Starting anything the right way can pay great dividends down the road, especially with data analysis. I do a lot of data analysis at CloudApp and I've started analysis the right way and I've started it the wrong way. You see, the wrong way is to just dive in and start clicking buttons, performing calculations, and pulling numbers with no direction and forethought. When I've done this I've found myself doing a whole lot of work but not really finding anything meaningful. Data analysis goes much easier if you start the right way.
What is the right way? Data analysis starts with a question. It starts with deep thought and mental analysis before a keystroke is ever made. The purpose of analysis is to find something that can be used to act, how are you going to find something useful if you don't know what you're looking for? You first need to think of a question.
I find that my analysis goes much better and my results are more meaningful when I start out by thinking deeply about current processes and events and forming hypotheses about why things are working out the way they are and ways they could be improved. I'm sure that to my coworkers or the casual passerby it looks like I am just staring out the window daydreaming when I am in my "deep thought" phase of my analysis but I can tell you that the mental energy I expend there is far better spent than if I were just clicking and typing around aimlessly looking busy "analyzing".
To me, the easiest part of data analysis is the actual analysis. The real work is figuring out what questions to ask, what data will be useful, and what would be actionable and make a real impact.
Data analysis without the hard work of thinking beforehand is just analysis for the sake of analysis and that's not very valuable to anyone. So do yourself a favor and spend some time thoroughly thinking things through before you start your next project, you'll be glad you did.
Comments