I've had a few conversations over the last couple of weeks about whether or
not UCD (or what ever you like to call it) adds cost to a project -
therefore making us less competitive in the estimation that we give to a
The argument is with technocrats who believe that on a five day project,
four days would be development (i.e. bashing out the features as best as
they can) and the last day for testing. In my experience this is where the
proliferation of poor practices start.
I argue that with one day's design up front, sat with the user, iterating
through the few possibilities, a model could be built (lets say a form on a
web page) to communicate what needs to be done to programmers, where the
major design decisions have been distilled from the logic and a process of
'hooking up' is required. That means IMO that four days development would
turn into one day of design and three days development, therefore adding no
extra cost to the project.
There is no question however of the value of UCD in a large project in my
company, where best practice is perceived to add more value to the product
than using best practice on a small five day-er. I beg to differ. I believe
it's a case of all or nothing.
SO! Is UCD scalable enough at a small project level to integrate it into all
processes? Well I believe that it is, but it requires responsible selling
and that's perceived as a problem to those who sell.