The Data Product Marketplace: turning potential value into tangible outcomes

This deck goes with our Big Data London 2025 talk, co-presented with Roberto Grandi of Eni.

The core message: federated strategies like data mesh don’t work if they stop at “data as a product.” Domain autonomy alone isn’t enough. Without reuse and collaboration across domains, duplication explodes, costs soar, and efficiency vanishes.

The real challenge? Matching supply with demand. Organizations need a marketplace where valuable data products can be discovered, understood, accessed, and combined.

We’ll show, with Eni’s real-world example, how to make that marketplace a reality.

Transcript

1. The Data Product Marketplace Turning potential value into tangible outcomes 24 September 2025
2. What is it and why does it matter 2 Data Product Marketplace
3. The Prestige How new ideas get sold 3 A c t 1 The magician shows you something ordinary. Perhaps he asks you to inspect it to see if it is indeed real The PLEDGE The Centralized Monolith A c t 2 The magician takes the ordinary something and makes it do something extraordinary… The Mesh Nirvana The TURN A c t 3 … but you wouldn't clap yet. Because making something disappear isn't enough; you have to bring it back. The PRESTIGE The Viable System
4. The End Game Re-thinking the data mesh golden circle 4 Naive Mesh Pragmatic Mesh Delivering Fast Adapting Fast Modularize & Distribute Share & Co-Create Scaling Supply Scaling Demand WHY HOW WHAT
5. Economies of Scale Beware of what you want to scale 5 Supply-side Economies of Scale Demand-side Economies of Scale The costs of a product or service decrease as output increases. The value of a product or service increases as more people use it. Economies of Scale Diseconomies of Scale Cost per Unit Output/Scale Minimum Efficient Scale LONG RUN AVG COSTS Value per Unit Users/Scale Critical Mass Point DIFFUSION CURVE FOCUS Reducing cost per unit FOCUS Increasing value per user Bootstrap Problem
6. The Viable (Data) System You can't always get what you want, or not? 6 Variety Efficiency Viability VALUE V V V DISTRIBUTED GOVERNANCE Efficiency + Variety but Low Viability 👉 Rip & Replace (Tragedy of the Commons) Ex. Scheme-on-Read Data Lake FEDERATED GOVERNANCE Variety + Viability but Low Efficiency 👉 Dubious ROI Ex. Naive Data Mesh CENTRALIZED GOVERNANCE Efficiency + Viability but Low Variety 👉 Bottleneck (Tragedy of the Enclosure) Ex. Modern DWH Pick two It seems that in data management, you can't have all three….
7. Bipolar organizationa The tragedy of private governance 7 Business V wants Variety + Efficiency V so it asks for Freedom that generates Complexity and reduces Viability IT wants Viability + Efficiency so it asks for Control that generates Bureaucracy and reduces Variety 🤜💥🤛 Organization Operanting Model & Priorities Steer Steer Return Return
8. Data Code Infra Recommoning Data Management Form Modularization & Distribution to Sharing & Co-Creation 8 Business Distributed Governance + Variety + Efficiency - Viability + Viability + Efficiency - Variety V V Data Code Infra Siloed Data Application Monolithic Data Platform IT Centralized Governance Data Code Infra Data Code Infra Data Code Infra Data Code Infra Data Code Infra + Variety + Viability ? Efficiency V Business +IT Federated Governance Shared Pool of Reusable and Composable Resources Data Products Platform Services Somethings that start here mybe better managed here Somethings that start here mybe better managed here Create Resilience Create Differentiation
9. XOps Platfrom Mobilizing the Data Ecosystem 9 Data Product Marketplace Data Product Catalog Data Product Developer Platform Consumer Experience Operation Experience Developer Experience Vendors (just a random sample) XOps Platform Promotes Purpose Autonomy Mastery Drives Participation Accountability Cooperation Data Ecosystem Experience Continuum
10. Platform Strategy It's a long way to the top if you wanna rock & roll 10 create Data Product Developer Platform To infinity and beyond… govern Data Product Operating Model contextualize Data Product Catalog connect Data Product Marketplace Inception Foundation Mobilization Scaling Platform Platforming Early Adopters (13,5%) Innovators (2,5%) Early Majority (34%) Early Majority (50%) CHASM 🎯 co-create DIFFUSION CURVE Active Population Adoption Phases
11. Data Produtc Marketplace Capabilities From transactions to relations 11 Data Product Marketplace Trust Search What types of hammers are available? Asset What is the best hammer for me? Access Ok give me this hammer Reduce Cost of Transactions focus on finding and using ingredients (like Amazon) Transactional Engine Increase Value of Relations focus on collaboration and sharing of receipts (like Wikipedia) Relate What I need to hang a painting on the wall? Compose What I have to do to hang it? Co-Create Who can help me? How I can help others? Learning Engine
12. There are no silever bullets Complex problem require complex solution 12 Experience showed that people often figured out ways to organize and manage common resources through community-based governance methods rather than governmental ‘Leviathan’ or private property. Elinor Ostrom The only way to scale a complex system is by decomposition to the lowest level of coherent granularity and then allowing recombination; not repetition or aggregation. Dave Snowden
