You are on page 1of 8

KM

COLUMN
NOVEMBER 2004
What is usability?

There’s something wrong with our computer This second aspect is sometimes described
systems. Much of our time is filled with frus- as usability engineering, and is more accu-
tration – websites and intranets that don’t let rately described as user-centred design.
us find the information we want, word proc-
essors that lose hours of work with a click of
the mouse, remote controls that have more ‘Usability’ refers to the quality of
buttons than we could possibly need.
‘Usability’ is widely seen as the answer to
a system and the process of
many of these frustrating interactions with designing a usable system
technology. There are usability books, web-
sites, guidelines and checklists, so why are
products and systems getting harder to use Usable – Usability as quality
instead of easier?
There are a number of definitions for the
There is some confusion about what usabili- quality aspect of usability, including this
ty is and whether businesses are ‘doing usa- one from ISO 9241-11:
bility’ or not. There are many aspects to
“the extent to which a product can be
usability – more than just running a usabili-
used by specified users to achieve
ty test at the end of a project.
specified goals with effectiveness, ef-
This article provides an overview of what ficiency and satisfaction in a specified
usability is (and what it is not). It provides context of use.”
ideas on how to include more usability ac-
This definition includes four elements that
tivities in projects and the types of activities
are necessary to create a usable system:
that are needed in order to create more usa-
ble systems. 3. There are specified users of the system.
4. The users have a set of specified goals.
What is usability?
5. The system should allow user goals to be
‘Usability’ is an umbrella term that encom- met (effectively) in an efficient manner
passes two related concepts: and the users will be satisfied with the
1. Usability is an attribute of the quality of process or outcome.
a system: 6. The system will be used in a particular
“We need to create a usable intranet” context (e.g. within a physical location, a
business process).
2. Usability is a process or set of techniques
used during a design and development A usable system is one that is designed to
project: consider all of these aspects.
“We need to include usability activi- Why is usability important
ties in this project” A highly usable system offers benefits to
both users and business.
Donna Maurer is a senior mem- The primary benefits to users are that they
ber of the Step Two Designs con- are able to achieve their tasks easily and ef-
sulting team, and a leading ficiently. This sounds simple, but the feeling
specialist in usability and infor- of achievement that people get when they
mation architecture techniques. use a computer system without frustration
Donna has worked extensively in
should not be underestimated.
the Australian public sector, with
a particular focus on intranet re- Unusable systems can result in substantial
designs. costs to the business, not just in terms of lost

© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au


intranets • usability • information architecture • knowledge management • content management
sales, but in customer satisfaction, staff pro- • If people have to use a difficult system,
ductivity and support requirements. they will waste time.
Where work arounds are not possible, a
difficult system will cause staff to be in-
efficient for a longer time than necessary.
People will go to great lengths to It will take them time to learn how to use
avoid using difficult systems the system and any task that they don’t
do frequently will take longer and result
in more errors.
Systems that are difficult to use result in the • More support is necessary
following business costs: A system that is difficult to use requires
• If the system is difficult to use, people much more support than an equivalent
won’t use it. easy to use system. The need for docu-
mentation, training and help desk sup-
Whenever a choice is available, people port is higher.
will take the easiest option. This is partic-
• More changes are necessary
ularly common on websites, where there
is choice and a wide variation in usabili- If user needs were not considered when
ty. If people can’t find a product on your determining requirements (functionality
is missing) and user abilities were not
website, they will do so somewhere else.
considered when designing the system
The alternative may not be another sys- the system is likely to require more
tem, but another method or format. If changes and enhancements.
people have a choice between a simple,
familiar paper form and a difficult online
form, it is easy to see why they would The only way to tell if a system
continue with the paper version. Poten-
tial business benefits associated with the
is usable is to watch people
online form (including reduced process- using it
ing costs) are not realised.
• If people have to use a difficult system, How to tell whether a system is usable
they will do so as little as possible.
The only way to determine whether a system
When people are given no alternative but is usable is to get end users to use it for real
to use a difficult system (such as a core tasks.
business application or a standard word The main way this is done is through usabil-
processing package), they will go to great ity testing. In a usability test, users of the sys-
lengths to avoid using the system. They tem attempt tasks while being observed. The
will create elaborate work-arounds, put a observers don’t tell the usability test partici-
lot of effort into finding ways to avoid us- pant how to use the system and don’t answer
ing the system and will share this infor- questions – it is as if the participant were do-
mation with others. ing the tasks alone.
For example, on difficult business sys- The usability test identifies the key usability
tems, people fill in as few fields as they problems with a system (which enables
can, and put dummy data in those that them to be fixed); and/or collects quantita-
tive measures of efficiency, effectiveness
they can’t avoid (causing poor data qual-
and satisfaction before release.
ity and increasingly strict measures or
training efforts to improve quality). On a The key aspects of a usability test are:
difficult intranet, people print out large • Participants involved are existing or
documents or email them to colleagues future users of the system.
so they don’t have to look for them again, It is important to involve the right people
defeating the purpose of the intranet as a in a usability test. For example, for a busi-
place for accurate and up to date informa- ness system the day-to-day users should
tion. be involved, not their managers.

