First, I'm really intrigued by the design choices made on the page - the red line and the fact that the page is a good 1280 px wide (I have a 1280x800 pixel mon

First, I'm really intrigued by the design choices made on the page - the red line and the fact that the page is a good 1280 px wide (I have a 1280x800 pixel monitor and had a horizontal scrollbar). There's something absolutely brilliant about how the red line on the left reinforces the key insight of the research. (And it exactly matches up with the part of the text that would maximize that reinforcement.) The page width, I don't get - but Dustin's a design stud and I'm not a stud of any sort. Second, I suppose we could end up at optimal designs for what we wanted through iterative A/B testing. Given enough time, enough variations, and enough traffic, you could conceivably come to the ideal webpage for conversion (of whatever metric you wanted). Of course, as people change, the page would have to change as well. The Gladwell article posted earlier today is a good companion read to this piece, I thought.