How Valuable is Your Company's Data?

By now, most of us believe that our data is valuable. We protect it, work hard on collecting it, and invest generously in storing and managing it. But how valuable is this data really? Is it worth our money and time? 

The answer is yes.  It’s very valuable! So valuable that organizations and regulators are beginning to think about it as a balance-sheet-worthy corporate asset. In fact, last year, citing the increased role of data in economic value creation, the Chinese Ministry of Finance issued an “Interim Provision” allowing data resources to be included on the balance sheets, starting in 2024. Currently, GAAP and IFRS regulations don’t allow it in the US. 

So how should we think about assigning monetary value to data? And why now?

The availability of foundational Gen AI models and the functionality to securely augment them with private data (RAG) unlocked a huge range of topic-specific AI applications. However, these topic-specific use cases rely on complete, high-quality company data. For some organizations, this kind of data is the only differentiator – the raw material - the asset that defines their products. Those who get it right gain a significant differentiator and therefore a competitive advantage. They typically command higher valuations and attract significantly more interest from investors.

Much has been written about the potential effects of data on the economy and organizations. In general, experts agree that data can be considered an asset because it’s a resource that someone owns with an expectation that it will provide future value. But, how do we go from feeling that our data has some intrinsic value to putting a dollar figure to it?  

Several years ago, in his book Infonomics, Doug Laney proposed six methodologies for estimating the value of information. Unfortunately, his groundbreaking research came a bit ahead of its time and didn’t get much practical use. But the introduction of GenAI changed the data calculus and Doug’s estimation methods are more relevant now than ever before. I found four of them to be particularly useful today:

• Cost Value – The simplest and the most common way to value intangible assets. This method considers the cost of generating, collecting, and managing the dataset. Alternately it can also be based on the cost to regenerate or replace it.  

• Economic Value – This is the most strategically interesting approach, which attempts to estimate how data impacts revenue. If the revenue impact is unknown, there are two proxies for monetary outcomes: Business Value (benefit of improving a process) and Performance Value (benefit of improving a KPI) estimates.

• Intrinsic Value – This is a more complex method. It’s calculated by assigning scores to the relevant data attributes like completeness, quality, accessibility, timeliness, etc., and computing an aggregate. 

• Market Value – This method is for organizations that sell or license data directly. It is based on the potential sales price and the total addressable market.

Data is playing an increasingly important role in our digital economy and has become the primary value creator for many digital enterprises. And while assigning value to data isn’t straightforward, entrepreneurs, investors, and regulators are beginning to react and recognize that data is an asset that should eventually find its way onto companies’ balance sheets. 

Do you agree? Which one of the methods above would you use to value your data?