What is usability? • Page 2


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
• Participants attempt tasks that they • Asking another system developer
would normally perform on the system. The designer or developer of a similar
It is crucial that the tasks that are includ- system may be able to give you some ide-
ed in the test be realistic. They must also as about usability. However, unless they
describe the entire task and be written to are going to be a user of the system, they
remove bias. are likely to be different to the real user
group. For example, they are likely to
• The usability test is set up in a way that know more about the technology and less
is as close to the normal context as about the subject than the end users.
possible.
• Asking a usability expert
The usability test may be conducted at
A usability expert should be able to give
the participants desk or in their home if
you ideas on likely usability issues based
the location is relevant. Alternatively, a
on experience observing people using
usability test may be conducted in a usa-
similar systems. However, almost all ex-
bility lab, an office or a meeting room.
perts miss issues that occur for end users
The equipment used should be similar to
of systems. An expert may not know the
the user’s normal equipment.
tasks that need to be completed and may
know more about the technology than the
users. A good expert will tell you this be-
Demonstrating the system does fore offering an opinion.

not identify whether it is usable Usability checklists, guidelines and standards


Usability checklists, guidelines and stand-
ards can be useful to provide ideas about us-
How not to tell if a system is usable ability issues, but they must be assessed to
determine whether they are relevant to the
There are a number of methods in use that
users and context.
aim to gather feedback about a system. How-
ever, unless they involve end users using the As an example, a common website guideline
system to complete tasks, they will not tell is that all content should be available within
you whether a system is usable: three clicks. The intent of this guideline is
good – it highlights that people should not
• Demonstrating the system have to click endlessly for information.
Showing a system to users and asking for However, there are situations where it is not
their feedback is a good way of getting applicable. Large sites or sites with informa-
their involvement in a project, but does tion designed to present increasingly de-
not provide information on whether it is tailed pages may not meet the guideline, but
usable. This method only collects opin- may be very usable for the intended audi-
ions or personal preferences. In most cas- ence.
es people will mention things that do not Guidelines and standards are best used to
end up being problems and will fail to identify the most obvious usability problems
identify things that are problems. and fix them before a usability test is con-
ducted.
• Surveys
A survey can collect people’s opinions on
a system and might provide some inter- Checklists and guidelines
esting feedback. Respondents may be
able to report on problems with the sys- should be used to identify
tem or changes that they need, however,
usability problems can be subtle. It is dif-
usability problems before
ficult to know why problems occur with- testing
out observing people using the system.

What is usability? • Page 3


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
How to create a usable system
The best way of ensuring that the system Learn about human abilities and
is usable is to follow a user-centred de-
sign process (outlined below), where us-
limitations
ers are considered and involved in the
process, and usability testing is carried About – Learning about people
out at many stages of the design process.
The ability to design systems that are usable
for people requires an understanding of hu-
man abilities and limitations. The key char-
Learn about people, design for acteristics relevant for many everyday
systems are outlined below, but there are
the user group, work with the many more that will vary depending on the
audience and system.
users
• Memory
Everyone jokes about how bad their
User centred design memory is. The human memory does
have many limitations that affect our eve-
User centred design is a process or set of ryday life. For example, we have limited
techniques that can be used to create a usa- capacity in working memory, have to re-
ble system. User centred activities can be in- hearse to store in long term memory and
cluded throughout the design and remember better if we practice.
development process, from concept devel-
Usable systems minimise the amount
opment to final testing and beyond into sup-
that we need to memorise. This may in-
port and maintenance.
volve displaying relevant information at
User centred design is not a step by step appropriate points, providing compari-
magic methodology. It is closer to a philoso- son charts or providing flexible password
phy or way of looking at the world. It is options.
about genuinely wanting to create some- • Error
thing that can be used by the real audience
It is a fact of life that people make errors.
and about remembering that the system is
That users will make errors is one of the
not for the team creating it.
most predictable parts of systems design.
User Centred Design involves three key as- A characteristic of a usable system is that
pects: it is designed to be tolerant of errors. This
7. learning about human abilities and limi- can be done by acknowledging that errors
tations will occur, preventing those that are pos-
sible and allowing easy recovery from
8. designing for a particular group of peo- those that do occur.
ple
• Perception
9. working with these people throughout Perception relates to the way that people
the design process acquire and organise sensory informa-
It does take time to learn about the user tion. It involves the acquisition of infor-
group and may need some specialist skills to mation through the classic five senses
ensure that the design matches human abil- (sight, hearing, smell, taste, touch) plus
ities and limitations. However, the time and proprioception (body awareness) and
effort will be more than offset during the de- equilibrioception (balance).
sign and development process. The focus on Sight is the primary sense used in many
user abilities and needs means that the team situations and the one that is most rele-
has a much clearer picture of the require- vant for the design of computer-based
ments from the start. A good understanding systems.
of needs, and regular involvement of users Usable systems recognise our perceptual
also means that usability issues are identi- abilities and limitations and allow us to
fied early enough to fix them. use the sense that is best suited to tasks.

