User Interface Engineering

eBay shoppers are a peculiar lot. They just ain’t normal, statistically speaking.

When we conduct a study, we ask our participants a ton of questions. Some questions prove to be completely unrelated to anything we learn in the study. But other questions, well, they provide some interesting insights.

We’ve always asked questions relative to other web experience. We started with the usual gamut: How many hours a week do you use the web? How many years have you been using it? What sites do you visit? Do you ever shop online?

The answers to these weren’t very interesting. Our participants’ answers never seemed to predict any of the results from the testing. In other words, it just didn’t matter what they said. So, we kept changing our questions, hoping to find something that could actually predict a test outcome. And we found something, in the weirdest place.

In a study we were doing with online shoppers, we asked a simple question: What was your largest purchase on eBay? Because most people have never shopped on eBay, they typically answer $0. But, in our study, some of our users put in a value, ranging from $15 to $2,500.

So, we dutifully entered this data into our modelling tools, along with the 300+ other variables we collected from each participant and we set our analyzer off and running. And what we learned surprised us:

• eBay shoppers are more likely to read policies (privacy policy, return policy)
• eBay shoppers are more critical of the site. (They rate various subjective measures lower.)
• eBay shoppers find shopping online more frustrating
• eBay shoppers are more likely to feel the shopping process takes too long
• eBay shoppers tend to find the site content more confusing
• eBay shoppers tend to dislike the layout and appearance of the site
• eBay shoppers tend to feel the images are not helpful
• eBay shoppers tend to feel the site wasn’t fun
• eBay shoppers tend to feel they won’t return to the site to shop again

Now, we asked the same question of Amazon shoppers, Wal-Mart shoppers, and about 20 other different offline and online shopping experiences. None of them showed the same kind of bias as eBay shopping.

We call this attitude bias eBayness. The more you spend on eBay, the more eBayness you exhibit. And the more eBayness you exhibit, the more attitude you show about the shopping process.

(Here’s an interesting by-product of our analysis of eBayness: it doesn’t seem to effect perceptions of price. Folks who have spent more on eBay do not think any differently about a given retail site’s pricing as anyone else. So, they aren’t any more likely to be bargain hunters than anyone else.)

Now, what can we do with eBayness?

Well, when testing e-commerce sites, we can start to use eBayness as a way to refine our testing. When testing a new design approach, we could focus on users who have a strong eBayness to whittle away at the coarse edges of our design. Once we make it acceptable in those users eyes, we can then move on to fine-grain refinements with users who have little-to-no eBayness.

Controlling for eBayness in our testing now becomes a priority. And it begs the question: what other user groups display particular attitude bias? How are we controlling for them?

This entry was posted on Friday, August 12th, 2005 at 1:37 pm and is filed under Usability Toolbox, Users . You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.