Today there is much discussion about the limitations of usability evaluations that focus on the cognitive dimensions of user experience. Usability supposedly is relevant only to goal-directed tasks whose outcomes can be judged relatively objectively. In contrast, other kinds of interactive experiences aim to achieve a subjective state, often construed as affective, such as engagement, arousal, and pleasure. However, cognition and affect are not as independent as they may seem, and therefore usability and other aspects of experience overlap significantly. In this forum, I suggest that incorporating aspects of play experiences may help solve a pressing problem in user experience design: the need to draw people into discovering value in new features, starting from a point where, if they even know the feature exists, they do not yet have a clear vision of what it will do for them.
Although it has face validity, the distinction between cognitive usability and the subjective, affective aspects of user experience is overly simplistic. For example, attention is usually thought of as a cognitive variable, but fun certainly involves attentional phenomena. People having fun tend to feel as if they are responding to something that intrinsically draws their attention. In contrast, when directing and maintaining attention requires disproportionate effort, we are rarely having fun. Even in games that challenge the user's ability to maintain vigilance, the level of vigilance required must be sustained by the arousal the game creates.
Curiosity, the desire to know more, seems like a hybrid of emotion and cognition. When some stimulus arouses us and orients our attention, it leads us to explore more thoroughly. If we hear a twig snap in the brush beyond the edge of our campsite, alertness, arousal, and fear occur spontaneously, along with a question, such as, "Was that a bear or my partner returning from the latrine?" We may consider this primarily an affective experience because, at the outset, arousal and attention are automatic, making the experience seem spontaneous. But it includes perceptual and cognitive processes, such as distinguishing the sound of the twig from the background noise and consciously directing attention in an effort to assess the situation.
Sometimes, even the sensory pleasure that some designed experiences deliver can be intertwined with cognition. In shopping for a car, the pleasure we get from the supple leather seats is meaningful because it suggests that other aspects of the car will live up to the impression of quality. We may confirm this impression by testing the feel of the door locks, window controls, and gear shift. The smooth but solid feel of the gear shift does not just give sensory pleasure, but also reduces ambiguity about its position.
While affect and cognition are entangled throughout, behavior at the beginning of these examples seems relatively spontaneous (alerting to the twig snapping, feeling the leather, etc.), and becomes more specifically goal-directed and planned as the person becomes more engaged. This progression looks a lot like play, especially play that someone is tempted into and then becomes hooked on. Like millions of others, I began playing Angry Birds by casually "checking it out," but soon I was consciously trying to plan the trajectories of the birds to hit what my experimentation led me to believe were weak spots in the pigs' defenses.
Play and learning are intertwined. Just as a kitten at play carries out a built-in program that causes it to learn hunting skills, human play propels us on a pathway toward consolidating new skills and knowledge. This process is exploited in the educational movement known as constructivism, which uses physical and mental engagement and experimentation as the basis for conceptual learning. A great constructivist teacher tries to entice learners into a playful interaction that leads to learning, beginning with and building on their existing experiences and mental constructs.
Years ago, a brilliant clarinet teacher gave me a great example of this. Clarinetists live in fear of "squeaks," the loud, high-pitched squawking sounds the instrument makes at the worst possible moments. Once, when I flinched after a squeak, my teacher took away all of the instrument except the mouthpiece and upper section, leaving me with no keys. He then challenged me to learn to play the remainder as if it were a trumpet, after showing me that this was possible. At first, I managed only to get out a few raucous sounds. By the end of the week, though, I could play simple bugle calls on it. I had learned the trick of "overblowing," which is how a trumpet or flute can play different notes with the same fingering. For each note, there is a sound spectrum of higher overtones. In the flute, the octave is the strongest overtone, making it easiest to reach. I learned that the clarinet does the same thing, except its strongest overtones are at intervals of a twelfth, making its accidental leaps particularly startling. I could apply my new skill to avoid a high note's awkward fingering by overblowing the appropriate lower note. Meanwhile, my tone improved, squeaking greatly decreased, and concepts about the instrument's acoustics became much more real to me.
