Megan Leigh McDonald
Get Megan's Resume
‘Don’t make me think.’ Steve Krug

Megan Leigh McDonald

A Limitation of Click Map

Today I learned something from my Product Manager. He had the look of a ‘bearer of bad news’ but he gave me the medicine anyways. I have heard in the past that ClickMap is not always reliable and noticed even that the data didn’t always match the next page reports I pulled.

But today, I understood at a new level of how this tool, as great as it seems to a designer, is sometimes wildly innaccurate. As you might have read on this blog before, I love this tool mainly because of its visual appeal and in some cases, when the link leads to a popup or a page that doesn’t have omniture tracking code, it can be really helpful.

However, when using templates, one must beware. It’s probably also a factor of not being coded correctly or with great care, but templates can give some very misleading numbers that represent the total clicks of all the links in a defined module. The module, such as a sidebar or navigation menu, might have more than one link but the flag will appear on only one of the links, leading one to misinterpret the data. Sigh. Well I hope that Omniture really looks at ways to improve the implementation of code for this tool or the tool itself because the visualization of clicks on live site pages seems like a really great idea.

About Face

In re-reading this book, I have re-discovered valuable excerpts and insights into what I do everyday. There are a few quotes I want to blog about. Nothing fancy today, just a few somewhat random thoughts about User Experience.

“…good design makes users more effective”

Truly, this is the point of good design, although the term ‘design’ is such a broad-based, multi-interpreted word. In the ‘real world’ the title ‘designer’ is often used as a ‘catch-all’ for someone who can do graphics, maybe produce a website and possibly provide some basic usability. But looking more closely at this word, it probably should be more narrowly defined. I read very quickly a prediction (on Adaptive Path or some random Tech article) that the job descriptions calling for a ‘do-it-all’ kind of person would be supplanted by very specific, more narrowly defined descriptions.

Consequently, the kind of designer that I naturally became and am still becoming is a more user-centric one. I studied media user behavior in Graduate School and appreciate a more academic analysis of website usage. I remember back in the day with WebTrends, how the data was focused on uniques and visits and clicks and various user statistics like browser type. But now, using Omniture’s SiteCatalyst I’m completely blown away and (dare I say?) excited. It’s the kind of thing that gets me up and moving in the mornings, for lack of a caffeinated beverage.

Something about examining the numbers is kind of like being a detective. I get to root out where the problems are, where people have trouble using a piece of software or web application. I can’t help it, I did love Nancy Drew and Trixie Belden. So this aspect of my job is the one part that gives me the greatest satisfaction. I get to improve the experience for the user. I get to make the user more effective at what he/she does.

There is a huge difference between teaching the user to use the system and designing a system that works for the users because ‘good design makes user more effective’.

Lastminute.com’s 700% Click-thru Increase

I was just reading over a PDF that I downloaded from Omniture. It’s a case study for this travel company lastminute.com which I follow a bit in the news. The two guys who started that company have also started one of my favorite travel social networking companies — www.wayn.com - Where Are You Now?

So this PDF, available for download off of the Omniture website, lays out the business problems that lastminute.com was experiencing. Their problems were overarching big picture problems like trying to figure out what the steps were that customers took between typing in keywords and booking trips, not to mention the customers that dropped off in between these steps.

They used a term I’m not familiar with — ‘look to book’ — I like that. It really describes the kind of user role that would be a ‘browser’ essentially. This really brings me back to graduate school and the persuasive theories in my ‘User Experience Communication’ course. I used this one particular bell curve in a paper I published about CyberHate websites. Basically on one end of the bell curve you have a certain percentage of people who are never going to subscribe to ‘hate sites’ or buy your product. Then on the other side is, for the sake of this example, an equitable percentage of people who will absolutely buy your product. That being said, it’s the large grey area in between those two groups that comprise the biggest opportunities in terms of convincing a person to convert from a ‘maybe’ to a purchaser.

I digress though.

To get back to lastminute.com’s case study, they did an A/B test through SiteCatalyst and discovered that the horizontal menu bar was more successful than a vertical one. After implementing the horizontal navigation, they increased their click through rate from the top navigation menu by 700% which is so insanely successful I can’t imagine what the UE and PM teams must have felt when they crunched the numbers and saw that kind of jump. It must have been an ‘awesome’ moment of triumph.

The other interesting angle to this is the SEM side of things. They were able to analyze how particular keywords performed by tracking the drop off rates on landing pages. When they identified high fallout rates, they looked at the keywords and the content of the page to determine how well they matched. In many cases, they weren’t EXACTLY what the customer was looking for and they were able to improve the experience and reduce the fallout rates.

It definitely makes me think about how to relate this to work. I already have an idea and can’t wait to whip it off to the appropriate person(s).

Omniture ClickMap Tool

I have to say that I really love this tool in Omniture.