Who doesn’t love traditions? While I am not the most nostalgic guy, I certainly have a handful of traditions that I hold dear. I think as we blaze bravely ahead, it’s important to remember where we come from, both good and bad to influence a better future. Sure, we all have traditions on a personal level, but what most people don’t realize is that most businesses carry their own traditions. The problem, is that most of these traditions are bad for the business itself.
The Process Problem
The sanctity of the process creates many problems with data. Poor data quality usually stems from a bad input process. Poor query performance usually is related to a bad process that needs to be corrected at run time. Poor technology acceptance may be part of a bad process; if users can still go elsewhere, why come to you, the data supplier?
Many times I have seen organizations hamstrung by data issues. And, not infrequently, I have seen organizations have data issues due to business processes. Sometimes the data comes to the end of the process and is incorrect, sometimes codes need to be transformed into something useful, sometimes the data is dirty. I’m sure you can think of a few examples of your own. And most times I see this issue, the solution is to fix the data after the process is complete (and the data has already been written), not during. Usually this is the case because “we can’t change the process”.
I’ve said this before, but a good tool does not fix underlaying data issues. During data initiatives, this is especially applicable when people begin to defend the processes used. Pushing back on the process to get better data in and out is a win for everyone.
“Well, our process won’t allow us to do that.” This a roadblock phrase that I have heard numerous times when it comes to fixing data during the process. Sometimes this is actually accurate; sometimes the process is mandated and enforced by regulation, as seen in financial services and healthcare industries. Most of the time however, it is not a by product of mandate. Most of the time the process won’t allow something to happen because it takes effort to change the process.
“The process has always been like that”. This is another excuse that I hear often. Let me tell you a story to illustrate why this is such a poor thinking.
It’s holiday season and the whole family is together. Cousins, aunts, uncles (even the crazy one), grand parents, brothers, sisters and parents are all in attendance, and it will be a grand time. Of course, the best part of the day is eating, and my, what a feast it will be. During all of the excitement, one of the children wanders into the kitchen and sees some of the adults making the food. She stops and stares while she sees her mother cut the ham in half, place one half in the roasting tray, and the other half in the garbage. The little girl asks “Mommy, why did you do that?” The mother replies, “Well, that’s how your grandmother always made it when I was a girl. She always made the best ham, so I am just doing what she did.” The little girl looks to her grandmother and asks “Grandma, what does cutting the ham in half do? What makes it taste so good?” The grandmother laughs and says “I always cut the ham in half and threw half away because the entire thing didn’t fit in our small oven! There’s no need to do that now!”
I cannot say this strongly enough: do not be like the mother. Don’t blindly accept that the process is the “secret sauce” to making the ham taste good. If we stop and question “the process”, we can start to think about what is necessary, what is the goal, and how to get there. Since most organizations do not think of data as an asset, or reporting, business intelligence or analytics as a positive business activity, it creates many of the issues I referenced above.
Here is an example of accepting tradition as truth. I was working with a client on migrating from a deprecated analytics platform to a newer one. One of the requirements was to ensure that the data matched between the old platform and the new one. As I was building metadata to handle the calculations of some of the metrics, I discovered gross errors in the math being used, which was leading to very different (i.e. incorrect) figures. I brought this up to the client and proposed a new method of calculating the metric in question that will correctly measure what they were trying to understand. Flabbergasted, the client said “That can’t be right. We have always calculated it the other way and it has always been right.”
The trouble with traditions is that at one time they made sense. They were deemed a good way of pursuing a goal, or remembering, or instilling consistency based on the understanding at the time. But as businesses grow more sophisticated, it is important to take off the blinders and restrictors that hold back understanding, higher quality and truth. That’s the challenge.
This train of thought also directly correlates with themes from my other blog posts. If the business lacks a vision for data, the data will not be considered when the business acts. And if the business does not care about the data, then data supply will be affected due to poor quality. This forces people to drill what I call “data wells” — silo-ed data marts, Access databases or even Excel workbooks that serve the needs for a small community. As I’ve discussed before, data should be a utility in the organization; one that provides access and value to all interested parties.
I’ll close with a quote from Jeff Bezos, who seems to know a thing or two about being an agent of change. In his last investor letter, he outlines ways to stay a “Day 1” company, which is a different topic for another time. However, he does have some words of warning about processes…
Good process serves you so you can serve customers. But if you’re not watchful, the process can become the thing. This can happen very easily in large organizations. The process becomes the proxy for the result you want. You stop looking at outcomes and just make sure you’re doing theprocess right. Gulp. It’s not that rare to hear a junior leader defend a bad outcome with something like, “Well, we followed the process.” A more experienced leader will use it as an opportunity to investigate and improve the process. The process is not the thing. It’s always worth asking, do we own the process or does the process own us?