[thelist] Project Price

Martin Burns martin at easyweb.co.uk
Thu Dec 6 18:58:57 CST 2007

On 3 Dec 2007, at 16:43, Matt Warden wrote:

> On 12/3/07, Portman <mrport at sbcglobal.net> wrote:
>> BTW, I often tell people that if they think that my price is too  
>> high, I
>> will work hourly for them. It ALWAYS comes out more....
> Really? I think you may want to reconsider how you calculate fixed
> price work, then. The client wants fixed price work because they will
> feel more comfortable knowing what they will pay. You are taking on
> additional risk, because your estimates must account for many
> unknowns, many of which are under the control of the client, and
> others (scope) which require a lot of effort to manage such that your
> risk does not increase.
> My experience is that estimates for fixed price work are *heavily*
> padded, and the client knows that. But often they're willing to pay
> that premium for the knowledge of exactly what they are getting and
> exactly what it will cost.

Couldn't have put it better myself - clients pay a premium for you  
taking on all the risk of overrun.

So to do fixed price work, you need to be absolutely rock solid on  
scope, and have a *completely* strict change control process in place  
to keep your risk within sane levels. Now, if the client isn't  
disciplined about what they want, this can get really expensive for  
them, really quickly, and it then has the worst of both worlds: an  
escalating cost at premium pricing.

If you're being asked for a fixed price deal on anything even  
moderately complex, then it's usually A Good Idea to attack it in  
stages; the (fixed) scope and fixed price for each phase is part of  
the deliverables of the phase before.

One way of attacking a T&M deal is with an upper budgetary figure,  
which you control by either:
1) Agreeing in your SoW that your responsibility is discharged once  
all the money's gone (easier when you're not actually accountable for  
the end deliverable; you're merely providing assistance services)
2) Keeping the client very close to the ongoing spend & forecast to  
complete, and at all times presenting "OK, if you want *this* thing,  
then what else do you drop?" That may not be pure functionality by the  
way; it might be a desired launch date, or desired quality (say by  
reducing testing).


 > Spammers: Send me email -> yumyum at easyweb.co.uk to train my filter
 > http://dspam.nuclearelephant.com/

More information about the thelist mailing list