Testing helps you develop
designs that won’t frustrate
Why should you test?
When it is easy for people to use a design, they are more likely to be
happy, to spend money, and to be truly engaged with your design
(and maybe even your organization).
For other technology, it can mean fewer support calls, more forgiving
customers, and longer, happier customer relationships.
Put together a design
prototype, pick an interviewer,
ﬁnd users (participants), and
ﬁnd a place to watch them use
What do you need?
It’s really simple. You don't need recording equipment, but you might
want to take notes, so a clipboard can be handy. And you probably
want an envelope or folder to hold materials like your paper prototype
Test when you have an idea of
what is going to go into the
design or when something has
When should you test?
Testing is useful when:
• you’re not sure what is frustrating about a design
• you have a question about how to design something
• there are arguments among the people on the team about the
• you want to make sure what you’re designing will work for the
Usability testing answers
questions about how and why
people use your design.
Know why you are conducting a usability test.
Usability tests can answer questions like these:
• How easily and successfully do people ﬁnd what they are looking
• Where do people get lost as they navigate the design?
• How close does the design match how people think about what they
are trying to do?
Start testing with a few users
trying out the ﬁrst versions of a
prototyped design, one at a
who is needed
to run a test?
4 test participants
If you’re testing with types of users, ﬁnd four people of each type. You’ll be
interviewing them one at a time.
Make sure it’s not someone who designed what is being tested.
Someone to take notes and wrangle participants.
Everyone who contributes to design decisions should watch at least 2 users go
through the test experience.
Test with what you have
available. Test again when you
have a version that appears to
work. Test again when you have
the ﬁnal version.
What do you test?
When you do your ﬁrst usability test, you might want to practice on someone
else’s design. That way, you won’t feel so bad when you test yours.
Otherwise, you can test:
• mock-ups or early versions
• competitors’ products
• prototypes at various stages of completeness
• a nearly ﬁnal version
People who are like your users
are everywhere. Go to them.
where should you test?
Choose a place where you can ﬁnd people who would normally use your design:
• food courts
• farmers’ markets
• street fairs
• buses, planes, or train
Use your imagination. Don’t be afraid to ask a stranger for feedback on your design.
Follow these steps to run each
session of a usability test.
in a usability test?
1. Introduce the session.
• Go over what will happen.
• Give interactions
• Give the participant the design.
2. Watch the participant use the design.
3. Listen for questions (don’t answer them) and comments (write them down).
4. When they are done, ask the participant to walk you through what they did and why.
5. Thank the participant profusely.
What is the role
of the interviewer?
As the interviewer, you guide the participant through the session,
watch what she does, and take notes (if you can).
Do not help the participant use the design. (Well, not until after you
have learned what you need to learn.)
Ask open-ended questions, like, “How did that go?” follow up with a
statement like, “Tell me about how youdid that.” But not too often.
What for wrong turns, listen
to questions, look for
you look for?
• ask for help with instructions or using the design?
• ask question?
(If so, what questions?)
• make comments?
(Again, note what they say.)
• take out reading glasses or lean way in?
• ﬁnd their way through the design efﬁciently?
• have trouble mavin through the design, or make wrong turns on their way to doing what they wanted?
• seem confused, puzzled, or frustrated?
Review what you saw and
heard. Tally the types of
problems participants had.
What do you do with what you ﬁnd out?
Look at what parts of the design caused questions, comments,
mistakes, or request for help.
This should tell you what is confusing to users, what is unclear, and
why. It should also tell you what might need instructions, messages, or
a different label.
– COLLIS TA’EED
“Things you think are obvious often aren’t, text you thought
explained something doesn’t even get read, and generally
speaking users do things they weren’t supposed to do. Even if the
only user testing you ever do is asking some friends to use a site
while you observe them, you’ll already be better for it.”