What were the key ingredients of this process? First, I had a general motivation to increase my mastery. Next, my teacher intervened at a moment of ripeness, just after I squeaked. He created a challenge for me that aroused my curiosity and focused my attention. The element of surprise and absurdity made it sound like fun. By demonstrating, he showed me that the goal was attainable and gave me a model to imitate. Finally, he enabled me to enter a process of playful experimentation, without risk.
The process of engagement often requires users to begin interacting with something before they can envision how it will ultimately benefit them. They may form a detailed idea of how they will exploit new capabilities only after a period of casual discovery, exploration, and experimentation. This means we should think about how the process of engagement begins.
I tend to think about the process of engagement with a product or feature in terms of the following stages:
- Awareness: Acquiring the notion that something of potential benefit might exist
- Discovery: Locating a feature and recognizing its potential relevance to some particular needs or goals
- Exploration: Attempting to dig deeper, to confirm the impression that it could be beneficial in principle
- Experimentation: Verifying that the capabilities of the product are relevant, by applying them to some personally meaningful task
- Adoption: Continuing usage of the feature that progressively incorporates it into the user's "standard" practices
Of course, you may conceptualize stages of engagement differently. For example, many usability professionals might group discovery and exploration together. Also, the progression is not strictly linear. Experimentation may lead to discovery of more facets of the technology, generating more exploration and experimentation. Finally, some stages may not apply in all cases.
The goal is to remove cognitive obstacles to progressing through the stages. But since we cannot rely on a user's conscious purpose to drive them through the earlier stages, we need to make the entire pathway enticing, rather than simply making work during the final stage easier to execute. Unfortunately, product teams sometimes load software with features in the hope of delivering value to hypothetical already-committed users, giving too little attention to the challenge of leading them into a process that will result in commitment.
This is relevant even in a domain as classically "task oriented" as business applications. It is misleading to think that user experience issues beyond efficiency, learnability, and accuracy are unimportant in business applications because their use is non-optional. While usage for routine administrative tasks may be obligatory, usage of many "value add" capabilities by knowledge workers, who have discretion about how they work, is optional. Lacking adoption, their usability and usefulness are like the sound of one hand clapping, a concept without a real manifestation.
However, as potentially valuable capabilities are continually added to software, more things compete for the user's attention, features are harder to locate, and the higher-order functions they enable become more abstract and harder to envision using. Thus the challenge of enticement and engagement increases, and we need more active strategies to draw users in.
Rather than trying to make applications look superficially like entertainment or games, we need to think more deeply and less literally about the underlying characteristics of play that make it enticing.
We need to go beyond crude approaches, such as trying to cram a link to every feature into the limited real estate of a home screen. Enticing users into playing with new features may help address this need. This may sound obvious, and many product teams may object that they already try various things to do this. However, too often these seem like add-ons, rather than being integrated into the experience.
Sometimes, product teams simply overlook the entry process. Often, the value of a proposed product or feature is promoted internally in optimistic scenarios portraying its benefits for hypothetical future users. These scenarios, typically written in the present tense, describe people who have already adopted the technology and are richly and successfully applying it in their jobs. Sometimes, these stories are repeated so often that people lose sight of the boundary between truth and fiction and forget that they depict the imagined end point, not how someone gets there.
The concept of the "early adopter" may divert attention from the design as a source of enticement. Often, early-adopter status seems to be construed as a deep personality variable, one that may even generalize across technologies. While every product will have some who adopt it before others, relying on the presumed internal motivation of hypothetical early adopters, or assuming that early adopters will start a viral process, may relieve product teams of too much responsibility for engagement.
Conventional usability evaluation tasks often start by assuming the user is already trying to do something with the software, and therefore skip or compress the first few steps of the process. It is difficult in the lab to study the process of experimentation, because the limited realism of a simulation makes it hard for users to apply newly discovered capabilities to things that matter personally. And the partly spontaneous process does not usually unfold in a 60- to 90-minute test session. Thus, we have to rely on opportunistic, indirect evidence, such as when a user departs from the task scenario because of curiosity about something else noticed on the screen.
