credit card
  • last 30 days of content only
  • evaluation and testing use only
Plan 3
  • Higher API tiers
  • starting with a 10K plan — seven paid offers in total
  • with amounts shown through an interactive slider rather than in the page
Plan 4
  • Support levels attached to the tiers — Limited
  • Basic
  • Enhanced
  • Advanced and Premium
  • the last adding a personal account manager
Plan 5
  • Custom API plan — specific data sources
  • delivery formats and volume pricing
  • arranged with the sales team
Plan 6
  • Web platform (ER Monitor
  • ER Intelligence
  • ER Compare) — priced individually per company
  • quote only
Special offers — Academic discount: users working with the data exclusively for academic purposes can request a discount by writing from an academic email address
Prices and plans listed above may evolve. Always check the official pricing page before subscribing.
Trust & Privacy

Data, GDPR & hosting

A consolidated view of how Event Registry handles your data.

GDPR overview

Concrete signals exist but the record is uneven. Event Registry d.o.o. is established in Slovenia, inside the EU, so no Article 27 representative is required and none is named. The News API page shows a GDPR-ready badge alongside its uptime and SOC 2 claims. The privacy policy grants the usual rights — access, correction, erasure, objection — all exercised by writing to a single address, and it devotes a section to the right to be forgotten. Against that: the policy, last updated in October 2025, still anchors itself in the repealed Directive 95/46/EC and the Slovenian Personal Data Protection Act, and never cites Regulation 2016/679. No data protection officer is named, no dedicated privacy mailbox exists, and no data processing agreement is published or offered.

Who owns the data?

Event Registry keeps full ownership of everything it produces: the datasets it assembles, the metadata its models generate — extracted entities, categories, sentiment, event clusters — and any derivative of them, along with the platform, the APIs and the interface. A paying subscriber receives only a limited, non-exclusive, non-transferable and revocable licence to access and use that data for commercial purposes. Ownership never transfers. The underlying articles remain the property of the news publishers who wrote them, and Event Registry states plainly that it neither grants nor brokers publishing rights over that third-party content, leaving copyright compliance entirely to the user.

Reuse rights

Reuse is licensed, not free. A paid subscription lets you query the service and use the results commercially, and you may export data and visualisations, but the licence stops there. You may not share, sell, sublicense or redistribute the data, nor republish or embed article content without obtaining rights from the original publisher, nor build or market a competing media intelligence, news aggregation or event detection service. Circumventing rate limits or creating several accounts to stay inside the free tier is treated as misuse. On the free tier, use for any project, research or application producing commercial or institutional value is forbidden, even indirectly. On the personal data side, Event Registry says it never rents or sells personal information, and that query history is used for internal company purposes only.

Data retention & training

Retention summary
Usage data — the queries you run, their frequency and the IP address they come from — is kept until you withdraw your consent, or for nine months after you register a free account without upgrading to a paid one, whichever comes first. Everything else is kept for as long as the purpose it was collected for requires, and longer where the law demands it. Deletion is not self-service: you email the company's general address, and requests are handled case by case, with a dedicated right-to-be-forgotten process aimed mainly at republished articles containing inaccurate or incomplete personal data. No specific duration is published for logs, backups or billing records, and no anonymisation timetable is described.
Subprocessors disclosed
Yes
GDPR contact

Hosting summary

The site never states where customer data is hosted. No country, no region, no cloud provider and no data centre is named, and there is no trust centre or security page to consult. The only geographic statement in the privacy policy runs the other way: it warns that information may be transferred outside the European Union, giving the payment provider's and the newsletter provider's servers as examples, and promises only unspecified appropriate security measures without naming standard contractual clauses or an adequacy decision. Third parties handling personal data are named in the body of the policy — Google for advertising, Microsoft Clarity for session analytics with anonymised IP addresses and country-level location only, Drift and Chatwoot for chat, and Mailchimp for newsletters — but no formal subprocessor register exists. Jurisdiction is clearer than hosting: the vendor is established in Ljubljana, Slovenia, the terms are governed by Slovenian law, and disputes fall to Slovenian courts. As a network observation rather than a vendor statement, the public website itself resolves to an IP address hosted in Slovenia.

Watch-outs

Things to keep in mind

Risks and trade-offs to weigh before adopting Event Registry.

  • The site's calls to action say free trial, but the terms describe an evaluation allowance: 2,000 non-renewing tokens with commercial use forbidden, even indirectly, so a prototype that turns into a product breaches the terms
  • Compliance claims are decorative rather than documented: SOC 2 and GDPR-ready are badges with no certificate, no report and no trust centre, and the privacy policy still cites a directive repealed in 2018
  • Coverage figures should not be quoted verbatim in your own reporting, since the site itself states both 50+ and 60+ languages and both 30,000 and 150,000+ publishers
  • Sentiment scores, categories and event clusters are model output, not fact; treating an automatic sentiment score as ground truth in a risk or investment decision transfers the model's errors into yours
  • The vendor states plainly that it does not verify the content it delivers, so responsibility for checking the reliability of any article stays with you
  • Copyright exposure sits with the user: the licence covers querying the data, not republishing publisher content, and Event Registry explicitly brokers no publishing rights
  • Reaching a human is harder than it looks: one shared mailbox, a booking form instead of a contact page, and a Clients Support link that no longer resolves
Setup

Setup & Integrations

Technical difficulty

Low for developers, moderate for everyone else. The web products need only an account and a browser. The API is equally quick: the key is issued at signup, the sandbox builds a query and returns deployable code, official Python and Node.js SDKs sit on GitHub, and the MCP server installs in one npx command. The friction is administrative rather than technical — signup demands a company or academic email address plus a confirmation step — and token consumption must be planned before running archive queries. There are no no-code connectors and no publicly documented webhooks or ready-made CRM integrations.

