If someone asked me what a company’s most valuable asset is, I’d probably say its products, customers, or brand. But the more I observe how businesses are evolving, the more I think there’s another asset quietly climbing that list: the knowledge they accumulate every day.
Every purchase, delivery, support ticket, software click, and machine operation leaves behind more than a completed transaction. It reveals something about customer behavior, operational efficiency, or market demand. For years, companies used those insights simply to improve their own decisions. Increasingly, they’re discovering those insights can become products in their own right.
What fascinates me is that many businesses don’t even realize they’re building this asset.
They still think of themselves as retailers, manufacturers, airlines, or banks. Yet behind those familiar business models, they’re steadily creating proprietary datasets that competitors can’t easily replicate.
I think we’re entering an era where every business produces two products: the one customers buy and the knowledge created while delivering it. The second product is often invisible, but it may become just as valuable as the first.
| Era | Primary Core Asset | The Secondary “Invisible” Asset |
| Industrial Era | Factories, machinery, physical infrastructure | Assembly line metrics & labor efficiency logs |
| Early Digital Era | Software code, user bases, digital platforms | “Operational exhaust” (server logs, basic clickstreams) |
| Modern AI Era | Core products & services | Proprietary behavioral datasets that train exclusive AI models |
Every Interaction Produces Two Products
I’ve started looking at businesses a little differently. Every interaction, in my view, produces two products.
The first is obvious- the product or service a customer pays for. The second is the knowledge created while delivering it.
A grocery purchase reveals buying habits. A logistics shipment uncovers traffic patterns and delivery bottlenecks. A payment transaction exposes spending patterns. A customer support call highlights recurring product issues. Individually, these interactions don’t seem particularly valuable. Repeated millions of times over months and years, however, they become something much harder to replicate: a proprietary understanding of customers, operations, and markets.
For decades, companies treated this information as operational exhaust. It helped optimize inventory, improve customer service, and streamline internal processes, but it rarely left the organization.
That mindset is beginning to change.
Increasingly, businesses are realizing that every interaction leaves behind an asset that compounds over time. In some cases, that asset can become valuable enough to support entirely new products, services, or revenue streams.
The more I think about it, the more I believe many companies now operate two businesses simultaneously. There’s the visible business customers recognize, and then there’s a quieter knowledge business that grows stronger with every transaction. Most executives still spend most of their time thinking about the first. I suspect the second may become just as important over the next decade.
From Operational Exhaust To Commercial Asset
The idea sounds abstract until you start looking at companies that are already doing it.
Take Kroger. Most people know it as one of America’s largest supermarket chains. Less visible is 84.51°, its data analytics business. Built on billions of grocery purchases, it helps consumer brands such as Procter & Gamble and Unilever understand shopping behavior, optimize promotions, and make better merchandising decisions. Kroger isn’t simply selling groceries; it’s commercializing years of customer understanding.
John Deere tells a similar story from a completely different industry. Sensors embedded across its agricultural equipment continuously collect information about soil conditions, crop health, weather patterns, and machine performance. That data powers precision agriculture tools that help farmers improve yields while creating recurring software and analytics revenue far beyond the original equipment sale.
Neither company set out to become a data business. They became one by recognizing that the knowledge created while running their core business had value beyond internal decision-making.
What strikes me is that neither company is really selling data. They’re selling better decisions. One helps consumer brands understand shoppers more clearly. The other helps farmers make more informed decisions in the field.
That’s an important distinction because businesses rarely pay for information alone. They pay for reduced uncertainty.
The Biggest Asset Isn’t Data, It’s Proprietary Behaviour
The phrase “data is valuable” has been repeated so often that it’s almost lost its meaning. Not all data is valuable. In fact, most of it isn’t.
Public information is widely available. Weather forecasts, economic indicators, demographic statistics, and even many market reports can be accessed by almost anyone. If everyone has the same information, it rarely creates a lasting advantage.
What companies can’t easily buy is a decade of proprietary customer behavior.
Amazon knows how millions of customers respond to delivery delays. Visa can identify spending patterns across industries. Netflix understands not just what people watch, but when they pause, skip, or abandon a show. These insights don’t exist because the companies collected more data than everyone else. They exist because years of real-world interactions created knowledge that competitors simply can’t recreate overnight.
I think that’s where the real value lies. The most defensible datasets aren’t necessarily the biggest, they’re the most unique. They’re built through trust, scale, and time rather than purchased off the shelf.
As AI makes information easier to analyze, proprietary behavior becomes even more important. If every company has access to similar AI models, the real differentiator won’t be the technology itself. It will be the exclusive datasets that teach those models something no one else knows.
The Hidden Risks of Monetizing Knowledge
Of course, turning knowledge into a business isn’t nearly as simple as packaging a dataset and finding a buyer.
The biggest challenge isn’t technological, it’s trust.
