CHI 97 Electronic Publications: Tutorials
Product Usability: Survival Techniques
Jared M. Spool, Tara Scanlon, Carolyn Snyder
User Interface Engineering
800 Turnpike Street, Suite 101
North Andover, MA 01845 USA
+1 508 975 4343
jspool@uie.com, tara@uie.com, csnyder@uie.com
ABSTRACT
Product developers are typically faced with small budgets, tight schedules, and
over-committed resources. To deliver high-quality products under these constraints,
developers need an understanding of basic design principles, techniques that allow
them to work effectively with materials on hand, and a development process that is
built around the use of such techniques. This workshop explains how low-fidelity
prototyping and usability testing can be used in a process of iterative refinement
in order to develop more usable products.
KEYWORDS
Design principles, usability testing, usability evaluation, prototyping, low-fidelity
prototyping, process management, product development, practical techniques.
© 1997 Copyright on this material is held by the authors.
PART 1: DESIGNING FOR USABILITY
In product development and interface design, the notion of "getting it right the
first time" is a myth. Rather than focus on getting something perfect from the
beginning, we advocate a process of successive refinement that allows for frequent
measurements against predefined goals.
The Cause of Usability Problems
Our research shows that the vast majority of usability problems come from a single
cause: the development team didn't know a key piece of information. If they had
known that information earlier, they would have designed the product to accommodate it,
and the usability problem would never have occurred.
The Four Stages of a Prototype
In a sense, the current release of a product is just a prototype for the next release,
just as the previous release was a prototype for this release. Every prototype goes
through the same four stages: Plan, Implement, Measure, and Learn. In the Plan stage,
the team looks at their areas of highest risk for the product and determines what
information they need for the key decisions they are facing [1]. During the Implement
stage, the team builds just enough of the product so that they have something to test.
In the Measure stage, the team collects data that will help them make decisions.
Usability testing is one very effective way to collect this data. And in the Learn
stage, the team responds to what they've observed by adjusting the interface,
functionality, or schedule for the product.
Successful Iterative Refinement
The key to this four-stage process of successive refinement is to conduct the stages
very quickly, ideally in a matter of hours rather than weeks or months. Implementation
is often the most time-consuming stage, so reducing implementation time will allow a
development team to go through more cycles, and discover key information earlier in
the project.
PART II: PAPER MOCK-UPS
The purpose of a prototype is to provide feedback on a design. But implementing,
testing, and shipping an entire release is an expensive way to get critical
information. Using paper mock-ups substantially shortens the Implement phase,
allowing more iterations than would be possible with electronic prototyping techniques.
The Basics of Paper Mock-ups
Using common office supplies, (markers, index cards, transparency film, etc.) the
development team can quickly construct a fully-functional prototype of the product
interface. This technique has proven effective even for sophisticated, high-tech
products. We have found that, with a little imagination, you can effectively
prototype almost anything!
Usability Testing with Paper Mock-ups
Usability testing of a paper mock-up is quite straightforward. Using their finger as
a "pointing device," users can select from menus, click on buttons, and otherwise interact
with interface elements. One or two members of the development team simulate the
behavior of the computer, taking the appropriate action in response to the users'
requests. Paper mock-ups are easy to modify, so the team can even make changes in the
middle of a usability test!
Advantages of Paper Mock-ups
An iterative development process that employs low-fidelity prototyping is an
excellent method for ending "opinion wars" and other project-killers. Instead of
wasting time on subjective arguments, the development team can instead focus on
objective usability goals (e.g., an untrained user can complete the installation
within 5 minutes) and test different approaches until the first satisfactory one is
found. At that point, the team can turn their attention to the next issue on the
priority list.
Comparison with Electronic Prototypes
An electronic prototype (such as one developed in Visual Basic) seems more attractive
at first glance because it has a more polished "look." However, when it comes to
prototyping the "feel" of the product, i.e., its behavior, some form of programming
is generally required. In a paper mock-up, the "feel" is simulated by the person
playing the computer, so this programming time is eliminated. In our experience,
most of the risky issues in product development pertain to the "feel" (Does the
product have the right set of functions? Can users figure out how to do their work?)
rather than the "look" (Are the icons clear? Does it conform to the style guide?).
Drawbacks of Paper Mock-ups
Paper mock-ups are not a cure-all. For example, it is not possible to assess response
times with a paper mock-up, and it may be harder to get accurate feedback on the
visuals of the product. Also, sometimes it is necessary to fake data that would be
available in an actual running system. We generally recommend using paper mock-ups
early in the product development process, when their advantages are greatest.
PART III: BASIC DESIGN PRINCIPLES
Product developers need a working knowledge of the key factors in good interface
design. We have found that a majority of usability problems can be addressed (or
prevented) by careful attention to three principles: affordances, mental models,
and tool time.
Affordances
Affordances [2] are the built-in "clues" that communicate to users how the interface
is to be used. Buttons, icons, words, and other controls make up the visual language
which helps users know what to do. In many cases, usability problems can be attributed
to false or missing affordances.
Mental Models
Users will subconsciously develop their own cause-and-effect mental model of how an
interface works [2]. By focusing on the users' mental model, designers will better
understand which details of the product's inner workings should be communicated to
users (or hidden from them!).
Tool Time
Tool Time is the portion of time that users spend being distracted by the interface
("how do I indent this paragraph?") instead of focusing on their own work ("what do
I want this memo to say?"). Reducing tool time allows users to be more productive.
Whenever developers can identify tool time, they can often minimize it through means
such as eliminating steps or moving key functions closer together.
DESIGN COMPETITION
The design exercise and competition is the high point of the tutorial. Using supplied
materials, participants work in small teams to construct a full working paper
prototype of a fast food ordering kiosk. The designs are then evaluated by a series
of usability tests, and each team has the opportunity to make changes after each
round. The team that has the most usable interface wins fabulous prizes!!
The competition gives participants "hands-on" experience in the concepts presented in
the course. Over 2,000 professionals have participated in past competitions, and many
tell us that this is the best part of the course.
REFERENCES
1. Snyder, C. Managing Risk by Using Paper Prototypes in Software Developer and
Publisher, Sep/Oct 1996, 38-40.
2. Norman, D. The Design of Everyday Things, Basic Books, New York, 1988.
CHI 97 Electronic Publications: Tutorials
