Barb Wixom and MIT CISR on managing data like a product

If data is the new oil, too many CIOs are still stuck building barrels instead of businesses. Despite steady investment in data platforms and governance, many organizations still struggle to extract lasting value from their data. According to Barb Wixom, principal research scientist at MIT Sloan’s Center for Information Systems Research (CISR) and co-author of Data Is Everybody’s Business, the answer isn’t more tools or talent. It’s a mindset shift: treating data as a product, the way information businesses do. 

Wixom traces the roots of her recent research brief back more than 25 years, to her time teaching in the University of Virginia’s MS program in the Management of IT. “Several students, leaders from data-savvy companies, kept pressing for real-world practices they could model,” she recalls. “They wanted examples they could put into action, but often came up short looking at traditional industries.” Her advice was simple and catalytic: “Study organizations whose entire business runs on data. Watch what they do, and you’ll see the future.” That challenge sparked decades of investigation into how information businesses master data, and what every other enterprise can learn from them. 

The case for managing data like a product 

In information businesses like comScore or LexisNexis, data isn’t just an input, it is the product. These companies succeed by creating reusable, scalable data assets, building solutions around them and monetizing the duo. “[In information businesses] you’ve always had owners for both the data asset and the solution,” Wixom explains. “They’re core strategic resources; these companies treat them as such.” 

That principle holds true in any industry. Organizations unlock value by managing data across a lifecycle: assigning owners, setting financial goals and continuously evolving offerings based on demand and usage. 

Two products, two playbooks 

Wixom emphasizes the importance of distinguishing between the two types of data products: 

• Data assets are curated datasets meant for broad reuse. Their owners aim to maximize utility across internal consumers. Think of these as “raw materials.”
   
• Data solutions are the vehicle for value delivery: dashboards, APIs and applications that yield specific outcomes. These often have owners with P&L experience and are focused on revenue, cost savings or customer value.
   

While both types require a product management mindset — ownership, lifecycle thinking, user-centric design — their performance metrics differ. “Asset owners think in terms of reuse and internal market development,” says Wixom. “Solution owners think about cost, margin and actual dollar value.” 

Managing data like a product also means product owners must focus on customers, cost management and continuous improvement. 

Customer satisfaction comes first. For asset owners, that means staying tightly aligned with solution teams to ensure their datasets evolve to meet real needs. The relationship is symbiotic: assets are only as valuable as the solutions they enable. For solution owners, success hinges on deep user engagement: co-creating with stakeholders, running experiments and analyzing how value is delivered. 

Pricing plays a key role, too. Data assets are often priced indirectly, via chargebacks or cost allocations, to promote efficient use. Solutions may be priced directly when external, or indirectly when used internally. 

And profitability can’t be overlooked. “If we’re managing data like a product,” says Wixom, “we need to do so profitably.” Asset owners instrument their offerings to track usage and understand value. Solution owners control costs and adapt continuously to evolving market needs, often requesting changes to the underlying assets as a result. 

Why traditional data management falls short 

Most data governance models are designed for compliance and access, not value creation. “You might have a system where someone pulls data for a one-off analysis, then moves on,” says Wixom. “That’s not a data product.” 

Building true data products, she argues, requires standardization, instrumentation and iteration. “It can take one to four years to build a true data asset,” she notes. “But once you do, you unlock scale, and you build organizational capability in the process.” 

This shift also raises the bar for governance. High-quality, “liquid” data assets must be versioned, cataloged and trusted, because they’ll be reused in dozens of ways. 

Where tech chiefs should start 

Not every company is an information business, but any company can learn from one. Wixom urges data leaders to start by identifying data that is both broadly relevant and strategically important. “You may have lots of data, but few true data assets,” she warns. “Assets are clean, curated and ready to go, with little wrangling needed.” 

Customer data is a typical starting point, especially for companies seeking margin gains or better customer experience. “Think about data that can power multiple initiatives – analytics, personalization, mobile features,” she explains. “That’s your candidate for productization.” 

The path forward is incremental. “You can start with just six fields,” she says. “The key is assigning ownership and building from there. That’s product thinking.” 

From stewardship to strategy 

Wixom’s message is clear: managing data like a product isn’t a future state, it’s a competitive necessity. As organizations pursue AI, automation and digital efficiency, reusable data products become the fuel for innovation at scale. 

“We’ve proven it in the research,” she says. “You don’t have to sell data to monetize it. If your data improves the customer experience, cuts costs or enables new features, that’s value.” 

For digital and tech leaders, the takeaway is simple: stop treating data as a static resource. Start managing it like a strategic product. The next phase of your transformation may begin not with a new tool, but with a new mindset. 

This article is published as part of the Foundry Expert Contributor Network.
Want to join?