Why Chief Data Officers Need to Go Beyond Data Catalogs to Deliver Value
Businesses across the globe can proactively turn their data into value by sharing it effectively with both humans and AI.
- By Franck Carassus
- October 14, 2026
Organizations are generating and collecting increasing volumes of data, and chief data officers understand the importance of scaling its use across the organization and beyond. This is especially true with the adoption of AI. Without access to high-quality, reliable data, large language models (LLMs), AI agents, and generative AI can hallucinate, deliver biased results, or simply not work correctly.
However, CDOs are often held back by issues around complexity, silos, governance, and trust. Available tools for data sharing are technical, rather than meeting the needs of business owners. The result is that data remains solely used by data experts, rather than being delivered widely to business employees, reducing its value and adding to the pressures on data teams. AI use cases can’t be scaled due to the lack of trustworthy, contextualized data foundations.
Essentially, the data is there, but it cannot be confidently accessed by those that need it. I’ve talked to companies where each department has its own jealously guarded set of data, each of which is subtly different. No one trusts data and there’s no single source of the truth, leading to constant infighting and confusion.
Why the Modern Data Stack Is Not Enough to Deliver Value
Organizations have invested heavily in data management solutions, from storage and integration to tools such as data catalogs which provide an inventory of data. The issue is that all of these solutions are focused solely on reactive management and governance and are designed to be used by technical and data experts. Business users find it hard to use tools such as data catalogs, which don’t give direct access to data in formats that they can understand and easily work with. One company I worked with complained that business users had just given up on their catalog, meaning there were less than two users accessing each report there. That’s a real cost and wasted opportunity.
It’s not just about the particular catalog vendor. One large business I know moved away from its existing data catalog to select what they thought was a more "usable" solution targeting business users. But there was no real change in business adoption. That’s because catalogs don’t ensure that data is standardized and packaged in ways that are both human and machine readable. All of this holds back the benefits of data consumption, such as improved decision-making, greater efficiency, and underpinning more effective AI agents and models.
The Importance of Data Products to Data Value
Data products, highly-consumable, business-focused data assets, help overcome this data availability challenge. They package data and any tools needed to access it into a single, usable and machine-readable product designed around a specific business need. Usage is governed by a built-in data contract, and every product has a visible owner who can be contacted in case of queries. Data products are continually updated based on user feedback, ensuring they meet business requirements. Building data products actually allows data teams to reduce the number of dashboards and reports they have to generate. One enterprise software provider was able to retire over a quarter of its 1,000 dashboards, saving $10 million dollars.
However, just creating data products is not enough. Just like consumer products, they have to be available, easily discoverable, and understandable by target users if they are to be consumed.
The Rise of the Data Product Marketplace
This is driving the emergence of data product marketplaces within large businesses. These provide a single, centralized storefront that gives users self-service access to all of an organization’s data. Built on the same intuitive experience as an e-commerce marketplace, they don’t require technical knowledge to use and are designed to quickly connect humans and AI with the data they need. Granular access management capabilities protect data from unauthorized access, while lineage enables uses to be tracked and shared.
Essentially, they provide a discovery and consumption layer for the organization’s data, bringing together information and assets from cloud platforms, data warehouses, data storage, and other business systems. Features such as AI-driven search and chatbots further help non-technical business users access the information they need.
Marketplaces deliver “AI-ready” data that includes semantic, contextual, and usage information that guides AI agents to prioritize the best data for their needs. A business glossary provides a standardized vocabulary around data terms and what they mean in that organization’s specific business context. This helps create a common language around data across the business, with different departments using terms consistently to avoid misunderstandings about what data actually refers to. No more fights between departments about which data is “correct.”
The Importance of Culture and Experience to Increasing Data Product Consumption
Data marketplaces provide the technical foundations for greater data sharing by the business. However, technology alone is not enough to engage business users, many of whom aren’t familiar with working with data on a day-to-day basis. Success therefore requires a focus on two further factors: culture and experience.
Organizations need to train their people to confidently use data independently, which comes down to building trust that the data they are accessing is understandable and useful to them. That means taking a step-by-step approach—some companies immediately create hundreds of data products, rather than focusing on those that people want, which risks confusing business users with too much choice before they’ve even started.
Backing this up, data consumption platforms have to engage business users with an intuitive interface that makes the whole process of discovering, accessing, and consuming data as simple as using any consumer website. I’ve seen technical solutions designed by IT that no one uses because they can’t work out where to start, even if the data in them is relevant. Without these two essential pieces, the data consumption jigsaw won’t be completed.
Businesses across the globe are now looking to move from reactively managing their data to proactively turning it into value by sharing with humans and AI. Success requires a new focus, going beyond catalogs by adding a data marketplace to complete their data stack, backed up with the training and experience needed to turn data into long-term value.
About the Author
Franck Carassus is the co-founder and CSO of Huwise. Since 2011, Franck has helped 200+ organizations improve their governance or operations through data sharing. Prior to Huwise, Franck spent 20 years driving sales and business development for companies including HP Software, Symantec, and Novell. Contact him on LinkedIn.