What is usability? • Page 4


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
• Decision making For – Designing for the users
People believe that decisions are made in The second aspect of user centred design is
a logical way – that we evaluate options that it is about a particular set of people – the
and make rational decisions. In practice, current or future users of the system. To de-
most decisions are not made this way. In- sign a usable system, it is crucial to gain a
stead, we often satisfice – accepting the good understanding of who will be using the
first reasonable answer that fits the situa- system. It is also important to articulate this
tion. so that the knowledge about the user group
is available to relevant stakeholders.
Designers of systems that help people to
make decisions (such as information sys- Who are the users?
tems) must recognise that people are like- For many systems, it is easy to determine
ly to use the first piece of information who the users are. For example, most busi-
they see, may not read beyond that, and ness applications are created for a specific
are very unlikely to read disclaimers or purpose with a limited audience. However,
additional information. there are cases when it can be more difficult
• Information seeking behaviour – many government websites are designed
for ‘the Australian population’ without con-
Most information seeking behaviour is sidering what this really means.
one of two types:
If you don’t know who the users of an exist-
– Known-item: when people will ing website are, do some initial analysis of
know exactly what information they site statistics, search logs and referrer logs or
are looking for, know that it exists run a short survey to find out who is using
and know how to describe what they the site before undertaking some of the fol-
are looking for. lowing steps to learn more about them.
– Exploratory (unknown item): when
people are not sure what information
they are looking for, may not know Learn what the users need by
how much information there is and
do not know how to describe what
observing them, not asking
they want. them
Some methods of information retrieval
suit some tasks better than others, requir-
Learning about users
ing an understanding of how people will
approach a particular set of information. Once you know who the users are, there are
a range of ways to learn more about them
• Mental models and what they need from the system.
One way that people make sense of the The most useful techniques involve observ-
world is to form a mental model of how ing people and talking with them about their
things work. The mental mode changes tasks, as opposed to asking them what they
over time with experience, but it may need. Asking directly is usually misleading –
never be accurate. people describe features they have already
An understanding of users’ mental mod- seen or used and are unlikely to know what
els can provide useful input to the way a possibilities exist. In addition, when asked
system is designed. The mental model of to describe how a task is done, people usual-
users can be used to guide the design and ly describe the way it should be done – ob-
increase the chance of people being able serving them shows how it is actually done.
to use it. • Interviews
Interviews involve one-on-one discus-
sions with users of the system. They are
good to identify key needs and uncover
A system is used by a group of important issues. They are conducted in
people – learn what they need the user’s normal context and are useful
for gaining practical information.

What is usability? • Page 5


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
• Task analysis • Scenarios
Task analysis is used to break down the Scenarios are short stories that describe
tasks people will perform with the sys- how users complete tasks within a partic-
tem. It identifies goals of the tasks, the ular context. They can be used to inform
steps involved, inputs, outputs and infor- the design process and communicate
mation needs.
about tasks that people do. They are also
• Contextual enquiry used during usability testing.
Contextual enquiry involves observing • Task lists
people undertaking their normal tasks.
The key difference between this and in- Task lists are simple listings of all of the
terviews is that contextual enquiry focus- tasks that users will want to do with the sys-
es more on observing the task and less on tem. They are most suited to systems that in-
discussing how it is done. volve a large number of discrete tasks such
• Focus groups as informational sites. They may include the
types of information that are needed to com-
A focus group is a facilitated discussion
plete the tasks.
used to elicit attitudes, feelings and ideas
about a particular topic. It is useful to
gain ideas about issues early in the sys-
tem design process. It is best conducted Involving users throughout the
with small groups (6-8 people) who are or process results in fewer
will be users of the system.
• Surveys
usability problems at the end
Surveys and questionnaires are an effi-
cient way to gain input from a large With – Working with users
number of people. They are best suited to
gathering opinions rather than specific The third aspect of user centred design is in-
information about tasks. To ensure that volving the users throughout the design
the results are valid, care must be taken process. There are many ways that this can
when developing the survey and analys- be done, from concept design to implemen-
ing the results. tation and beyond.
In the early stage of a system design process,
the techniques discussed above are most ap-
Share user research outcomes propriate. Using these techniques ensures
that you gain a good understanding of what
so everyone can focus on the the system needs to do.
users Further into the system design process, oth-
er ways to involve users include:
Communicating user needs • Collaborative design
It is important to communicate the charac- Collaborative or participatory design in-
teristics and needs of users to other people volves business representatives, design-
so that everyone can focus on user needs. ers, developers and users working
• Personas together to design the system. It is a good
Personas are fictitious users that repre- way to involve users throughout the
sent the needs of specific user groups. process, but needs to be managed careful-
They are used to guide design activities ly. With too much exposure, the users in-
and to communicate about users to other volved can become experts in the system
stakeholders. They are created as a result and fail to represent the user group. Col-
of user research activities. laborative design sessions are to learn
For more information, see our article “An more about user tasks and how the sys-
introduction to personas and how to cre- tem will support them, not to get users
ate them” at www.steptwo.com.au/papers/ (who are not experienced designers) to do
kmc_personas/ the design.

