Bear with me for a bit. Here’s something to try at home:
Hold your hands as far apart as you with arms outstretched. That represents the scope of a typical project the day it kicks off.
Now bring your hands together so that they’re a little less than shoulder width apart. That’s about the scope you’ll be left with once you’ve gone through requirements review processes, procurement elimination, cost/benefit analysis, etc.
Now clap your hands together and leave them an inch apart. That’s the requirements that your user base will ultimately desperately need every waking day whilst they are working on the system that you have designed, procured, built, tested and delivered over the last however many months. The noise you heard was the big bang of implementation!
Now, if you could get from furthest apart to closest together in a dramatically faster time, would you be prepared to sacrifice the loss of functionality in return for the faster return on investment, the more rapid delivery of business and customer benefit and the overall reduction in risk? Then you could start work right away on delivering the other functions that were less useful.
Be honest. What would it be worth to you, your business, your customer base to deliver the right functions ahead of time? Would it be worth enough for you to take what someone else had already done and exploit? After all, how many of you have written your own word processor recently? How do we get other key applications recognised as the equivalent of “word processors” – i.e. things that you get someone else to build for you and then you deploy rather than things you have to define, procure, build, test, deploy a la waterfall model?