Rewired, 2nd Edition

CHAPTER 26Data products: The reusable building blocks for scaling

Think about how airports operate. Each airline doesn’t build its own gates, baggage systems, and runways. That would be impossibly costly. Instead, terminals provide standardized gates, security processes, baggage systems, and air traffic workflows that any airline can use. This allows thousands of flights to run each day, safely and efficiently.

Data products are like that terminal. They provide clean, structured, reliable data—ready for many downstream “flights.” You build the terminal once; countless flights use it.

As companies increasingly rely on data to innovate and grow, data products have become a top strategic issue for scaling AI.

With data products, use cases can be delivered as much as 90% faster. The total cost of ownership of AI solutions, including technology, development, and maintenance costs, can improve by 30%. And finally, risk and data governance burden can be substantially reduced.¹

This value results from teams packaging data into well-defined, reusable “products” each built with a clear purpose, clear documentation, and clear ownership. They are built in such a way as to be easily added on and used for a wide variety of use cases. Data products have made data easier to find, trust, and use by anyone who needed it for analytics, AI, or decision-making (Exhibit 26.1).

The costs of developing a data product are largely one-time investments that, in effect, amortize as the data product ...