Usability On The Cheap

Paper Prototyping

For this test, sketches or textual navigation is placed on paper cards, or printouts of interfaces mocked up in HTML or created with Photoshop are used. Paper prototypes can be used throughout the development cycle, or before the "real" interface is created. Task-based testing is completed using the paper prototypes, and user feedback is gathered verbally and/or by observing their usage of the prototypes.

Advantages of using the paper prototyping method are:

• it's inexpensive,
• problems can be fixed properly,
• paper prototyping can be used at any stage of development and most importantly,
• the test can be conducted anywhere.

The main disadvantage of paper prototypes is that it is difficult to manage complex prototypes.

Heuristic Evaluation

Heuristic Evaluation should be performed either early in the phase of development, or after the usability test has been conducted. In this method, the testers perform an informed, critical inspection using a set of rules, checking for usability problems, lack of adherence to standards, and inconsistency of design -- so this method doesn't need involvement of test participants.

Heuristic Evaluation methods are quite inexpensive and consumes the least amount of time. However, the task list to be reviewed can be exhaustive. Evaluators' inexperience and team members' resistance to the findings can reduce the effectiveness of this method. And while the task list helps verify any problems, this method doesn't provide any design solutions.

For more information on Heuristic Evaluation, read this article.

Scenarios

Creating scenarios is another popular discount usability method. According to Jakob Nielsen,

While the horizontal scenarios are effective in evaluating the site navigation (during the early phases of development), the vertical scenarios are helpful in the later phases, when the development is in progress and the search or submission systems are being finalized.

While being inexpensive, scenarios allow the simulation of independent areas of the interface for usability evaluation. However, the limitation of using scenarios is that the user has to follow a predetermined path, with options predefined by the evaluators.

Conducting the Test

In a study conducted by Rolf Molich and Christian Gram at the Technical University of Denmark, 50 teams of students were made to conduct usability tests of commercial Websites as part of a user interface design class. The average time spent by each team was 39 hours. Hence, it was inferred that on an average, a discount usability test should not take more than 39 hours.

And Jakob Nielsen argues that:

The guidelines for conducting discount tests are simple: use a limited number of participants, and conduct the tests in an informal environment (thus making the participants comfortable with their surroundings). Don't isolate the participants completely as they might feel as if they're "being tested" and hence more conscious about their actions, which would, in turn, generate inappropriate results. On the other hand, don't have too many distractions or noise around the participants, as this might also impact on your results.

If you've selected the participants properly, you'll have a mix of individuals with varied skills and experience for your evaluation. Before you start the test, give them an overview of the process and what they can do to help you get the best information. Be honest with your instructions. Assure the participants that it is not they who are being evaluated, but the site. And most importantly, ask them to "think aloud".

The method you've designed for the evaluation will be influenced by the goals you've defined for the test. However, it's important to build in the flexibility to deviate from the anticipated procedure, in case the participants supply useful information that wasn't planned for during the course of the test. Responses will be more insightful when the evaluation is run like a conversation rather than an interview or test.

At times, there are some nitty-gritties that can be missed out during the test: it's hard to moderate and collect feedback at the same time. So it's best to have at least one person observe the test to note down the responses of the participant in each situation. Make sure the observer clearly documents the problems faced by the participants, and also the solutions (if they suggest any). If you have a Webcam, try to capture the movement of your participants while they perform the task they're assigned.

Don't lead the participants: let them vocalize their thoughts and lead you through their experience. Avoid any phrase that can influence their views on the product. Ensure that the observer records all the changes in facial expressions or body language the participant makes as they perform the task. More often than not, these expressions are more revealing than the participants' verbal communication.

Evaluating and Reporting your Findings

Identify the key comments, phrases, problems and expressions of the participants at each stage so that you can make a critical analysis of the evaluation. Clearly differentiate between your observations and the verbal response of the participant. This could make all the difference in the way you analyze the report.

Base the conclusion of your report on the usability shortcomings of the Website rather than focusing the attention on how the individual participant performed the task he or she was assigned. Mention the task-list, participant's profile, and the version of the Website being tested (with a screen-shot, if possible). After you compile the report, share the experiences and findings with the team so that they understand the strong points and drawbacks of each evaluated feature.

And once you feel that the feedback from the test has been incorporated in the latest version of the site, perform another evaluation before uploading the Website, and continue to test once the site's been launched. Remember -- testing is about trial and error. Each evaluation provides new challenges, but each amendment delivers a more useable site.