As requests go “How long will it take?” is one of the tough ones. Taleb’s book Black Swan makes an thought provoking point about it. Like most of that book the underlying question is what statistical distribution the data is drawn. Naturally, that should color our expectations.
Some project durations are reasonable, Gaussian. How long you will live, for example. As the project unfolds the expected end time draws progressively closer. Delay in these projects push out the end date; yes but, you do get always closer to the finish line. Such projects can be standardized.
But many projects are unreasonable, their duration is drawn from a highly skewed distribution. Chock-a-block with extreme cases and little black swans. In this case as the project unfolds it’s end date moves further out. Each delay increases the expected time to complete the project. Taleb’s example is the refugee who imagines that each passing day brings closer the day he will return home, but as that is likely the this second kind of project, these passing days infact push out that day.
You manage these two kinds of project in very different ways. Most projects are a hybrid. The nature of hope is very different for one v.s. the other.
See also: Time to Market