In my current assignment I was recently in a meeting where we were discussing a new technology that I have some previous experience with. The purpose of the meeting was for the project to understand how we would go about learning more about this technology that could possibly be used by us.
I don’t want to bore you with the specifics of the technology, because that is not really the point of this post, but let’s just say that it is a standard for exchanging messages about financial transactions. The world of financial trading is complex, multi-faceted and multi-dimensional. On the one hand you could talk about separate classes of financial instruments (equity, bonds, fx etc), on the other hand you can talk about separate classes of transactions (spot, option, future etc) and on the third hand you could also talk about separate types of information being conveyed by these messages (transactions, positions, events etc). I have found that navigating such a multi-dimensional world in various discussions is hard and on many occasions I have been in meetings where people did not understand each other because of confusions on where they were in this multi-dimensional world. In fact, I once wrote an article for The Rational Edge on just that topic but for system models, let me see if I can dig it up again and post here….
Back to my meeting…
Early on we were discussing what asset classes that the technology in question supported. The people in the meeting were mainly project managers and requirements people (and me…) The project managers being who they are quickly made a list of asset classes and gave the task to the requirements people to investigate what parts of the list were supported and what parts were not. Now I felt I had to say something… My understanding of the technology in question is that it mainly deals with transactions and events and not in positions. For our purposes we were interested in positions. This is a separate dimension so it doesn’t matter if the technology supports messages abouttransactions of various asset classes, since we were interested in the positions of those asset classes. But somehow I was not really able to get this point across to the other people in the meeting.
Maybe I did a poor job of communicating, but I don’t think that is the whole story. By making that first (one-dimensional) list, the project manager inadvertently locked peoples thinking into that the list was the only important thing. If we could just cover the list, everything would be fine. Furthermore, the list was easy to check off in the documentation of the technology so the job at hand would be easy to do… My objection forced people to wake up from this illusion which is always hard to do and therefore usually meets some resistance.
My experience is that many project managers are so focused on creating a (one-dimensional) list of things to do and then to check off that list that it is hard to for somebody else to point out that the real world usually does not lend itself well to be described by one-dimensional lists.
<Update 140401: I have found the article. You can find it here>