For many organizations, data is a bit like the comedian Rodney Dangerfield: it never gets any respect. Or at least, not the respect it deserves.
I’ve been working in the data field for more than a decade, and I frequently hear clients say that data is one of their organizations’ most valuable assets. But when I ask a follow-up question about exactly how valuable that data actually is, I usually get a blank look.
And when I ask even more basic questions, such as how many business data elements (BDEs) the enterprise has — that is, unique categories of data such as customer date of birth, SSN, etc. — the answer is either vague or far off the mark — sometimes by a factor of 10 or more.
For the record, the typical enterprise tends to use somewhere in the range of 1,000 BDEs to operate all its business processes and activities. This fact leads to something we call physical redundancy, since there are actually tens of thousands of physical data elements stored in multiple locations, including mainframes, databases, spreadsheets, tables and so on.
One more thing to note: Not all data is an asset — in fact, some is a liability. Older organizations, or at least those with significant amounts of legacy data, may have personally identifiable information (PII) and/or other data issues honeycombed throughout their legacy data. My sense is that newer tech companies such as Facebook, Amazon, Google, etc. don’t have to attend to the legacy problem that older organizations do (or at least not as bad); therefore, their ‘data debt to asset’ ratio is more favorable.
But in any case, I think that for most organizations, data as a commodity deserves more respect. If it’s truly one of your most valuable commodities, shouldn’t you be able to accurately measure how much of it you have, and what its financial value is?
The right tool for the job
Fortunately, the emerging science of data valuation can at last provide the hard-nosed, impossible-to-ignore numbers that will help the value of data take its rightful place next to (or even above) the value of such physical assets as buildings, desks and computers.
Before I get to the key elements that comprise the data valuation process, however, a word of caution. There are many highly-regarded companies that claim to offer technology solutions that will facilitate the data valuation process. The challenge is, there are three aspects of data that define its value — physical, logical and conceptual — and the leading technology solutions only provide insight into the first, a little of the second, and none of the third. As a result, they only provide a partial solution, and potentially misleading results.
A complex process, but definitely worth it
In contrast, a more effective data valuation process includes several interrelated components and processes:
Metadata repository. This repository contains all the “tribal knowledge” about the physical and logical aspects of your metadata — that is, insights into your data from the people who use it everyday. There are six categories of metadata to be gathered: business, technical, core, data quality, people, and search.
Business data element inventory. You also need a defined and living inventory of your physical and logical BDEs that allows you to compare and account for data value. When used together, your BDE inventory and metadata repository can basically function as the chart of accounts for all of your data.
Data quality and data governance operations. Start by profiling your data, then work to assess and improve it. Meanwhile you must implement data governance — preferably overseen by an Office of Data led by a Chief Data Officer — to provide for a more systematic approach to information asset management.
Collaboration around data. You also need to make sure your data stakeholders can collaborate around the data in real time — much like the way accountants share information about physical assets throughout each monthly or quarterly close cycle.
Analyze data’s bottom line impact. Choose your most important metrics — whether it’s EBITDA, net income, or even monthly cash flow — and use regression analysis to find meaningful correlations between changes in your data and your bottom line. You’ll gain insights into your data’s financial value today, and also the ability to make more accurate forecasts of future performance.
In case you’re on the fence about taking on the challenge of data valuation, consider just a few of the benefits. Data valuation allows you to:
Develop an accurate understanding of how and where your organization should allocate resources to improve the quality of data for better ROI
Give your management team the material they need to make more accurate operational forecasts for the organization
Gain insights into setting defensible prices for your data, should you decide to pursue opportunities to monetize data
Keep data value on your radar
Going back to my original point about data not getting enough respect as an asset with actual, demonstrable value, I might change my analogy. For many organizations, it’s not just the Rodney Dangerfield of assets, it’s more like the Amelia Earhart — completely missing from the balance sheet.
This is why I think that data valuation is so much more than an academic exercise. It actually helps you make smarter and far more strategic decisions about your data — and specifically, how to leverage it for maximum advantage.