If you want to know when new interviews go online,
subscribe to the WebWord.com Usability Newsletter!

Usability at Women.com - Part II

A follow up interview with Bill Skeet, Director of Network Architecture at Women.com 

Conducted via email by John S. Rhodes (28-Jan-2001)

(Read Part I...)

What kinds of usability tests do you do? What were the results? What did you find?

We've practiced heuristic evaluations and have hired firms to recruit and perform one-on-one usability. We've also hired a firm to perform eye-tracking in conjunction with traditional usability. I could tell you the results, and what we've learned, but I'd have to shoot you! HA! Actually, what we've learned ranges from a lot of minutia that wouldn't interest many other folks to some tidbits that are truly original and valuable enough that other companies could use it.

OK... One tidbit: A recent study has affirmed an old mantra of mine: Every page is a front or "home" page. Probably nothing new to many, but the saying was born from my frustration about how much time the organization spent obsessing over details on the home page while the design of the pages deep in the site (article pages) were ignored. Meanwhile, to the users, these pages deep in the site were the goal or destination. Many people actually entered through a link sent to them from a friend or a search engine or some other "backdoor". To these people, the article literally *was* their "home page." It was burdened (or not) with representing the brand and the breadth and depth of the site. The content page must attempt to offer more information that would encourage this person to stay a little longer and check out the site. These pages were every bit as important as the root or "home" page of the site.

How do you measure the ROI on the usability testing you are doing? 

We've been fortunate thus far to be able to point to traffic increases... The correlation between traffic and revenue is pretty well accepted in our company, so I haven't had to sell it hard. We've also been applying usability on design projects that were already approved and we've simply made the case that a round of testing will help catch some mistakes before they go live... and we've done the usability fairly inexpensively. So it's remained under the radar thus far... I'm trying to keep the organization positive by publicizing results such as a 25% increase in traffic for one of our sites within 24 hours of a redesign. And we're talking about an additional 25% on a site with thousands of page views a day.

Are you doing to be doing more or less, and why? 

We're doing about the same. Right now, we are working on some large-scale projects that touch broad parts of the company... essentially superseding the need to test other parts of the site.

What are some of the areas of the site that have received a workover as a result of usability data?

The home page of Women.com and Astronet. And I expect there's more to come.

Do you have any interesting case studies? Horror stories? Pitfalls to avoid? 

OMG, yes. It would take some time to write the stories and change the names to protect the innocent and guilty (and myself!)

Here's a quick summary without details:

• I feel I must actively elicit the flaw or issue from parts of our organization who come to me with solutions. I urge folks to explain and fully understand what is broken (and what isn't) before fixing things. (It's surprisingly difficult for some to re-trace the tracks that led them to a solution that included flash and a set of new templates.) We've used user-testing to help discover and define issues.
  
  
• Understanding the relationship between form and function is a big struggle. I don't think I've figured this one out yet... Generally, people either a) see them as independent and try to solve them with separate teams, b) address the aesthetics before the function or c) mistake form as function-- and only address the look and feel.
  
  
• Many people don't *really* know what usability testing is... so there's need to re-educate people before providing results. I've found that people often mistake it for other types of research, such as focus groups. They don't understand how and why one research method can deliver a certain type of information and what it *cannot* deliver.

What next?

If you want to know when new interviews go online,
subscribe to the WebWord.com Usability Newsletter!