RESEARCH METHOD

29 Desirability Testing

When there is disagreement about which design direction to pursue, desirability testing shifts the conversation from which design is “best” to which design elicits the optimal emotional response from users.

First impressions matter, and within seconds of being introduced to a product, people will make judgments about it. Most of the snap judgments are based on how design elements make people feel, and designers know that interface elements that trigger an emotional response are difficult for non-designers to identify and articulate. But there is a method designed to explore this emotional space—desirability testing. Desirability testing goes beyond helping teams to simply identify the “best” or “most popular” aesthetic design direction. Instead, it explores the effective response that different designs elicit from people, so that the team can focus design efforts on shaping the exact emotional response they want people to have while using their products.

Desirability testing provides people a way to identify and articulate how a design makes them feel. It accomplishes this by providing participants with a range of positive, neutral, and negative adjectives that help them to tell the story of their experience¹ using simple, handheld tools—index cards with adjectives written on them. To begin, write each adjective/descriptive phrase on its own index card, and place all of the cards randomly on a table. Show participants a prototype mock-up, and ask them to pick the 3, 4, or 5 adjectives that best describes how they feel about the design. Record their selections, and ask the participant to talk about what each card means to them as it relates to the design.

When this process is applied repeatedly with twenty-five or more participants per user segment, the team can begin to compare the words that are most frequently chosen, and explore the groupings of positive, neutral, and negative word clusters. There are multiple ways to visualize the results,² and you can continue to refine and retest the design prototypes until there are enough responses that elicit the intended emotional responses.

The method can be conducted using low-fidelity prototypes, or on existing products already in the public domain as a baseline before the team embarks on a redesign. It can also be used to explore the emotional responses people have to competitor websites.³ If there are too many strong and varied opinions on your multidisciplinary team about the direction a design should go, help everyone refocus their energies on identifying what emotions they want the product to arouse in people. When used this way, the method becomes a helpful consensus-making tool that focuses the team’s attention on actual responses from end users, instead of on personal opinions and preferences that often leave teams at an impasse.

1. Desirability testing was first developed at Microsoft and documented by Joey Benedek and Trish Miner in their UPA 2002 paper “Measuring Desirability: New Methods for Measuring Desirability in the Usability Lab Setting.” The adjectives and phrases they used to run their studies were chosen from market research, prior user research, and team brainstorming, and were selected to align with specific project goals.

Barnum, Carol M., and Laura A. Palmer. “More Than a Feeling: Understanding the Desirability Factor in User Experience.” Proceedings of CHI 2010 (2010): 4703–4715.

2. See note 1 (Barnum and Palmer) above.

3. Hawley, Michael. “Rapid Desirability Testing: A Case Study,” 2010, www.uxmatters.com.

4. Microsoft allows for free use of the cards with the following disclaimer: Developed by and © 2002 Microsoft Corporation. All rights reserved. Permission granted for use.

Further Reading

Williams, Don, Gavin Kelly, Lisa Anderson, Naomi Zavislak, Dennis Wixon, and August de los Reyes. “MSN9: New User-Centered Desirability Methods Produce Compelling Visual Design.” Proceedings of CHI 2004 (2004): 959–974