What is usability? • Page 6


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
• Design walk-throughs • Paper prototyping
Design walk-throughs are used to check Paper prototyping involves creating
design decisions and collect feedback on mock-ups of the system screens with
an early version of a design (a version enough detail to demonstrate the system.
that is too early to usability test). It in- Users work through scenarios on the sys-
volves walking through the design using tem while a facilitator shuffles screens.
real scenarios and discussing actions and Paper prototypes are best hand drawn
aspects of the interface. (rather than computer drawn) so partici-
• Card sorting pants can tell that they are still under de-
velopment.
Card sorting is a quick, simple technique
used as input to the structure of an infor- • Usability testing
mation system such as a website. Users Usability tests should be conducted
are provided with a list of potential con- throughout the design cycle, first on pa-
tent (written on index cards) and asked to per or early computer prototypes, then on
sort them into groups. a more complete system. This will help to
• Free listing identify usability issues as they arise – it
is much easier to make changes to early
Free listing is a technique that is used in designs than a final system.
group sessions. Participants are provided
Small, frequent usability tests provide
with a topic and the group has to come up
much better value than one large usabili-
with as many examples for the topic as
ty test.
possible. For example, the group may be
asked to list as many business forms as
they can. It provides insight into items Conclusion
that are related and the words people use Usable systems deliver substantial benefits
to describe the topic. It is useful as input for business and users. A system that allows
to the content and labelling for informa- people to complete their tasks easily, quick-
tion systems. ly and without frustration will save time and
• Prioritisation sessions money for the business and allow business
goals to be met.
In a prioritisation session, users are pro-
vided with lists of new functionality and It is possible to create usable systems – by
content and asked to sort them in various considering human abilities and limitations,
ways. They can sort them according to learning about the needs of the user group
usefulness, how often they would use the and involving the users throughout the de-
function or how quickly they need to ac- sign process. Taking these steps may involve
cess it. This is a practical way to prioritise a little more effort in the early stages of the
requirements. process, but ensures that the end result is go-
ing to be more useful and more usable, and
therefore used.

More articles! Did you like this article?


Find out when new papers are published: Send your thoughts and feedback to:
www.steptwo.com.au/papers/subscribe contact@steptwo.com.au

Step Two DESIGNS


www.steptwo.com.au • contact@steptwo.com.au
knowledge management • content management • intranets

What is usability? • Page 7


© Copyright 2006, Step Two Designs Pty Ltd • www.steptwo.com.au
TM

The Intranet Roadmap™ provides the first truly


comphrehensive methodology that describes all the
activities required to develop (or redevelop) an
intranet.
Beyond just implementing software or redesigning the
site, the Intranet Roadmap covers activities in five key
streams:
n strategy
n design
n content
n change & communications
n technology
The Intranet Roadmap does more than just indicate
what activities need to be conducted, it clearly shows
the sequence of activities required, and the techniques
that will help at every stage of the project.
The Intranet Roadmap is delivered in two forms:
n full colour A1 wallchart
n supporting 54 page booklet
The wallchart lists the key activities required in each of
the project streams. It also highlights which activities
(such as usability testing, affinity diagramming,
personas and collaborative design) can be used to
support individual activities.
The supporting booklet then provides an overview of
Task Focus every activity and technique listed on the Intranet
analysis groups
Roadmap, as well as linking to further resources and
information.
Usability Develop authoring Staff
testing
guide interviews The combination of the wallchart and booklet will be
invaluable for any team looking to develop or
Training needs Design
analysis patterns redevelop an intranet, and it will assist in both
Train authors planning and reviewing the approach taken.
Train small
groups

Personas
Write and acquire
us$120 + $15p&p
content
Accessibility
testing
Secure online ordering:
www.steptwo.com.au

Step Two DESIGNS