Rama Ramakrishnan

I have data. I need insights. Where do I start?

This question comes up often.

George,

Have you seen this role in action often? If so, have you found it to be effective in addressing the question you lay out in the title of your article?

That’s an interesting question. While I haven’t seen someone formally play this role, as I think back to successful projects, the data…

The approach I proposed isn’t in lieu of asking the business what’s important to them. As I say at the outset, if the business points you in the right direction and gives you a clear problem definition, that’s data science heaven.

But in practice, this frequently doesn’t happen and there’s often an impasse between data…

Mike,

… but isn’t the maturity of data science to understand many such “hmms” and react in quasi automated mode?

In my experience, we haven’t made much progress in automating this. This stands in sharp contrast to our deep-learning-fueled success at certain “intellectual” tasks like image…