You know those surveys that happen every year? Those ones that talk about what l&d should be doing and they are not or what the top l&d teams do that makes them stand out? Apart from the fact that they make me cringe with envy or I feel inadequate, there is one thing common to these surveys, they alway point out the fact that hardly any l&d team is doing proper evaluation. Forgive me if your team does do proper evaluation but you really are an outlier or an anomaly. And then we have those arguments around how smile or happy sheets (whatever you call them) are a waste of time or Kirkpatrick being outdated or not usable or even the whole ROI being too complex and how ROE is the proper approach. The truth is, no matter what approach you pick when it comes to evaluation, one thing is common to them all, they all need data to work. If you don't collect good data, you can't evaluate. It doesn't matter whether you use the best evaluation framework in the world, effective evalaution still comes down to collecting the right kind and quality of data. Therefore I submit to you that evaluation is not a frameowrk problem, it is a data problem.

So what is our data solution then? Simple, we need to become adept at data analytics which is the process of defining a question (in this case the evalaution question), identifying and collecting the relevant data, cleaning the data to make it usable, analysing the data to derive insights and then presenting the insights in the right way, to the right person and at the right time. This begs a question - does l&d need data science?

I will try and answer that question in my next post.