Searching for information is an emotional experience. We've all experienced confusion and frustration while searching the Web, as well as the occasional joy of discovery. But how do we conceive of emotions within search design?
A key message from recent investigations is that feelings affect thought and actions. To some degree, this is something interface designers have tacitly known all along; we've just not clearly articulated the role of feelings, or we've confused them with something else. For instance, the main thrust of Don't Make Me Think  is really, "don't make me frustrated." (Of course we want users to think.)
But even fields such as "experience design" seem to lack a clear place for emotive criticism and development. Therefore, I advocate formalizing and systematizing considerations of emotions in Web search design. We can begin by turning to existing theories of information seeking for guidance. In particular, Carol Kuhlthau's, Information Search Process (ISP) considers three levels of searching for information: actions, thoughts, and feelingsthe last setting her model apart from others .
Kuhlthau's ISP has six stages:
- Initiation: The user recognizes a gap in knowledge. Feelings of uncertainty and apprehension are common.
- Selection: The task is to identify the topic. Uncertainty often gives way to optimism.
- Exploration: Returning feelings of uncertainty and doubt are critical: If a focal point does not form, the search may be broken off.
- Formulation: Rising confidence and a sense of direction mark a turning point.
- Collection: Focus is clearer and confidence increases.
- Presentation: Upon completion, feelings of relief are common, as well as feelings of satisfaction. The user makes sense of what was learned.
A fundamental principle within Kuhlthau's model is uncertainty, which has the affective symptoms of anxiety and lack of confidence. Uncertainty in earlier search stages is often caused by the introduction of conflicting information. Moreover, uncertainty is frequently associated with perceived complexity. Both aspects are key stimuli that trigger emotions.
Kuhlthau's theoretical ISP serves as the framework for design. The objective is to create a tailor-made ISP, reflecting variations in user actions, thoughts, and feelings for a specific search context. There are six steps:
- Segment users and create profiles. An ISP only applies to a particular target group.
- Identify the search stages and user goals for each. Established phases (described above) serve as a starting point, but must be adapted.
- Note typical feelings, thoughts, and actions at the individual stages.
- Summarize each phase with a user requirement: What do they need to achieve their goal?
- Develop and arrange features to help users achieve their goals at each stage.
- Map business goals back to user goals and features.
This yields a clear representation of the search process from a user's perspective. Once completed, an ISP enjoys a great deal of longevity. It also scales to different situations and considers business needs.
The ISP framework is primarily intended to facilitate conceptual design, in the same vein as user personas and scenarios. In fact, it fits nicely within other user documentation, expanding and strengthening knowledge of user contexts. Articulating the search process can, for example, provide a common understanding across teams. The ISP can also guide and organize user research, allowing users to make comments about their feelings "on-the-side."
Matching the search interface to emotional ups and downs is the ultimate goal. But developing an ISP may not lead to "killer" features or a totally new, "emotionally charged" search experience. Subtle changes often make the difference between a negative or positive experience, though. In some cases this may mean simply reordering existing features, and by that, giving them a new rationale.
Reducing complexity at points of high uncertainty clearly emerges as a key tactic. This is not to say a minimalist approach need dominate the entire design. Instead, how and when design elements are presented is vital: Timing is essential.
Focusing on the "Exploration" stage, here are some preliminary techniques and examples that illustrate this approach:
- Remove unnecessary options and navigation on results pages, while allowing destination pages to be more complex.
- Repeat key elements to provide a sense of familiarity (e.g., the original search query).
- Exaggerate visual priorities to increase focus on key tasks.
- Vary page templates to communicate progress within the search process.
- Provide next steps and recommendations for assistance.
Consider the Search on BBCi (www.bbc.co.uk). Moving from a rather busy homepage to the results page, we get a different experienceone that matches typical seeking patterns. The layout of the results page and over-sized labels create a clear focus. Redundant elements and manual recommendations ("BBCi recommends") provide a safety net for users in doubt. Overall, a reduction of complexity acts to ease uncertainty.
Beyond avoiding uncertainty, try to foster optimism and increase interest. Take the travel portal Opodo (www.opodo.co.uk. See image). While waiting for results to appear, a screen provides more than mere system feedback: Images communicate a relaxed atmosphere. And, concrete information about the database and search status not only quells frustration, it offers a sense of beauty. ("Wow! This is huge and fast.")
Another example is the clippings feature on the International Herald Tribute Web site (www.iht.com). This supports browsing behaviors by allowing readers to skim headlines. (Click any "+" sign to add that title to the clipboard.) This site also varies page templates and reduces clutter at key points in the search process.
Consideration of emotions is not limited to visual appeal: Users also have emotions while interacting with information. When designing search interfaces, the challenge is to make affective considerations part of the design process. Raise "feelings" as a topic within your organization and allow decisions to be based on them.
Human Factors Engineer
©2004 ACM 1072-5220/04/0900 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.