Estimating is more an art than a science, especially for software projects.
There is plenty of literature about estimation but is not easy to find the right numbers.
Why these man /day are so hard to be decided?
First of all is the customer that:
1)Doesn’t know what he needs or what he can have with his budget
2)Have, at least, a romantic idea on what can been done.
3)Would like to have something fast and cheap and perfectly running from day one.
Then there are the software developers, that can be differentiated in three sets:
1)The monkeys in front of a keyboard.
2)The missing link between God and humans.
3)The I’m doing my part, I don’t care for others.
The phases of a project kick off, are to be short, 3:
1)Inception.
2)Offer document, to the customer.
3)Technical document to the developers.
After writing a technical specification document I’ll share it with the Development guys.
As project manager you cannot choose, usually, what kind of people you will got in your team, so how you can estimate?
In my case I use the simile estimating technique, in how much time a normal it programmer could do it.
The trick is finding epics and then user stories that can be easily calculated.
You can use the “self-fulfilling estimate epic” method, in three easy steps:
- Take two or more epics.
- Sum the estimate that the developers say they need.
- Add 10% to the estimate and you have a “buffer”.
Using this technique, you can have rough numbers on project cost, the cost is feasible?
If not, start some haggling with customer and developers… to be continued.