Customers may willingly share information to receive a better product or service, but that doesn’t automatically mean they expect their behavior to become part of a commercial asset. Regulations such as Europe’s GDPR, California’s CCPA, and India’s Digital Personal Data Protection Act have made it clear that companies must earn that trust through transparency, consent, and responsible governance.
Even anonymization isn’t a perfect solution. As AI becomes better at connecting information from multiple sources, researchers have shown that supposedly anonymous datasets can sometimes be re-identified when combined with other publicly available data. That raises both ethical and reputational risks.
I think this is where many companies will succeed or fail. The businesses that create lasting knowledge products won’t necessarily be those with the most data. They’ll be the ones customers trust enough to keep generating it.
In the long run, trust may become an even stronger competitive advantage than data itself. Without it, the second product disappears.
The Real Competitive Advantage Is Time
Most competitive advantages can eventually be replicated. Competitors can build similar products, adopt the same technologies, hire experienced talent, and even copy successful business models. Given enough time and capital, many advantages become temporary.
Proprietary knowledge works differently.
A competitor can buy the same AI software, install the same machines, or launch a similar digital platform. What it can’t buy is ten years of customer behavior, operational decisions, supplier relationships, and real-world outcomes. Those insights can only be accumulated through continuous interaction and experience.
That’s why I think time is becoming an increasingly valuable competitive asset. Every year a company operates, it isn’t just generating revenue, it’s expanding a proprietary knowledge base that becomes harder for others to replicate.
AI only reinforces this dynamic. As advanced models become widely available, access to the technology itself becomes less of a differentiator. The companies that stand out will be those feeding these models with exclusive, high-quality data that reflects years of real-world experience rather than publicly available information.
This is also why data licensing isn’t simply about creating another revenue stream. It’s about recognizing that one of the strongest competitive moats a business can build isn’t physical infrastructure or even technology. It’s the unique understanding that compounds over time.
The businesses that recognize this early may discover they’ve been building their most defensible asset without fully appreciating its value.
The Companies That Win May Commercialize Understanding
I don’t think every business will end up licensing datasets or launching a data platform. For many, that may never become a meaningful source of revenue.
But I do think almost every business will start asking a different set of questions. Instead of viewing data as a by-product of operations, they’ll begin treating it as intellectual capital that can strengthen products, create new services, deepen customer relationships, or, where appropriate, become a business in its own right.
In other words, the opportunity isn’t simply to sell data. It’s to commercialize understanding.
That distinction matters because businesses aren’t creating value from raw information.
They’re creating value from the insights, predictions, and decisions that emerge after years of observing customers, markets, and operations. Those are assets that become richer with experience and far more difficult for competitors to replicate.
Looking ahead, I think the companies that build the strongest competitive positions won’t necessarily be the ones with the largest factories, the biggest customer base, or even the most advanced AI models. They’ll be the ones that consistently transform everyday interactions into proprietary knowledge and then find new ways to put that knowledge to work.
For decades, we measured businesses by what they produced. Increasingly, we may also measure them by what they learn.
A Simple Audit for Your “Second Product”
If I were leading a business today, I’d probably ask four questions before thinking about data licensing.
First, where does the business generate repeated interactions at scale? That’s usually where proprietary knowledge begins to accumulate.
Second, what do we know that competitors can’t simply buy or scrape from the internet? Public information rarely creates an advantage. Exclusive behavioral insights often do.
Third, can we use that knowledge without compromising customer trust? If monetization damages privacy or confidence, the long-term cost is almost always greater than the short-term revenue.
Finally, are we selling data or helping someone make a better decision? The most valuable knowledge businesses don’t monetize spreadsheets. They reduce uncertainty. Whether that’s helping a farmer improve yields, a retailer forecast demand, or a bank detect fraud earlier, customers ultimately pay for better outcomes rather than more information.
I think those four questions are becoming increasingly important because they shift the conversation away from collecting more data and toward creating more value from the knowledge a business already possesses.
Conclusion
It’s easy to think of data as something businesses collect. I’m beginning to think that’s the wrong way to look at it.
The real asset isn’t the data itself, it’s the understanding that emerges from years of serving customers, running operations, and solving the same problems repeatedly. That’s the kind of knowledge competitors can’t simply purchase or recreate overnight.
Of course, not every company will build a business around licensing data. For many, the bigger opportunity will be using their proprietary insights to develop better products, improve decisions, or create entirely new services. But the underlying shift remains the same: knowledge is no longer just supporting the business, it is becoming part of the business.
That’s why I believe the companies that stand out over the next decade won’t just be those that make the best products. They’ll be the ones that learn the fastest, accumulate the most distinctive insights, and find new ways to turn that understanding into lasting value.
Perhaps the most valuable thing a company creates isn’t always what leaves its factory, warehouse, or website. Sometimes it’s what stays behind after every interaction, the knowledge that quietly compounds over time.