Deployment

Web appAPIPlugin

Integrations

Claude Desktop Claude Code Cursor Windsurf Gemini CLI VS Code Python Node.js
Company

Behind Event Registry

Company name
Event Registry d.o.o.
Founded
INFORMATION_NOT_FOUND
Country of origin
🇸🇮 Slovenia
UBO
Gregor Leban
UBO country
🇸🇮 Slovenia
Domain registrar country
INFORMATION_NOT_FOUND
Legal contact

Social

Official links

Resources

All the official URLs gathered for verification and reference.

Compare

Alternatives

Tools that compete with or complement Event Registry.

M Meltwater
FAQ

Frequently asked questions

Is there a free plan, and what does it include?
Yes. Creating an account gives you a free API key with 2,000 tokens and no credit card. It covers only content published in the last 30 days, and the terms restrict it to evaluation and testing: any project, research or application producing commercial or institutional value is excluded, even indirectly.
How much does the cheapest paid plan cost?
The 5K Plan is 90.00 USD per month. It opens up the historical archive from 2014 and allows custom sources on demand. All prices are quoted in US dollars and billed monthly.
Is the web platform priced the same way as the API?
No. Only the News API rates are public. ER Monitor, ER Intelligence and ER Compare are priced individually per company, and you have to book a call through the vendor's Calendly link to get a quote.
How are tokens consumed?
A search over recent articles costs one token, while an archive search costs five tokens per year searched. Event searches cost five tokens, or twenty per archived year. Text analytics is billed per document, from 0.01 token for language detection to 0.05 for semantic similarity or article extraction. Unused tokens are not carried over to the next month.
Is there an API, and is the documentation public?
Yes. The documentation is openly accessible, with a sandbox that generates ready-to-deploy code, official Python and Node.js SDKs on GitHub, and an official MCP server that connects the service to AI assistants such as Claude Desktop, Cursor, Windsurf and VS Code.
How much news does the service actually cover?
The vendor claims more than 150,000 publishers, over a million articles a day and an archive back to 2014. Be aware that the figures are not consistent across the site: the API pages say 50+ languages while the homepage says 60+, and the homepage social metadata still mentions over 30,000 publishers.
Who operates the service, and what about NewsAPI.ai?
The service is operated by Event Registry d.o.o., a company incorporated and based in Ljubljana, Slovenia. NewsAPI.ai is the same company under a second brand aimed at developers: it shares the same terms, the same privacy policy and the same pricing page.
How can the company be contacted?
Only one email address is published, info@eventregistry.org, and it is used for legal notices, data protection requests, invoicing and academic discount enquiries alike. Sales contact goes through a Calendly booking form. There is no postal address, no contact page and no dedicated support address on the site.
Can the data be republished or resold?
No. Sharing, selling, sublicensing or redistributing the data is prohibited, as is republishing or embedding article content without rights from the original publisher. Building or marketing a competing media intelligence, news aggregation or event detection service is also excluded.
How long is personal data kept?
Usage data is kept until you withdraw consent, or for nine months after registering a free account that is never upgraded to a paid one. Everything else is kept as long as necessary for the purpose it was collected for. Deletion is requested by email and handled case by case.
Conclusion

Should you pick Event Registry?

Event Registry is a genuinely technical product in a market that often sells dashboards. Its value is not that it collects news, but that it hands the news back already parsed: entities resolved, categories assigned, sentiment scored, and coverage of the same development folded into one multilingual event. For a data team that would otherwise build that layer itself, and for analysts working on risk, geopolitics, investment, ESG or supply chains, that is a real saving.

The commercial posture is refreshingly plain for this sector. The API price list is public, the token cost of every call is itemised, and the entry point is a permanent free tier of 2,000 tokens with no credit card. At 90 USD a month, the cheapest paid tier opens an archive going back to 2014, which few competitors match at that level. The developer surface — public documentation, a code-generating sandbox, official SDKs and an MCP server — is well ahead of the norm.

The weakness is the company behind it. The site publishes no postal address, no contact page and no data protection officer, and routes legal, privacy, billing and academic questions to the same single mailbox. SOC 2 and GDPR readiness appear as badges rather than as documented, verifiable claims, and the privacy policy still anchors itself in a directive repealed in 2018. The site also contradicts itself on how many languages and publishers it covers.

The verdict follows that split. If you are buying data to plug into your own systems, this is a strong and unusually transparent option on price. If your procurement needs an auditable vendor with a signed processing agreement and a documented compliance posture, expect to ask for a lot that the website does not currently provide.