It was March ‘95. I was in Michigan, “fresh off the boat” from Australia and my American cousin — Matt — was driving me to a movie theater.
About halfway into the trip, I asked Matt to stop somewhere so I could grab tea. I’m pretty sure I got a side eye look. The look said “sure, I’ll be polite and accommodate your weird Aussie needs”. I assumed he was concerned that we’d miss some of the movie.
You see, “tea” in Australia has a couple of meanings. It usually refers to the leaf- based drink known the world ‘round but it can also mean dinner (yes, that last meal of the day). And, being naive, I thought nothing of using the word for this purpose.
So Matt dutifully navigated his car through the back streets of Kalamazoo with a new destination in mind. After a few minutes, we arrived at Meijer (a large supermarket chain). Now it was my turn to give a side eye look.
Whenever I find myself in a peculiar situation, I tend to rationalize the weirdness away. Having heard about supermarkets in the U.S., I convinced myself that this one had some kind of cafe or food hall. Why else would Matt bring me here? We entered Meijer, barreled past the cooked food, zig-zagged through a maze of seemingly endless aisles and it didn’t dawn on me until we reached the ‘coffee and tea’ section why we were there!
The knowledge that tea was synonymous with dinner was so engrained that I was blind to the possibility that it didn’t exist everywhere.
Fast-forward 23 years later, I’m a little wiser (certainly more assimilated) yet I continue to be confronted with similar challenges. Thankfully, no more unsolicited trips to the supermarket but dealing with data forces me to be more cognizant of definitions and the possibility that seemingly simple words might mean different things to different people. For example, an invoice could be both a document and a process. IT might view an invoice as the process of creating a bill while Tom, in accounting, is adamant that invoices can only be documents used to request payment from customers. And, while these differences in opinions may seem harmless on the surface, they can have serious negative consequences when critical decisions, models or systems are dependent on them.
A few months ago, I was working with a senior engineer from a team tasked with tracking product usage. They were tracking customer events and one of those events was product websites log ons. With product teams free to name events as they saw fit, there were as many “log on” events as there were “log in” events leaving my colleague with an indexing nightmare. A difference of just one letter led to index saturation and poor system performance due to the need to overcompensate for poor standards.
Without a common business vocabulary or standard naming convention, ambiguity will thrive and subject the most buttoned-up organizations to vulnerabilities, misinformation, inefficiencies and missed opportunities. This is where standards come in. Actually, this is where governed standards come in. Put simply, an enterprise must breakdown silos and work together to gain consensus on names and definitions of the business terms and data elements upon which they’re most reliant.
Get our Updates as They Happen
Subscribe to our blog to get new posts delivered straight to your inbox.
I’m not going to sugar-coat it. It’s hard. Really hard.
It takes buy-in from stakeholders at all levels then coordination, collaboration and alignment across teams and departments. And even when you do achieve this holy grail of standards, if left unmanaged, the glossary becomes stale and is relegated to some unmanned corner of an organization’s intranet with fewer hits than Buford, Wyoming’s travel information site.
Fortunately, there are self-service tools that employ crowdsourcing and social collaboration to accelerate and simplify the task of building, maintaining and governing business glossaries. Even a basic wiki will work but some go further with traceability to physical data sources and seamless integrations with analytics platforms. Integrations, bells and whistles aside, the important attributes are — ease of use, accessibility, built-in collaboration and low maintenance. With the hard part out of the way and a glossary established, a smart investment in the right tool will go a long way to ensure the information remains relevant, accurate and trustworthy.
If it’s so hard, what’s the point? According to IBM…
The average information worker spends 3.5 hours each week searching for but not finding the information they were looking for. For a company of 1,000 employees, that is a staggering US$5,250,700 per year!
My cousin is now a Doctor. Like me, he relies on data standards to do his job efficiently. In his line of work however, the consequences of data ambiguity can be a lot more dire.
Building a business glossary is the first big step toward an effective data governance competency. It will help break down organizational and technical silos and deliver the kind of insight, control and collaboration necessary to leapfrog competitors in any industry.
As Director of Enterprise Analytics, James helped Thomson Reuters establish data management capabilities and an enterprise-wide analytics competency.
- Data and the Post-Pandemic World - June 30, 2020
- Data-Driven Decision Making in Recessions - June 16, 2020
- What You Need To Know About Data Driven Decision Making - May 26, 2020
Get our Updates as They Happen
Subscribe to our blog to get new posts delivered straight to your inbox.
