A famous scientist once told me that in science the answers are easier than the questions.  “Formulating the right questions” is the most intricate and demanding part of the scientific method, he opined.  Certainly a fascinating point of view.

On reflection, I began to think that this maxim could apply as well to much of what we do in the world of data.  While not as grandiose as science, the data work we do in business depends a great deal on the degree to which we can formulate problems to solve; solutions can be built but formulations require deep knowledge and imagination.

Take the issue of business-users needing data to make decisions.  We all know that once we get the right data that we can at least attempt to build a set of processes and procedures whereby we visualize and apply this data to the problems at hand.  We might or might not succeed but the problem is tractable.  If, however, we have no idea what sort of data we need, have no idea what the right questions to ask “the system” then, we’ll, garbage-in, garbage out.

In a world awash with data, being able to cut through the noise by asking the right questions is crucial.  Then one must go through the process of determining which data is relevant, finding sources of that data, ingesting, digesting, and visualizing the data.  Then one has a fighting chance of being data-driven in the business.

To do this well, the key is to accelerate the process so that when the hard-work of problem formulation is done, the rest can be done in a timely fashion without the need for resources not in your control.  That’s where automation comes in. 

Data Questions are the beginning.  Automation is the middle.  How you end the story.. that is up to you!