This nice blogger, Victoria Morehead of the optimization outfit Brooks Bell) designed an A/B test on their blog for a Cisco.com page. Their suggestion was to test the effectiveness of video ("A") vs no video ("B") vs segmented videos: http://www.brooksbell.com/blog/2012/03/id-test-that-determining-the-conversion-power-of-video-with-a-split-test
What we know about video already:
- Visitors who view videos stay longer, go deeper into the site, and return more frequently
- People who watch a video are in fact twice as likely to return
- People who watch a video are more likely to complete a "success event" such as a chat or downloading relevant information
- We use video interaction to help personalize some elements and behavior on your site journey
The blog points out some places where a good A/B test could reveal even more good data. You can bet we're talking about running a test like this.