For a very long time, business analysts and system developers have studied the concept of Cost of change.
The cost of correcting a defect increases almost at a factor of 10 as the defect skips one stage to another, including Requirements to Design, Design to Coding, Coding to Testing, and Testing to Deployment.
This has prompted a behavior where organizations try to 100% correct up-front. This means having extensive processes, exhaustive templates, and review processes to capture requirements. However, the cost of change ignores an important aspect: The Cost of Delay. All of us recognize that business opportunities typically follow an inverted U form. They are low initially; the opportunity can grow, mature, and finally decline. The organizations which spend significant time in the up-front analysis will generally tend to lose out on taking advantage of the growth phase of the opportunity. Again the opportunities curve for an internal application can be substantially different from applications (products) developed for the market.
Wherever possible, let's try to estimate both the cost of change and the cost of delay before we decide on investing and developing new applications and products.
I will summarize my thoughts with what Reid Hoffman, founder of LinkedIn, said, "If you aren't embarrassed by the first version of your product, you shipped too late."