Jason Fried’s mantra while testing was: We need to test radically different things. We don’t know what works. Destroy all assumptions. We need to find what works and keep iterating—keep learning. (I’m paraphrasing here…) We tried out a radically different design with these results:

The Person Page was far shorter. There was less information about Highrise. However it had a 47% percent increase in paid signups than the Long Form design.

This insightful post from 37 Signals shows the evolution of A/B testing pages for one of their landmark products, Highrise (a customer relation management system). It shows how good design and clear objectives can see significant improvements to calls to action.