Data sharing is the pooling of data between different entities, to create value that could not be achieved by remaining compartmentalized. In this article, discover the importance of Data Sharing and why, now more than ever, it's time to talk about it again.
Contents
Why is it appropriate to talk about Data Sharing now?
A changing regulatory environment
The European Union has clearly highlighted data as a strategic lever. After the RGPD (on the personal data protection side), it is preparing for the application of new rules, such as the Data Act (expected to come into force in September 2025).
The aim: to provide businesses with a secure, harmonized framework, while promoting European digital sovereignty. As the Commission explains in its European Data Strategythe aim is to enable the free circulation of data within the EU, while respecting fundamental rights and competition.
The rise of generative AI and technological sovereignty in focus
The arrival of generative AI (GenAI) is reshuffling the deck at an unprecedented speed, forcing companies to rethink the way they manage and exploit data. As a recent report (AI générative: s'unir ou subir), data sharing is a major strategic lever. "We are all collective creators of data, which needs to be extracted, refined and enriched like any other raw material".
However, to feed large-scale language models and their advanced uses (agentification, automation, etc.) in a relevant way, the volume, diversity and quality of available data are decisive. In other words, generative AI can only deliver its full potential with access to vast deposits of data, ideally pooled and structured.
In this context, the European Union is putting forward the idea of "unite or suffer". On the one hand, if economic players forgo cooperation and simply adopt the solutions of a few non-European giants, they run the risk of suffering increased technological dependency (cloud, AI models, etc.) and seeing the value of their data captured elsewhere. On the other hand, the gamble of joining forces, by facilitating the pooling of information (data spaces, sector portals, trusted artificial intelligence platforms), would enable Europe to equip its sectors with generative AI adapted to their realities, while respecting privacy protection, data security and technological sovereignty.
A colossal business opportunity
According to a study on data sharing ecosystems (Data sharing masters: How smart organizations use data ecosystems to gain an unbeatable competitive edge), organizations taking advantage of data sharing ecosystems could improve their annual revenues by 2.4% in a pessimistic scenario, and up to 9% in an optimistic scenario, thanks to new revenue sources, cost reductions and improved productivity.
What is Data Sharing?
Data sharing is the pooling of data between different entities, to create value that would otherwise be unattainable on a stand-alone basis. There are several different types of data sharing.
Data sharing between companies (B2B)
We're talking here about data from the private sector exchanged between economic players, for example industrial data produced by IoT (Internet of Things) devices, or the provision of commercial data to create new services.
Despite strong economic potential, B2B sharing remains limited by a lack of trust, fear of losing a competitive edge and legal uncertainties (who can do what with this data?).
There is therefore a real need to establish a clearer framework, particularly for "co-produced" data (produced simultaneously by several parties, as in the supply chain).
Data sharing from the public to the private sector (G2B)
The EU is actively promoting the "liberation" of a wide range of public data (open data) - weather data, geographical information, statistical databases, etc. - in order to stimulate innovation and the creation of new services.- to stimulate innovation and the creation of new services. The Open Data Directive urges administrations to make more high-value data sets available in usable formats (e.g. via APIs).
An example of open access to public data is the platform data.economie.gouv.fr. Managed by Bercy, this platform is being modernized with a number of objectives in mind:
Making data easier to understand
- An editorial animation explains the content of the datasets (explanatory sheets, contextualized articles).
- Data visualizations (interactive graphs, dashboards) are made available to users to give immediate meaning to information, in addition to simple raw files.
- A catalog of data visualizations lists different productions and examples of use, to encourage reuse by the community.
Federating an open data community
- Users (journalists, researchers, citizens, start-ups, etc.) are invited to exchange with producers (Bercy departments and services) via discussion channels and feedback.
- The feedback from these exchanges is first processed automatically (LLP, or natural language), in order to categorize the discussions, monitor usage indicators, and adapt the data offering as effectively as possible.
Improve dataset exploration and discovery
- The portal offers more filters to refine searches (data type, theme, format, etc.).
- Each dataset is accompanied by richer metadata: update date, person responsible, license, functional description, etc., enabling easier, more transparent access to information.
