In the course of my day job as director of research in data warehousing at Giga Information Group, I answer a lot of client inquiries. I recently received a question from a customer who wanted to know how to model the value of her firm's data as part of an insurance policy covering data warehousing disaster recovery. What is the quantitative measure, in dollars, for which the policy is to be written?
Perhaps the value of the data is determined by the cost of capturing, managing and consuming it. Thus, the data inherits features of the hardware, software, user interfaces, software development capabilities or business processes that surround and define it and which the data, in turn, enables. In an oversimplification, add the costs of the computing plant and utility function and divide this sum by the weighed number of data components. That is the proposed value of each component. Call this the "cost model." This approach has limitations, though it may be useful for disaster recovery purposes.
It is easy to provide a counter-example to this approach. When I'm hanging from a rope off the side of Mount Everest, the value of the rope is much greater than the replacement cost. The reason for this is perhaps open to interpretation. However, one answer suggests that the data inherits features of the context and of the way in which it is used. This sometimes gives the data value extending far beyond the immediate context. The cost of the data is significantly different than its value. Value includes cost, but the total economic impact (i.e., the value) of the data includes other business benefits and risks. What has actually been captured here is the cost; the value has escaped.
This at least gets us to a definitive answer: the data, itself, is worthless. The value exists in the application, service or product where the value-adding function appears and, incidentally, for which the customer pays. No one is interested in the data which is dull, dumb and useless as dirt. However, that too seems like an oversimplification. Maybe the data is worthless in isolation, but it is valuable when it is exposed at an end-user interface where a business analyst can use the data to answer a business inquiry. Thus, the question is not really about data; it is about information. No one is interested in the data, but the information has value.
The third suggested answer goes to the other extreme. Without the information, the firm in the question goes out of existence. So, perhaps the value of the information is equivalent to the value of the firm itself. Disaster-recovery planning in an IT department suggests a scenario such as this. Without a backup of the databases, the firm is out of business.
So, we have the proposed equation: value of the data = value of the firm. But what about the contribution of the other factors of production labor, land and capital? Subtract those factors, and the leftover is the value of the information, the so-called intellectual capital. For example, if the value of Megasoft Consultants' assets is $10 billion and its market capitalization is $100 billion, then the value of the information and information-like capabilities such as brand would be $90 billion. That might be one way of arriving at an estimate of the value of the information. This works as a possible macroeconomic approach.
Now consider a true story. A bankrupt dot-com company wanted to auction its customer list on a site such as eBay. Had such an auction been allowed to go forward (the U.S. Federal Trade Commission stepped in and canceled it on the grounds that it would violate commitments to customer privacy), we would have a data point as to the value of that customer list. Let's say a one million-entry customer list sold for $1 million. We would then know the value of one customer record.
What's the model by which to value data? This is a trick question. The model is provided by the economic principle of supply and demand that sets the value of the exchange of goods and service by a market exchange mechanism. That model is documented in any economics textbook. However, that the value will be impacted by disruptions in the demand or discontinuities in the supply is not generally an insurable risk. Still a useful lesson is provided.
The recommendation to clients is to recognize that data is worthless. The economic value of data is in its use as information in the context of the functioning business operations it enables. Costs of data creation and maintenance are relatively easy to capture but are incomplete. Much of the value of data escapes narrow cost analysis and becomes visible in the scenario in which lack of a data backup means the company is out of business. Data (as useful information) is an essential link in the value chain whose value point is determined in a market transaction (supply and demand). That is the model for establishing the value of information.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access