About 450,000 articles a day from 9,000 European publishers in 24 languages, with entities, IPTC topics, sentiment and duplicate clustering, by API, webhook or RAG endpoint.
Data product · Listing updated 11 Nov 2026
Europe
Median 4 minutes from publication to API
About 450,000 articles a day from 9,000 publishers
Startups, Small and medium businesses, Mid-market, Enterprises
Publisher RSS feeds and sitemaps, plus full text licensed from partner publishers
Feed polling every two minutes, language detection, entity extraction and near-duplicate clustering
A real-time feed of European news coverage, enriched so you can filter by company, topic and language without building your own NLP pipeline.
Vistula polls publisher RSS feeds and sitemaps every two minutes and fetches only what publishers make publicly available or license to it. Language detection, entity extraction and topic classification run within seconds of ingestion. The licence covers internal analysis, display of headlines and snippets with a link to the source, and retrieval grounding with attribution. It does not include training models on full article text; publishers that allow that are listed separately on request. Coverage is strongest in Poland, Germany, France, Italy, Spain and the Nordics, and includes all EU member states plus the UK, Norway, Switzerland and Ukraine.
12 fields as documented by Vistula Web Data.
| Field | Type | Description | Example |
|---|---|---|---|
| article_id | string | Vistula article identifier | — |
| url | string | Canonical article URL | — |
| publisher | string | Publisher name | — |
| published_at_utc | datetime | Publication time | — |
| language | string | ISO 639-1 language code | — |
| headline | string | Article headline | — |
| snippet | string | Licensed snippet, up to 300 characters | — |
| full_text_available | boolean | True when the publisher licenses full text | — |
| entities | json | Organisations, places and public-role people with IDs | — |
| topics | json | IPTC media topic codes | — |
| sentiment_score | number | From -1 (negative) to 1 (positive) | — |
| duplicate_cluster_id | string | Groups syndicated and near-identical articles | — |
Real rows from the provider's sample file. Personal data is masked before anything is shown.
| published_at_utcstring | publisherstring | languagestring | headlinestring | topicsstring | sentiment_scorestring |
|---|---|---|---|---|---|
| 2026-11-03T07:42Z | Demo Gazette | pl | Example Rail opens new freight terminal near Łódź | economy/transport | 0.41 |
| 2026-11-03T07:45Z | Example Daily | de | Sample Chemie cuts 2026 outlook on energy costs | economy/company results | -0.52 |
| 2026-11-03T07:51Z | Demo Tribune | fr | Demo Foods launches recycled packaging line | environment/recycling | 0.28 |
Prices in EUR, as published by Vistula Web Data
headlines, snippets and metadata
per 1,000 API calls above plan
including archive since 2019
Converted amounts use DataStoa display rates and are approximate. Vistula Web Data quotes the final price, taxes and terms.
Request detailed pricingEurope.
Time coverage: 2019 to present.
Provider-reported
Latency: Median 4 minutes from publication to API
DataStoa has not audited these figures. Ask the provider how they are measured, or test them on a sample.
The licence is agreed directly between you and Vistula Web Data. This summary is provider-reported; the linked terms prevail.
Not listed: AI / ML model training, redistribution allowed, marketing and outreach, research use. Ask Vistula Web Data if you need any of these.
Public web data extraction, news, domains and company websites
Vistula Web Data builds and runs crawlers for public web sources and packages the results as clean datasets and APIs. Its engineering teams in Kraków and Gdańsk maintain collection pipelines for news, domain records and business websites across Europe.
Also available on: AWS Data Exchange, Snowflake Marketplace
What buyers say about Vistula Web Data. Reviews cover the provider, not only this listing.
from 2 reviews
Coverage of European business news is broad and the archive helped with our retrieval tests. Entity linking confused companies with similar names more often than we liked, so we added our own disambiguation step.
Laura V., AI engineer, Legal tech start-up
They run 30 sources for us and fix layout changes before our deliveries are affected. When we asked about a site behind a login they declined and suggested a licensed alternative, which we appreciated in hindsight.
Tomasz K., Head of data, Price comparison platform