February 12, 2003
All Those Opposed
"It takes experts to understand user behavior. We don't have that kind of money."
It's true that you can spend a fortune on research and analysis of user behavior
if you are so inclined: field studies, ethnographic research, and contextual
inquiry methods can require some pretty rarefied expertise. But adopting a user-centered
approach doesn't require such lavish expenditures. Taking a day to talk with
a few users over the phone can provide a wealth of useful insights into their
attitudes and expectations. And you don't have to have any specialized equipment
or an advanced academic degree to be successful.
"It doesn't take experts to understand user behavior. We'll figure it out
as we go." Successful user experience design calls for understanding the
nature of the problem before you set about constructing a solution. If you don't
understand the problem, the only way you can solve it is by accident. User experience
problems tend to be so complex that the odds are against you from the start
if you expect to stumble upon the solution. It doesn't take experts, but it
does take a willingness to plan your approach. Common sense alone isn't enough
to guide you to the correct answer.
"We'll fix it in QA." When there are problems with the user experience
of a site, they often aren't as easy to isolate as the kind of coding errors
that the quality assurance process is designed to catch. User experience issues
are frequently more systemic, requiring rethinking of fundamental aspects of
the entire site. And because QA comes near the end of the overall development
process, there's rarely time to undertake the kind of large-scale design reworking
that's often needed. (Besides, your QA team has plenty to do as it is, without
trying to identify user experience issues.)
"We can't make room for it in the schedule." In the heat of development,
it's easy to forget that the launch of the site is not the end of the process—it's
the beginning. Your attentiveness to the calendar is unlikely to mean very much
when the resulting site isn't meeting user expectations. The amount of time
you spend on user experience at the beginning of the project is nowhere near
the amount you'll spend retooling the site and taking another blind stab at
meeting user needs. The ultimate success of the project is not about getting
to launch. It's about what happens after that.
Reaching Accord
Creating a successful user experience isn't simple common sense, but it isn't
rocket science either. User-centered design isn't a methodology or a collection
of tools and techniques. It's a way of thinking about how you design the site
so that your decisions are rooted in an understanding of the underlying strategy.
It's a philosophy that seeks to remind us: If you're only looking at the problem
from your own point of view, you're only going to be, at best, half right.
Asking the Right Questions
Jesse James Garrett's new book The Elements of User Experience
(New Riders) deals with the fundamental concepts of user-centered design
for the Web. In this exclusive excerpt, Garrett talks about good and bad
ways to adapt user-centered techniques to fit limited budgets and schedules:
Facing the tangle of small problems to be solved in creating the
user experience can sometimes be disheartening. Occasionally, a solution
to one problem will force you to rethink other problems you thought
you had already solved. Many times, you will have to make compromises
and evaluate tradeoffs between different approaches. When you're in
the middle of having to make these kinds of decisions, it's easy to
become frustrated and question whether you're taking the right approach.
The simple fact is this: The right approach is one in which no aspect
of the user's experience is left to chance. Make every decision consciously
and deliberately, and ground each decision in your understanding of
the underlying issues at play.
Having the right frame of mind about approaching the problems you
are facing matters most. Every other aspect of the user experience
development process can be adjusted to fit the time, money, and people
at your disposal. No time to gather market research data on your audience?
Maybe you can find ways to look at the information you already have,
such as server logs or feedback emails, to get a sense of the needs
of your users. Can't afford to rent a usability test lab? Recruit
friends, family, or coworkers to participate in informal testing.
The worst mistake you can make is to gloss over the fundamental user
experience issues of the project in the name of saving time or money.
On some projects, someone will thoughtfully tack on some form of user
experience evaluation to the very end of the process—long after the
time to actually address those issues has run out. Racing toward launch
without looking back might have seemed like a good idea back when the
launch date was set, but the result is likely to be a site that meets
all the technical requirements for the project but doesn't work for your
users. Even worse, by tacking user experience evaluation on at the end,
you might end up launching a site that you know is broken but have no
opportunity (or money left) to fix.
Some organizations favor this approach, calling it "user acceptance testing."
The word "acceptance" is very telling here—the question is not whether
they like the site or will use the site, but rather can they accept the
site? This type of testing all too often happens at the very end of the
process, by which time countless assumptions have gone into shaping the
user experience without ever being examined. Those assumptions can be
extraordinarily difficult to uncover in user testing, because they are
hidden behind layers of interface and interaction.
Many people advocate user testing as the primary means of ensuring
a good user experience. This line of thinking seems to be that you
should make something, put it in front of some people to see how they
like it, and then fix whatever they complained about. But testing
is never a substitute for a thoughtful, informed user experience design
process.
Questions that focus on specific elements of the user experience
can help you gather more relevant input from your users. User tests
constructed without an eye toward the elements of user experience
can end up asking the wrong questions, which in turn can lead to the
wrong answers. For example, when testing prototypes, knowing what
problem you're setting out to investigate is crucial to presenting
your test subjects with an experience that doesn't cloud the matter
with unrelated issues. Is the problem with that navigation bar really
just the color? Or is it the wording that your users are responding
to?
You cannot depend on your users to articulate all of their needs.
The challenge in creating any user experience is to understand the
needs of the users better than they understand those needs themselves.
Testing can help you understand the needs of your users, but it's
just one of many tools that can help achieve the same end.
—JJG
|
Jesse James Garrett is a founding partner of Adaptive Path, a user-experience
consultancy based in San Francisco, and author of
The Elements of User Experience:
User-Centered Design for the Web. Contact him at
jjg@adaptivepath.com.
Previous Page |
1
|
2
