Wasting Money on Content in Visual Dead Zones
E*Trade
Visual dead zones reduced E*Trade's ability to communicate with their target customers, because low percentages of people read their promotional content.
This is an old story*, but it is an Eyetools favorites because it illustrates an important point:
We eyetracked E*Trade's homepage and then inserted gibberish into it to illustrate that content in a "visual dead zone" doesn't get read and might as well not exist.
To identify parts of E*Trade's homepage that weren't being viewed, we collected eyetrack data as people surfed E*Trade's website to decide for themselves whether they wanted to sign up for E*Trade's services. We found that about 1/5 of the screen real estate above the fold was being wasted, so once we located un-viewed areas, we came up with gibberish to replace the existing text on a cached version of their homepage (this was a lot of fun!).
Examples of the gibberish tested on E*Trade's homepage:
- FDIC distrusts us * No Bank Quality * Will Lose Value
- Not ready to event an insurance? Tax group of our manager discussion free of funds.
- Get $25 to close an E*Trade Bank Money Market Plus Advice! Tax a gear cool and ATM access!
- ...and more
Then, we re-tested the modified homepage in the lab. As a secondary test, we also sent it out to a large group of people who we didn't eyetrack. After people had seen the page, while evaluating whether they were interested in signing up for E*Trade's services, we asked them if there was anything strange about the homepage. Only 1 in 25 people noticed!
The moral of the story is this, if you wish to optimize your content:
- People can't click on, or buy, what they don't see.
- If content on your webpage is in a visual dead zone, it might as well not exist.
- Changes made to text in visual dead zones will have little impact.
- If the copy is critical, then visual dead zones need to be fixed before copy changes are considered.
- If the copy is non-critical, simply removing the copy is often (though not always) a solution.
* This is an old E*Trade website -- we ran this test in January 2001. Since then, E*Trade has become a client and changed their website.