The Data-Driven Difference
I’d like to share one of the best explanations of data-driven decision making that I’ve ever encountered; it comes from a Forbes article published back in March. I’ll paraphrase here, but I also encourage you to read the full article.
Most people have had the experience of buying a car. In making a decision about what car to buy, the first step is to create criteria that are important to you like gas mileage and price point, for example. Based on these criteria, you narrow down the field of all available cars to just a few that you’re interested in. During this process, you might use some outside data – Kelly’s Blue Book or Consumer Reports, for example – to support your decision making. From there, you test drive the cars and decide which one you’d like to buy. This is an analytic decision. You’re making a decision on which car to buy based on your criteria and facts about each car, and thus your decision is data-driven.
The opposite approach, the non-analytic approach, would start at the test drive phase without any pre-set criteria. As you were driving various cars, you’d figure out your criteria by which cars you rejected or you’d simply by the first car that “feels” right. In this case, your decision isn’t based on data, but rather a “gut feeling”. This car may or may not fit the criteria that are most important to you, and without having done much research, you can bet that there was probably a better option you could have chosen.
Doing your research first, as a sort of personal due diligence, is expected when you make large decisions in your personal life like where to live or where to work. This same diligence should also be commonplace in our business lives. We want to make sure that we’re making solid decisions for our companies based on all of the available data.
All of this isn’t to say that gut feelings aren’t based on data (in fact I’d argue that they usually are to some extent), but that they are based on subjective data. Gut decisions are often related to feelings, and feelings are mutable depending on the amount of time that has passed since the event and the intensity of a feeling, among other things.
For example, I recently visited an ice cream shop that I went to a lot as a kid. From memory, I’d rate their ice cream as top-notch; we’re talking better than Ben and Jerry’s or homemade. Years later, when I tried their ice cream again, I found that it wasn’t as out-of-this world great as I would have predicted. It was really good, but if I were to put it to a taste test, it would probably come in on par with other ice creams. My rating from memory is a subjective one (and is probably a bit inflated), whereas a taste test would be a good objective way to tell exactly how good this ice cream really is.
Like the taste test or the car information from Consumer Reports, data is the objective test that can be applied to expand your understanding of a business problem. Predictive modeling and other types of data analysis aren’t meant to replace human judgment – they’re meant to enhance it. If your data contradicts your gut feeling, it’s an opportunity to take a closer look at your dataset. Being data-driven allows you to see if there’s an angle you might have missed, diagnose problems with your data, and gives you a better understanding of your data in the process, all of which will help in the long run.
Harness the power of many.
Create and share reports and datasets across the enterprise, and put analytical power in the hands of everyone. Veera creates a truly data-driven culture. Try it for yourself today.