2026 IEEE World Congress on SERVICES
(SERVICES 2026)

Analytical data has become an important asset of most organizations. One driving factor for this is the increased application of machine learning and AI. This will further gain importance with the rise of agentic AI, where agents depend on the availability of well discoverable and processable data. As a foundation for this, the notion of data products has gained popularity. The concept of a data product is two-fold. On the one hand, it is a data asset that is managed like a product to be sold to consumers. On the other hand, it is a self-contained unit of analytical data that is self-describing, ready for consumption, and designed for composition. Regarding the latter notion, a data product shares many similarities with that of a microservice. It is the unit of decomposition of a larger system and it is an independently deployable unit of composition to create higher-order services from. In fact, a data product is often implemented as a microservice that wraps a data asset (which can range from a simple spreadsheet to a massive large-language model) and shares its data via an API.

For microservices, large companies often rely on a sophisticated infrastructure platform as well as service discovery mechanisms. These form a company-internal service ecosystem to build new services from. Similarly, companies that decentralize their data landscape rely on data infrastructure services. This way, companies can quickly compose new data-driven use cases and AI applications from existing data products. However, in contrast to microservices, which are mostly used for intra-organizational service composition, companies are often interested in sharing data products also across organizational boundaries. For creating such a cross-organizational data product ecosystem (DPE) that allows for seamless and secure sharing of data products between organizations, several additional challenges need to be addressed, e.g., how to deal with data privacy or data licensing.

The services computing community has been developing standards and technologies for the integration of software, applications, and collaboration of services and processes by overcoming system heterogeneity. Services computing technologies can play a significant role in data product ecosystems, in which novel tools, best practices, solutions, and platforms will allow: (1) for organisations to transform their data into data products with governance, and privacy protection, to extend and broaden data applications, and to enable data exchange and flow crossing organizations, domains and sectors; (2) for data consumers to search, query, explore, and combine data products for advanced data analytics and providing value-added services.

This symposium will feature invited leaders in the field to articulate the idea of data product ecosystems (DPE) and tackle the challenges that result from the complexity of DPE. The topics of the symposium include, but are not restricted to:

• Data product design, description, and packaging

• Methods and tools for data source to data product transformation

• Mechanism and tools for data product discovery and composition

• Services for data management and exchange in DPE

• Review of Data Mesh, Data Fabric and other relevant technologies for DPE

• Data transparency, provenance, and traceability in DPE

• Metadata management, semantic modelling, and enrichment in DPE

• Data product catalog generation and governance of DPE

• Data quality management and data curation in DPE

• Data contracts and data product testing

• Privacy models for data products in DPE

• Privacy and security perseverance for DPE

• Data sharing policies for DPE

• Ethical considerations and regulatory frameworks for DPE

• Data royalty collection and licensing for data products

• Review of Data Spaces, GAIA-X and other standardisation efforts for DPE

• Applications of AI and self-optimization in DPE