What is a Walkthrough?
 Download 'printer-friendly' PDF version (File size: 16 KB)
A walkthrough is a cost-effective technique for testing a user interface.
There are several variations - this document describes a simple paper-based technique.
When are walkthroughs appropriate?
Use walkthroughs to get early validation of design decisions, and to get feedback from
several people at once. Walkthroughs are particularly effective for testing gross navigation.
You can also conduct walkthroughs if you do not have the resources for formal usability
testing.
A significant weakness of paper-based walkthroughs is the fact that they do not show
interactive behaviour (such as button-presses). Walkthroughs can include an interactive
screen-based presentation to overcome this.
How is a walkthrough conducted?
Each participant is given a booklet containing tasks and screens.
Screens are presented one at a time, and participants are asked to specify the actions
they would expect to take on each screen (for example, entering data, clicking a button or
hyperlink).
Participants are encouraged to write on the booklet. After each screen is presented, the
group as a whole is asked to comment on any issues with the screen, and what elements did
and did not work. It is important to log these comments, as participants may not be clear
in their written comments.
After each screen, participants are told what action the hypothetical user has taken, and
the next screen in the sequence is presented.
The walkthrough continues until all screens and tasks have been completed.
At the end of the session, it is appropriate to administer a questionnaire to confirm that
the participants fit the demographic profile required, and to get subjective feedback and
comments.
A typical walkthrough session with four tasks lasts for 2 hours.
Preparing for a walkthrough
The tasks to be completed should be based on the scenarios used during the design process,
and should be representative of the tasks the interface is designed to support. It is
important to exercise as many elements of the interface as possible. If there are any
elements of the user interface that you suspect may be troublesome, make sure the tasks use
those elements.
Tasks should be worded in a simple and straightforward fashion, and should not describe any
aspects of screen interactions.
Each task should be printed on a page of its own in a large font. Consider a different
colour paper for the tasks so that participants can easily refer back to them. Using an
overhead projector to display the tasks makes it easier for people to remain focused on
the current activity.
Participants should be representative users, although the ability to have many participants
provides the opportunity to have other stakeholders participate.
Ensure that screens contain appropriate data. For example, if a screen would contain a
customer name, make sure that this appears on the relevant screen in the booklet.
It is critical to conduct a pilot session to uncover missing steps or data, and to identify
unclear instructions or questions.
Use large format paper (such as A3) in landscape orientation.
Guidelines
- Point out to participants that you are testing the interface, and not their abilities. Use a script to ensure you provide all necessary instruction.
- Include page numbers, so that you can direct participants to particular screens.
- Include the participant number, for cross-referencing with questionnaire or other data.
- Make sure you note all comments made during the walkthrough, and write up your notes as soon as possible after the walkthrough.
- Discourage discussion of any screen until all participants have finished writing.
- Encourage developers and business representatives to observe or participate.
- Discourage participants from changing their comments based on what other say.
- Ensure screens have a polished but simple look, to avoid participants being distracted by graphic design issues.
Download a 'printer-friendly' PDF version of this sheet
(File size: 16 KB)
|