13. Adoption and evolution towards Data Product Catalog and Marketplace 13 ENI Data Mesh Journey
14. ENI – Main traits 14 Distributed worldwide Hetherogeneus Businesses Innovative in order to support Energy Transition People and their Knowledge at the core
15. ENI Data Mesh Journey - Timeline 15 2000 - 2021 2022 - 2023 2024 - 2025 2025 - >> Other Approaches Foundation Mobilization Scaling Distributed ownership across company data domains. Data treated as a product. Federated Governance Blueprint as accelerator for data platform initiatives DWH by unit or company (high data fragmentation) Central Datalake with no clear ownership and data governance hard to implement Self Service Data Access for users through Data Product Marketplace Collaboration features. AI Capabilities to support user and application interaction with the marketplace features. Adoption of Data Product Catalog as entry point for data product search and exploration. End Users are involved in the Data Governance process. Data Mesh Adoption starts here!
16. Foundation | Processes and Responsabilities 16 Definition of a «toolset» that supports the Data Mesh Adoption with processes, roles and responsabilities • Data Mesh Adoption Workflow • Data Product Canvas • Data Product Map • Operational Framework With this framework we have been able to define the foundation of Data Mesh adoption. Operational Framework
17. 17 Cloud Provider Services Data Services Blueprint Data Service that follow company rules (e.g.: security, networking), ready to use within project initiatives (aka building block) Standard cloud provider services scouting and selection. Group of Data Service combined togheter in order to provide a ready to use artifact at disposal of project teams for data product realization Data Platform Team supports the standardization for (data) services and architecture solutions at three different layers • Cloud Provider Services • Data Services • Blueprint Blueprint act as accelerators for time to delivery of Data Platform Initiatives Data Factory Data Lake Databrick s Event Hubs Data Factory Data Lake Databrick s Event Hubs Foundation | Platform
18. 18 Data Infrastructure Technical Metadata Layer Data Product Owner Platform Engineer ENI Data Mesh Network of interoperable Data Products SALES CUSTOMER Data Engineer .. to define an initial layer of metadata supported by the usage of blueprints Metadata layer as repository of technical metadata for Data Products First experimentation with classical Data Catalog tools, again, more focused on tech than business users. Technical Data Catalog Difficult to engage business users and thus slow data governance adoption Foundation | Architecture
19. 19 Data Infrastructure Platform Control and Utility Plane Data Product Owner Platform Engineer Enterprise Ontology Data USers Data Engineer Data Product Catalog to bring onboard business users in Data Product definition and Data Governance Data Product Catalog as an entry point for Data Product and not only Data Asset Business Users actively involved in the Data Governance processes starting from definition of business metadata and ontology concepts. Control and utility plane layer improvements to support metadata management and computational policies Mobilization | Architecture Data Product Catalog 1 3 0 SALES CUSTOMER 10 50 ENI Data Mesh
20. 20 Data Infrastructure Platform Control and Utility Plane Data Product Owner Platform Engineer ENI Data Mesh Data Product Catalog 1 3 0 SALES CUSTOMER 10 50 Enterprise Ontology Data USers Data Engineer .. Adding a Marketplace to improve User capabilities when searching or creating new Data Product Enabling of data interoperability across the company even considering different cloud providers Increase in the indipendency of users when accessing or searching products AI Capabilities to support complex searches, metadata definition and data quality processes. AI Capabilities Collaboration Features Data Product Marketplace Scaling | Architecture
21. 21 Incremental Approach Start from Data Mesh pillars and real pain point. Then move to processes/roles and evaluate tools when the datamesh strategy adoption is clear Bring Value to Users Priorityze a roadmap where business become indipendent when searching, requesting access and composing data product by itself AI Capabilities AI features must help business and tech users to improve they activities on the platform / marketplace itself. Marketplace itself act as an «tool» for other AI applications. Let User Make Impact Even with the right tool Users need to be involved actively in the Data Governance processes since the beginning. Communities and Chapters of experts are welcome to consolidate relationship Key Takeaway 01 03 02 04
22. 22 @ Quantyca Cofounder of Blinadata.io & author of “Managing Data as a Product” andrea.gioia@quantyca.it Andrea Gioia CTO @ ENI Data Platform Delivery & Governance Center of Excellence roberto.grandi@eni.com Roberto Grandi Data & AI Architect Lead Thanks! Q&A & reach out! We’ll be at booth L102. Drop by to chat with us.