Business-to-business data sharing (B2G)
Public authorities increasingly need data from the private sector to improve their policies and services to the population: mobility planning, emergency response (floods, fires), more reliable statistics. Anonymized data from large platforms or mobile applications can, for example, inform urban traffic management or complement epidemiological monitoring.
A group of experts commissioned by the European Commission has stressed the need to create dedicated national structures to encourage this B2G sharing, while ensuring that corporate data is protected and used only for purposes of general interest.
Data sharing between public authorities
While we often talk about data exchange between administrations and businesses, data sharing between public authorities is just as crucial. Beyond the "Tell us once" principle, which avoids users having to enter the same information several times with different administrations, a new stage is taking shape: proactive administration.
What is proactive administration? Rather than waiting for a citizen to take the necessary steps to obtain a right or benefit, administrations cross-reference their internal databases to identify a priori potentially eligible users. They can thus :
- Notify users of their eligibility for grants, social assistance, etc., and simplify the process of obtaining them.
- Automate the allocation of benefits as far as possible, when no additional formalities are required.
Example: automatic allocation of certain national education grants to students recognized as eligible, without them having to complete a file.
What are the main trends and use cases for Data Sharing?
The rise of sector-specific data ecosystems
Initiatives are developing in many strategic sectors. For example:
- Healthcare: sharing healthcare data is essential for accelerating innovation in the medical sector. It enables health authorities to make better-informed decisions, helping to improve the accessibility, efficiency and economic viability of healthcare systems.
- Mobility: data sharing in the mobility sector offers the opportunity to develop innovative services, based in particular on information generated by connected vehicles. It makes journeys smoother and more efficient, reduces costs and emissions, and enhances road safety through predictive maintenance.
- Finance: data sharing in the financial sector accelerates innovation by facilitating the emergence of new financial services, notably through open banking: account aggregation, third-party payment initiation, fast loans and financing, etc.
Time-to-market-oriented uses and innovative products
Sharing upstream data speeds up prototyping and time-to-market for new products and services. Examples:
- Research & Development: several manufacturers can co-finance and co-exploit R&D data, each retaining ownership of the confidential part but sharing the exploitable part for all.
- Digital twins: we feed a virtual model (machine, city, building) with data from multiple sensors, in order to run simulations (energy consumption, probable breakdowns, etc.).
Augmented" data market places
Some Data Sharing success stories
Data sharing in maritime logistics
Shipping companies have set up a data ecosystem to record all incidents, accidents and near-misses on their ships (HiLo initiative).
By pooling this information, they feed a predictive algorithm to anticipate risks and improve safety for the entire fleet. Results:
- 72% reduction in rescue channel accidents
- 65% reduction in machine room fires
- 25% reduction in fuel spills
Data sharing between distributors and suppliers
In the 1990s, Walmart and Procter & Gamble entered into a Vendor Managed Inventory (VMI) agreement, based on the complete sharing of sales and inventory data. Results:
- Reduced breakage rate: P&G can more accurately predict when demand will increase (e.g. promotions, seasonality) and adjust production or deliveries.
- Lower logistics costs: fewer emergency deliveries, better transport planning, less overstocking and therefore lower storage costs.
The "augmented" data market place
In the past, a data market place was mainly a catalog of CSV tables or raw data feeds made available to customers or partners. Today, data market places offer new services: integrated data analytics and AI functionalities, the ability to share insights, reusable AI models, a logic of active collaboration by integrating a community component.
Read more with 5 key Data Sharing issues.
Opening up data assets with an AI agent: challenges and steps
While it may seem like a long way to go before you can deploy an AgentAI on your data assets, many companies already have sufficient data assets, but they are often poorly organized, poorly managed and poorly managed.
Setting up an ERP project: the keys to success
This article is the continuation of 4 mistakes in ERP / data repository implementation Every problem in the implementation of ERP and data repository projects has to be solved.
Data governance: definition and key principles
Data governance is a major lever for improving performance and accelerating business transformation. Today, data is a major corporate asset, with
