I think the old Cranfield Application Portfolio Grid is a way
of classifying applications (and the strategies for developing and
managing them) to address this problem.
you're not familiar with it, it is basically the Boston Square oriented around
IT. It maps applications on to a 2x2 according to rate of change and
operational criticality. The Quality, Cost and Timetomarket values vary across
the quadrants. I've listed them in priority order here:
rate of change, low criticality - HIGH POTENTIAL (Time, Cost,
high - STRATEGIC (Time, Quality, Cost)
High - KEY OPERATIONAL (Quality, Cost,Time)
Low - SUPPORT (Cost, Quality, Time)
all written up in John Ward's book on Strategic Information Systems
wrote a customised version of it up in the BA Business Systems Planning guide,
published in October 1993. (We had it externally reviewed by Hoskyns, actually -
chap called Robert Mann), with recommendations for different project approaches
in terms of justification, methodology and resourcing. I think most of the
advice still stands, except that systems are more interdependent today, so even
support systems could be quite critical.
Lambert from Cranfield is going to be doing a talk for us at the
EAC Europe 2009, and he did mention the grid, so it's possible it may feature in
his presentation. I will encourage him to do this, as I've always found it a
very useful model.