This is Part 2 in a series about actionable ways to start treating your data like an asset. Read Part 1 here.
In the previous article in this series, I left you looking around your office and maybe seeing that storage cabinet in a whole new way. I left you with this question: Why is data on a laptop treated more like the pens in a company’s cabinet than the laptop or cabinet itself? Is this corporate habit bad?
The answer depends on what’s valuable to you and your organization. An asset:
- provides some value or benefit to your organization (or has the potential to), and
- can be managed (sourced, maintained, and decommissioned or disposed)
Someone in your organization is responsible for doing each of those things. Someone figures out what assets to put on the balance sheet. Similarly, someone is managing those assets and trying to maintain their usefulness and squeeze value out of them.
Data can be treated similarly. Once you have figured out what data you have, you then have to find its value. As an easy starting point, identify and list what basic benefits your data provides to your organization - this can be as superficial or as detailed as you want it to be, but the goal is to prioritize the data that is truly important to the functioning of your organization and figure out how it’s being managed.
There are some exciting developments in the data valuation space that people like Doug Laney have started to formalize in Infonomics (2017). This is a great resource to start with if you’d like to go deeper and identify which assets are worth your attention, and which ones may be underutilized. In this series, we’ll look at it from the asset management perspective. We’ll take the starting point that you have some assets with business value (formally or informally defined), and you want to find out where to direct your efforts to preserve their value and manage asset performance.
Some of your data may indeed be no more valuable than a pen, or the bolts that hold your equipment together. That loose collection of spreadsheets you have strewn across your C drive where you tracked last month's office party supplies, or the pro’s and con’s of taking the bus to work, probably has limited value. That equipment tracking sheet that you have to nag everyone to update with their monthly usage? Harder to tell. But some data could easily be just as valuable as the equipment itself.
For example, picture your complex data landscape with multiple interacting data pipelines and all the storage and compute that goes with it. This looks less like an office with cabinets, pens, and paper, and more like an industrial plant with pipes transporting chemicals, tanks storing them, and pumps and reactors processing them to create new materials. Throughout this data asset management series, I’ll explore ways to manage our data assets within a data factory.
If your initial benefit assessment reveals that your data isn’t providing as much value as you thought, that’s not necessarily bad. While this doesn’t mean you can ignore the quality of data that is not treated as an asset, it provides you an opportunity to prioritize what is important, rather than trying to maintain 100% quality in areas that aren’t as important to your business.
Picture this: you manage a maintenance shop. Your staff tells you that they order boxes (read: database tables) with hundreds of thousands of bolts (rows) every month, but the bolts are usually of really poor quality and aren’t trusted by your technicians. Some of the boxes are stashed in a storage room to gather dust. Other boxes are meticulously inspected by hand, piece by piece, when a bolt is needed for use.
It might sound silly to store supplies in a place where they’re likely to never be looked at again, or to rely on someone to meticulously inspect each bolt before they can use it. But this is exactly what so many of us do at our companies. Have you really looked in your data lake lately? How much effort do your users need to put into inspecting the data that comes out of your lake, if they’re lucky enough to find what they need in the first place? I think you would consider either of these scenarios problematic, an inefficient use of time and effort.
In the bolt example, it probably wouldn’t be your first thought to create a quality management system and strive for 100% perfection. You would, however, likely find a way to improve quality along your supply chain. This lets you focus your attention on maintenance of true assets where you can apply more rigorous asset management practices.
Assets provide value, but also have risks associated with them, and those risks can be identified and managed. We pinpoint where to do maintenance by knowing these risks, having a plan for when they manifest, doing inspections, and tracking failures in operations. Finally, to bring asset performance to the next level, we track asset reliability metrics. In upcoming posts in this series, I will introduce actionable things you can do to truly start treating your data like an asset, and describe how this relates to the trend of data observability.
Find out in Part 3 of this series, coming soon.
Ultranauts helps companies establish and continually improve data quality through efficient, effective data governance frameworks and other aspects of data quality management systems (DQMS), especially high impact data value audits. If you need quality management systems for your data pipelines, Ultranauts can quickly help you identify opportunities for improvement that will drive value, mitigate risks, reduce costs, and increase impact.
Laney, D. B. (2017). Infonomics: how to monetize, manage, and measure information as an asset for competitive advantage. Routledge.
Author: Peter Dobson is a data quality professional with a M.Sc. in Mechanical Engineering and background in industrial inspection and maintenance.