My general strategy is to write the stories, estimate them, and add them up.
What happens next depends on what the estimate will be used for. If I am
making a fixed-price, fixed-scope bid I would like to have at least a little
of the system implemented before making a bid. If the date is fixed, there
is likely to be a fair amount of negotiation over the stories (splitting
stories, picking cheaper infrastructure to reduce estimates, picking simpler
stories). In any case, I would include slack in the schedule so I could make
honestly confident estimates.

Kent Beck
Three Rivers Institute

> -----Original Message-----
> From: extremeprogramming@yahoogroups.com
> [mailto:extremeprogramming@yahoogroups.com] On Behalf Of
> Jonathan Rasmusson
> Sent: Sunday, April 17, 2005 2:56 PM
> To: extremeprogramming@yahoogroups.com
> Subject: [XP] Sizing projects from a budget point-of-view
>
> Hi all,
>
> How does XP recommmend sizing projects from a budgeting point-of-view?
>
> After coming up with high level stories for the project (say
> 100 days),
> do you then take that number, estimate a team velocity, pick a date,
> and hence the budget?
>
> i.e. I am the customer, I am asking you the XP team for how much this
> project is going to cost. Is the number 100 days? Ok thats
> I will get
> the budget setup for 100 days.
>
> Or do you add in some contingency to the number?
> For example add 10% for this, 20% for that, and call the date 130 days
> (ala more traditional forms of project estimation).