As analysts, we need to work with complexity, while simplifying for end-users, yet avoid oversimplifying. Naturally, this is easier said that done …
Simplifying for others: This is incredibly important. If you can’t explain the problem and findings to someone in 25 seconds or less, you 1) will likely lose their attention, and 2) possibly don’t understand it well enough yourself to explain it yet. That’s our job. We work with the details and bring others in on the conclusions.
Oversimplifying: The balance required is to simplify the final conclusions without oversimplifying the problem, the data, or your analysis. The struggle, however, is that our brains are hard wired to simplify information.
Think about the amount of stimuli your brain receives every day. For example, you are crossing the street. I am outside. I see a long stretch of gray. That is a road. There is a red thing coming towards me. I hear a noise. The red thing is making the noise. That red thing is a car. Cars can hit me. It is going at 45mph. I am stationary. It will reach my location in 4 seconds. I will take 10 seconds to cross the street. I should not walk yet. And of course, I’m completely understating all that goes through our brains for even simple tasks. If our brains didn’t find a way to make sense of a high volume of inputs, we simply wouldn’t function.
Acknowledging Complexity: The challenge for analysts is to try to simplify the answer, without oversimplifying the questions along the way. If you make erroneous assumptions because it (over)simplifies your analysis, you could end up drawing the wrong conclusions. You will probably make your analysis easier, but render it less valuable.
We need to acknowledge, work with (and enjoy) complexity. (And we had better get used to it, because the digital measurement space is not getting simpler.) However, we need to avoid oversimplifying more than is necessary to sift signal from noise. We need to question what we know, evaluate what we assume, and separate fact from opinion. And if in doubt, invite someone else question you or poke holes in your analysis. Chances are, they’ll spot something you didn’t.