Functional Specs

The guys over at 37signals are spot on again; Get­ting Real, Step 1: No Func­tional Spec (Sig­nal vs. Noise).

The “specs are really hard to make until you’re sort of halfway through the project” is a clas­sic, and while the ‘just don’t do it’-idea sounds attrac­tive, I’ve often expe­ri­enced (as men­tioned by oth­ers in the com­ments) that clients often wants some idea of scope (and espe­cially price) pretty early on in the pro­ces. So far I’ve man­aged to basi­cally just drag the spec-writing by div­ing straight in to build­ing inter­faces and func­tion­al­ity that I know is going to be there any­way, which usu­ally ends up being a far bet­ter way to start dis­cussing the final details. On the other hand, not all clients allow this to hap­pen. Either because they’re trigger-happy, basi­cally not smart enough to be in the software-buying-business or what­not. And while I usu­ally have a pretty good idea of what not to accept with regards to the final pric­ing, I kind of like get­ting some­thing down on paper so I don’t end up in the “well, we thought this was included in our agreement”-problem. Most clients accept that at some point, there’s sim­ply not more money — but again, this is quite nice to have on paper…

Part of the prob­lem defi­nately has to do with the fact, that I’m a small player, work­ing with clients more con­cerned with price and launch than a good, con­struc­tive process — and/or work­ing on projects where the actual spec work takes up a sub­stan­tial part of the total amount of work because of the project size. A good dia­logue goes a long way, but what to do when you can’t aim for the ideal solution?

I really like the idea of ‘mov­ing tar­gets’ as men­tioned in the com­ments; it’s the agile approach and often what every project ends up being any­way. Only prob­lem is, how do you struc­ture that process so every­one is happy and under­stands what’ going on?

I’d like to write more about work-realted issues as a free agent type of busi­ness; this is what this com­mu­ni­ca­tion thing should be all about. And who knows, maybe some day I’ll even sug­gest solu­tions instead of just ask­ing ques­tions. What do you say, how about mak­ing it a bit more per­sonal, share some of your expe­ri­ences, see if we can become bet­ter at what we do by, well, doing some of tthe other stuff we do?

Also, White­space men­tions another clas­sic; free­lancers tend not to spend any real amount of time on mar­ket­ing. Which, when I come to think about it, is stu­pid. And bor­ing. And I don’t like doing it. But I like when I have nice projects. Go figure…