Joh and I held the fort for the Brighton Agile Forum last night by discussing a number of Agile issues over beers at The Eagle. One thing we discussed was what the ultimate targets of an Agile Manager are. Targets being the slightly abstract concepts that lead you to achieve the end goals.
Joh suggested that it was to create a process that will work without you, i.e. to make yourself redundant. Easy enough to test: when you go on holiday does everything crumble? or do you have to put in double hours the week before making sure that everything is all set up?
This reminded me of being a DBA or a Linux Admin. A test of whether you are doing it right is that people start to believe that you don't do anything, as you have set up the systems so smoothly that they either never go wrong, or if they do then it is easily corrected.
Our current process is simple but effective.
I started thinking about what it is that I do that isn't part of our current devlopment process. At the moment I believe that we have implemented a process that is quite efficient but doesn't see 'The Big Picture'. I think that the role that I provide is that of an Analyst. I sit between the development process and the business and decide what we should develop in order to enhance their work. This is a creative activity that involves experience and imagination. Once it has been decided 'what' we want to achieve, then the development process kicks in and actually produces it. However, the isn't all decided up front (as a BDUF - Big Design Up Front) but may appear as the development continues. So, how do we adjust the development process to lead the developers towards insightful decisions about the direction of the project as a whole?