How do you justify spending more time on design?

A very thought-provoking article from Fast Company about determining the ROI on design. Recently, we revamped our website at TheGoodBlogs. We've been through several iterations in our short life-time to try and keep the site fresh. However, we would struggle to answer if the effort involved (not insignificant) was really justified. Although subjectively, we think it did. To obtain a direct correlation is extremely difficult because the experiment is not isolated or repeatable with the same set of subjects. i.e. since our membership numbers are climbing month over month, how do we attribute the increase to a better interface? even if there is an increase in the growth rate, there is not telling whether that was a direct result of our design efforts. Ideally, you would run parallel sites over the same period of time with the same people visiting both. Unless you have deep pockets or a research organization like Nielsen, using usability labs or conducting real-world surveys is usually out of reach.

There was a conjecture this morning in the news where they attributed lower American Idol viewership to the changes in daylight savings time. i.e. more people spending time outdoors earlier in the season. We all know we can spin statistics to tell whatever story we want. But quantifying a lot of what we do is certainly hard especially in a startup where all the variables tend to change quickly and by huge margins.