Originally posted by where did my id go?
The biggest problem with fixed time & cost is the difference between what a customer wants prior to commencement of the project, and what they realise they actually want later in the cycle. If it is hard for us as software professionals to get our heads around the complexity of a solution - just imagine how difficult it is for the average business person!
Small businesses don't have money to throw around - they therefore want to be certain that they can accomplish business objective X with exactly Y amount of cash. If you were to be honest up-front about the possibility of extra expenditure resulting from changes later in the cycle, they view this as you "trying it on" and will simply go to another supplier who will not be quite so honest.
The only way I've ever got around this is to allocate a change budget and say something to the client like "The work will cost 10k, but in the event of our understanding of the requirements changing, you may need to be prepared to spend up to 12.5k". Generally speaking, they're ok with this - the potential cost of delivery is variable, but at least it's within a commonly understood range. In theory at least, you can keep within this range by managing/prioritising changes during the course of the project and learning to say "no" to the client when there's a good justification for doing so.
The size of the change budget is derived from the perceived risk of change - this can generally be derived by looking in detail at the personality of the client and how much you're able to hammer down the spec before development work begins.


If you're a developer, then do you have a specification of a project, or part of a project, and an end date written into the contract? Surely you couldn't have a detailed spec, that would require too much work in forming the contract, and the client might not want to divulge that to a third party (i.e. an agent).
Leave a comment: