There have been a few discussions recently on the ROI issue on a number of fora.
Sadly a few have generated more heat than light so I decided not to join in, but I did want to make a couple of points...which is why I have my own blog!
As a management discipline, communications should always act with data. I hope the days are long gone when the gut feel from hoary old hacks was considered to be a useful guide to action or an excuse for evaluation.
If you accept the need for data you then hit another problem.
What is an acceptable level of data for communications?
This is a problem which has exercised better minds than mine and I would praise the work of Angela Sinikas for anyone who wants to explore the subject further. And some of the work of Kathryn Yates at Towers Watson demonstrates the potential value of research into this area.
The central question is how to evaluate the impact of communications inputs in bringing about change. When behaviours at work are the product of many different forces, how do you reach a reliable or unequivocal assessment of the value of communications.
Angela has produced some very interesting case studies. She does tend towards accentuating tracking very specific behaviours - because you can isolate where behavuoural change has happened and hopefully isolate a limited number of potential drivers. The Towers Watson models operate on a more grand scale and are equally interesting.
Now, the methodologies of all practitioners are open to question and should be challenged. Only through challenge and debate do we move forward.
But because I might disagree (or in my case simply not understand) with the methodologies, do I think the endeavor is worthless?
Of course not.
It is my personal experience that the most effective communicators come to conversations with data. The stars of our profession are excited by what they can learn from the data that is available (check out Charlie Nordblum's Communicative Leadership group on Linked-in to hear what they are talking about).
It is therefore my prejudice that the mark of a potential leader in our profession is an interest in data and a willingness to explore issues like ROI. I don't know the answers but I do believe that they are worth worrying about.