Data is a Valuable Asset. But It’s Not Free.

In the first post in this series I defined an asset as something that has value to the organization and is legally owned and controlled by the organization. The usual assets we think of, cash, securities, inventory, and property and equipment, all meet this definition. Other assets such as intellectual property and brand value meet this definition as well, even though they are not tangible, and usually not liquid. They are still assets.

Data is not tangible, and in most cases is not particularly liquid. As we saw, however, this does not disqualify data from being an asset. Data, when acquired within the bounds of regulatory and legal requirements, is owned and controlled by the enterprise. (There are limits, as I’ll discuss in a subsequent post.) And data certainly has value. Given the definition of an asset, data certainly qualifies.

But as an asset, data has some unique characteristics.

Many assets wear out. Equipment is depreciated according to its expected useful life as an approximation to its decrease in value over time. When IP or brands are purchased and put on the balance sheet as an asset, they are amortized over time. IP loses value in the market as its novelty diminishes, and brands lose value without constant expenditures in marketing and brand management.

Data seems unique. Data doesn't wear out with use. Data can be copied repeatedly for free with no degradation in quality. This seems to make data a rather unique asset, since it can be repeatedly sold without giving up possession. The corollary to this characteristic is that data can be stolen without anyone realizing it. That data is infinitely reproducible has an upside, but also a large downside, which requires a considerable expenditure in information security to mitigate.

While data doesn’t wear out, data is perishable. When a customer moves, their old address is useful for historical and audit purposes, but it's useless for collecting on a past due invoice.  Specs on discontinued products are not very useful to a purchasing agent sourcing a new component. Data never disappears, but its value degrades without a continuous investment to keep it fresh.

Unlike many other assets, data can also be wrong. Erroneous data not only loses its value, identifying and correcting the errors costs money and damages customer satisfaction and operational efficiency. Just as with defective products that are returned and replaced, poor quality is expensive. And just like quality engineering in manufacturing, enterprises need to engineer in data quality through data quality control processes.

In summary, data is an asset, not metaphorically, but literally, just as cash, inventory, and plant and equipment. However, the value of data does not come for free. Generating, utilizing, and protecting data takes effort. Which I will discuss in the next post in this series.