Let's get honest: it routinely feels like a shot in the dark.
Here's the scenario:
Someone in management is legitimately trying to organize resources and push some projects forward. He or she may be pitching the project to an executive or at least someone who has the red-light/green-light control switch.
So, you are being asked, "How long will it take? How many people do you need?"
These are not bad questions, and it is reasonable for management to expect answers. The biggest problem is that estimates are hard to produce. Why?
Well, for one thing- it's sometimes hard to even determine how long a project took after the project is over! It is much, much harder to predict how long it will take for the project to get done before it begins.
Or, is that really what estimating is all about?
One thing is sure: under normal circumstances, the more real information known about the project scope, dependencies, resources, and risks - the better estimates become.
One of my colleagues here at work, Ade Miller, is good at calculating in my view: he climbs mountains. Miscalculating can be really costly when you are hanging off a chunk of ice thousands of feet up the skyline.
Here are a couple of books he recommended to me:
Agile Estimation and Planning – Mike Cohn
Software Estimation: Demystifying the Black Art - Steve McConnell
Rock Thought for the Day: Silversun Pickups:
- I'm still waiting, rather impatiently, to hear the new Smashing Pumpkins album.
- We need to end the euphoria about My Chemical Romance, the Killers, and bands like Fall Out Boy. They're not awful; they're just not great. Let's raise the bar a little higher.
- I like this blog entry about rock rivalries. Entertaining, and a sign that we have too much time on our hands.