Advertising certainly attempts to entice by communicating value propositions for the product that will attract users and motivate them through the early stages of engagement. However, depending on how well marketing and product development work together, the actual designed user experience may or may not be aligned with that value proposition. When there is a disconnect, this can create not only disappointed expectations, but confusion as well. A service and associated Web tool I once consulted on promised a highly personalized medical insurance signup process, for example, allowing different coverage levels and medical-provider choices in multiple specialties for each member of a family. Unfortunately, it presented users with a complete default set of choices as soon as they submitted identifying information. Personalization had to wait until 13 screens into the process, when users could "edit" these default choices (if they had not already dropped out of the process).
User assistance, such as tutorials and context-sensitive suggestions, are other attempts to make users aware of the potential value in new features before they have "tasted" them. Users often experience these as detours or distractions. Tutorials too often tell users what they could do, without giving them an idea of why and when they would want to do it. They rarely receive the same level of user experience attention as the main pieces of functionality themselves. This is unfortunate because there are plenty of user experience issues to address in these approaches.
The "gamification" movement is exploring the application of user experience lessons from games to other domains, such as education and business applications, to make applications more engaging or to motivate effort and productivity. However, sometimes these approaches seem like surface add-ons to the experience, rather than ways to exploit users' intrinsic interest to draw them in. Points, competition, leader boards, and game-like visuals may represent too literal and superficial an application of game design to these other domains. Recently, Google introduced a "News Badges" feature, in which people can earn points based on how many items they read on a given topic. I may be proven wrong, but I question whether points would add to the motivation of someone who is already interested in news, or whether anyone not interested in the news would care about earning news badges.
Another problem is that levels of difficulty are much easier to define in the world of games than in productivity and business tools, making it easier to design a fairly standard progression of experiences ("leveling") along which people can be lured. Productivity tools and business tools need to support a wide range of pathways based on the specifics of users' job design, tasks, and goals. It is also notoriously difficult even to determine what functions belong under "advanced," because this is so specific to users' roles and task patterns.
Rather than trying to make applications look superficially like entertainment or games, we need to think more deeply and less literally about the underlying characteristics of play that make it enticing. Here are some principles that may help draw users onto a pathway of exploration and adoption:
- Seek ways to arouse users' interest in new features with personally relevant information and examples, provided in context, rather than by explaining functional capabilities hypothetically or didactically.
- However, don't exaggerate intelligent software's ability to provide personalization or accurately targeted, in-context information. Users will be more forgiving of "misses" if they are presented tentatively.
- Exploit the power of imitation, for example, by using social computing tools to show related examples of value from peers, or by repurposing input from the user to provide demonstrations of new capabilities.
- Evaluate the expectations created by marketing messages, and then compare the early user experience to these. This can mean beginning usability evaluations with an assessment of the expectations aroused by marketing messages or home-page text.
- In storyboarding scenarios, conceptualize usage as evolving through a sequence of stages, making sure the steps are chained together plausibly. Design to support multiple pathways into more comprehensive usage.
- Provide a safe place for users to experiment in applying new capabilities to their own personally relevant information and tasks, without fear of losing work, contaminating data, or getting stuck in a blind alley of steps that have to be undone.
- While leveling may be difficult, some aspects of it may be transferrable. Use any sign of progressive engagement by users to show them that more is possible. Offer indications of additional potential value in small increments. Provide feedback to the user regarding their mastery or progress in expanding their toolkit over time.
As I mentioned earlier, many product teams may feel they are already trying to apply principles like these, but often the efforts are not integrated into the overall experience. These principles don't provide easy answers and their application will vary in different contexts. We also have to avoid hubris. Sometimes the obstacle to adoption is not lack of awareness of the feature, but that its claimed benefits are not meaningful to users, it does not deliver the promised benefits, or its cost in usability hassles is too high. Finally, just as my clarinet teacher needed to deeply connect with my goal of becoming a better player to have a chance of engaging me in a new, transformative experience, enticing users to discover, explore, and experiment with new features raises the stakes for deeply connecting with their motivations and mindsets.
David Siegel, vice president of Dray & Associates, is a user centered design consultant who has contributed his research skills to a wide range of technologies and to all phases of product development. He uses methods ranging from contextual field research to laboratory evaluation. He holds a Ph.D. in psychology from UCLA.
©2012 ACM 1072-5220/12/0700 $10.00
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 © 2012 ACM, Inc.