As a wise old man once said: You can prove anything with facts. Okay, it was Homer Simpson, but you have to admit, he has a point.
You need data to make data-based decisions, but the challenge is that the more data you have, the easier it becomes to misuse the data to serve the conclusions you want to draw. With large data sets and good tools, it may even be deceptively easy to do so.
When the data set is large, it is time-consuming to point out weaknesses in the analysis. And even if you do so, it might be too late to change anything, if actions have already been taken.
The ideal data processing workflow
If you are looking for genuine and sustainable improvements, the data processing workflow starts with an open mind. The ideal workflow looks like this:
- Ask open questions
- Collect data
- Refine data
And these steps need to be repeated several times to test different scenarios. Only then it is time for conclusions and decisions.
Doing high-level data analytics requires courage
This way of working requires some bravery, since results and facts can be ugly, unpleasant, or unwanted. To stay on track, you also need high morals and ethics. And you do want to stay on track, don’t you?