We’re excited to announce the revival of our French company data, including coverage of French territories! This marks a crucial step forward, considering the significance of French legal entities within the EU and the jurisdiction’s substantial size. Our database currently holds records for over 11 million companies, but these were last updated in June 2023.

In the coming months, we’ll be working to keep up with new incorporations and fill in the gaps as we go, adding any missing records and updating stale data. We expect to fill in the gaps by the end of December 2024 and ensure there are no stale records by the end of Q1 2025.

The challenge

This project has been a rewarding one, showcasing advancements in EU open data initiatives in France, particularly concerning legal entities (there are many exciting datasets in data.gouv.fr/ that promote transparency in the public sector). Previously, we relied on an outdated method that provided a single bulk file each month. While useful it posed several challenges:

Updates were infrequent, limited to monthly releases.
The file now contains over 27 million records, significantly increasing complexity when identifying changes or new incorporations, as this requires sifting through the entire dataset—a highly resource-intensive task.
Records associated with sole traders were excluded due to legal constraints in this case, though other limitations also add to the time and complexity of managing and processing the data effectively.

The solution

Today, thanks to the efforts of French institutions like Institut National de la Statistique et des Études Économiques (INSEE) and Institut National de la Propriété Industrielle (INPI), we can access open data through APIs that update on a daily basis. These APIs allow us to work with much smaller, more manageable data loads, focusing specifically on what’s new or what has changed. This daily access enables us to maintain more accurate, up-to-date records, ensuring the data we provide is timely and relevant for everyone who relies on it. Daily updates helped manage the workload but we still faced several challenges:

The API provides a huge amount of structured data. We’ve streamlined this to align with our other datasets, ensuring compliance with data reuse licensing requirements and legal constraints.
Although the API is comprehensive, we’ve enriched and combined certain fields, at times with additional sources (such as industry codes), to provide more value than just raw codes would.
While we can limit searches to recently created or updated records, we’ve encountered instances where large-scale updates necessitated handling substantial data volumes efficiently. As a result, we have adapted our solution to manage these scenarios effectively.
As these platforms continue to evolve, we’ll adapt accordingly. Our processes are designed to handle frequent updates and improvements, ensuring our dataset remains aligned with any ongoing modifications.

We’re now fully equipped to handle even the most extensive updates in France’s vast data landscape.

What’s next?

As we continue this work, our goal is to ensure that our French company data remains comprehensive and current. The data fields we previously maintained will stay intact, but this update will allow us to offer much more—particularly a more complete set of officerships through the Data INPI service. By having hold of the Système Informatique pour le Répertoire des Entreprises et des Établissements (SIRENE database), we’ll have the foundation to further enrich our dataset through other sources, with promising potential to expand our information on establishments, trademarks, and any other public announcements.

As with any information technology solution these sources are undergoing continuous modernisation, and we’re committed to staying aligned with their progress, seamlessly adapting our services to incorporate any updates or enhancements on your behalf.

We anticipate enhancements to the Sirene API, as well as a new revision of NAF codes in 2026 , which will bring even greater utility to the data. For instance it is a bit outdated that currently the French NAF codes are based on business models from 2008 (with a few touch-ups in the 2015 revision). Take Netflix, for instance—not quite a television broadcast right?

We’re excited about what’s to come and look forward to providing ever more accurate, detailed, and timely data for all who depend on it—whether for regulatory, commercial, or analytical purposes.

For more information

Contact us