In this chapter I’ll discuss usability testing and research and it’s role in User Experience Design. It dovetails nicely with last week’s post in which I discussed the design portion of the UXD profession. The post seemed to strike a nerve with one of the respondents in which he argued that User Experience Designers and Designers can be the same person or role. Although that was not the gist of the post the challenge with the Designer also being the UXD professional on the same project comes down to an inherent conflict of interest. Why you might ask? It’s the same reason why a developer can’t sign off on final quality assurance testing- a separate resource needs to provide fresh eyes and an untainted view of the state of the functionality.

As Designers we love doing new things or reinventing traditions in hopes to create something fresh. However, creating new metaphors for navigation on an eCommerce site can be devastating for conversion (i.e.- getting someone to find a product, add it to their cart, and checking out). That’s why UXD professionals research existing metaphors, common mental models, and interface design best practices. When you have to get it right, you have to prove that it’s going to work through research and testing.

So now we’re starting to stretch our User Experience legs a bit and branch off into areas that are not commonly visited by lay people. Bare with me as I attempt to make this area that is particularly thick with new terminology easier to understand.

I find that the best way to describe these types of things is to give real world examples. So I’ve pulled out a sample or two from my project files.

Understanding the User

I find it difficult to build something if I don’t know who I’m building it for. I can create a design based upon what I “think” is going to work but I’m never truly sure unless I know the audience. And that’s where the research comes into play. Late last year and earlier this year I had the opportunity to work on the redesign of a college web site. Now, I graduated from college so I had preconceived notions of what I thought goes on in one’s mind the last year of high school and the first year on campus. But therein lies the problem. The site is not supposed to be built for a guy who already went to college- it’s for the ones who are coming to and attending their first year. Also, this was a well-known community college in an area in which I didn’t grow up. You’ll be surprised how much location and environment influence decision making.

So the first thing we did (it was two of us) was identify our audience. Our focus was prospective and current students. The ages ranged from (16 to 19) years old with a mix of workforce development students in their mid-thirties and forties. Next we profiled them by interviewing them. Most interviews were conducted over the phone and others were focus group settings in the high schools. And the results were amazing! We found that the younger adults wanted to know the ethnic make up of the prospective school. Some wanted to be near their homes so that they could continue to help their parents in the family business. Others wanted to know exactly how much it was going to cost because they were paying for school out of their own pockets with little financial aid.

All of this information and more provided us with a mental model of their needs. We had a great illustration of the types of information they were going to be seeking in order to make a decision as to apply or move on. I can tell you that some of my preconceived notions were correct but some were not. And the ones in which I was wrong would have been detrimental to the entire experience. That is why research is critical to a site of this scale and purpose.

We utilized other methods that supported the activity I highlighted above but it would be a very long post to list and describe them. The point is that the time we invested in research netted a site that was dead on with our target audience. But what about testing our execution of the research?

Usability Testing

Now testing comes in all different shapes and sizes just like research. The idea is to use the right set of tools to get the job done. But generally we want to test our theories of certain things that we think will work based upon our research. And in order to do that effectively we have to set up a test with the right audience. This includes recruiting users who fit the target group. Now you can debate this point all day long but you don’t have test (100) people in order to discover the major flaws in your design. Actually you can begin to see trends in a sample as low (5 to 8). Our objectives here are to find at a high level what we did right and what we did wrong.

In Jakob Nielsen’s book, “Usability Engineering” he states that usability testing seeks to find: learnability; efficiency; memorability; errors; and satisfaction. I encourage you to check out his book to get more information on these usability attributes but for today I’ll just reference them. Once we’ve defined our audience, completed recruiting, and set up our schedule we executed our tests with a facilitator who follows a set of questions which in their essence addresses the five attributes we just mentioned.

I’ve conducted tests with paper and PostIt Notes all the way to working applications and web sites. After reviewing the results of the data I typically find trends. For instance, 6 out of 10 users had difficulty navigating to the Shopping Cart. Whereas 4 out of 10 successfully completed a transaction. This information tells me that we may be able to increase our transaction completion rate if we work on the navigation issue.

What you should be noticing is that User Experience Design is not about the visual design. Rather it is about the information that feeds the visual execution. If you’ve got a simple brochureware type of site then this type of research and testing may be too much for your budget to handle but if you’re building a web site or application that has to deliver then you must invest in UXD. Design can be so subjective. After all my green is not your green. But when you have research and testing behind your design you have a value that will net returns.

Wrapping Up

I’ll summarize this series on “Defining User Experience Design” next week. We’ll review the characteristics and value of UXD and how to implement it’s principals into your next project. Stay tuned!

POST A NEW COMMENT

Your Message

Name

Email

Website (optional)

Subscribe to replies