You can prototype a website until you’re blue in the face, but sometimes you should trust your gut when determining what users want. That’s something Will Marks, VP of business development at E! Entertainment Television, learned after going through a ‘very expensive, very time-consuming, very detailed’ redesign of E! Online a year-and-a-half ago. E! worked with a Web consulting shop to do user eyeball tracking and prototyping, but when the website relaunched, ‘people didn’t love the site,’ says Marks. ‘They didn’t love the new front door, didn’t love that the site was too slow to load, couldn’t find their favorite columns…’
A subsequent redesign of E! Online a few months later used a more intuitive approach to improving the site for users. This redesign, says Marks, was more internally developed, and played to E!’s strengths. The site got a new front door, an increased focus on blogs, more frequent updates and more user interactivity. ‘Even just changing the colors, fonts and the layout of the front page in a much less prototype, less user-tested, less rigorous way – the site’s actually been much more popular,’ says Marks. ‘Sometimes I think you can get too caught up in doing too many betas, too many prototypes, having too many focus groups and you don’t go with what your gut is telling you.’
Still, intuition can’t fully replace hard facts. Marks says his team evaluates the site’s logs and user reports daily to see how much time people spend watching clips, and how the content mix should be tailored to suit users’ tastes. ‘We see what’s working right away, and if stuff isn’t hitting, it’s gone,’ says Marks.
In the last 18 months, he says E! has added a lot of content across all of its Web properties, including video, user commentary and more blogs. One of the blogs is ‘Watch with Kristen,’ which receives regular feedback from users wanting to learn more about shows Kristen mentions. ‘It’s amazing to see how this interactive world between the columnist and the user has shaped what we cover and the frequency of what we cover,’ says Marks.
There are also viral signals that show what’s popular on the site. With these, says Marks, ‘You don’t have to wait two weeks for a rating to come out; you’re not going through some panel that’s run by Nielsen or anybody else. You know who’s emailing this clip around, who’s taking the time to give this clip a rating, who’s taking the time to move it to the top of their most-played list – those are things you see immediately.’ In the fickle world of Web surfers, every second counts.