User Experience Is All There Is

1.1 Joy of Use?

But let’s not get ahead of things. Back in 2002, HCI was very much research-oriented. Thus, the knowledge provided by Ergonomics and Human Factors, which was based on the psychology of cognition and a quantitative approach to research, resonated well. Most practitioners worked in usability departments located in research, just as Michael and Marc. Small specialized consultancies, such as User Interface Design (UID) already existed (UID was cofounded by Franz), but by and large, interaction design was rarely done by designers but by engineers. Usability experts in turn stood by, analyzed and tested to make sure that the technology was halfway usable, that is, effective and efficient. Fun, beauty, pleasure was considered out of scope. And if somebody pondered about “joy of use”, it was quickly dismissed as either some odd and superfluous version of good old user satisfaction or as too esoteric since we cannot seem to measure it objectively. After all, isn’t an efficiently working system the crown of pleasure and elegance?

Michael und Marc doubted all this for quite a while already. Marc, for example, remembers vividly his first ever scientific talk in Dresden, 1997. The conference had been the German conference on Human-Computer Interaction, when it was still called “Software Ergonomie” (software ergonomics, today it is the “Mensch und Computer”). While Marc presented his own work on quantifying(!) the severity of usability problems [39], he was most intrigued by a keynote. Bob Glass, Director of Strategic Technology at Sun Microsystems, talked about future challenges and opportunities for usability professionals. Glass stated: “I predict that Joy of Use will become an important factor in product development and product success” and that “[N]ew metrics and techniques to measure joy of use will need to be developed” [21, p. 18]. And elsewhere he said: “If you’re still talking about ease of use then you’re behind. It is all about the joy of use. Ease of use has become a given – it’s assumed that your product will work” [29]. Glass’ keynote had the title “swept away in a sea of evolution”. Marc was certain: he neither wanted to be behind nor to be swept away by evolution.

This and a number of other lucky circumstances spurned the first scientific attempts to get a grip on elusive concepts beyond usability, such as “joy” and “beauty”. Michael and Marc first met at Siemens, both working in a competence center for user interface design. Normally, they organized focus groups and site visits to elicit “user requirements”, designed prototypes of interaction concepts and ran usability tests to further refine systems. Yet, these methods left blank spaces.

To give an example: In 1998, Siemens Medical started to developed Syngo, a new, unified interface for medical imaging. In fact, while Siemens’ CT scanners regularly won prices for their industrial design, the user interface was a lame, standard OSF/Motif based affair. Not unusable, but quite ugly, already drawing unfavorable comparisons to shareware by customers. Axel Platz, the designer in our department, was charged with rethinking the interface. And he did, complete with a new information architecture, novel widgets, and a special color scheme (see Figure 2).

Figure 2

Siemens Syngo.Via. Source: Siemens Press, www.siemens.com/press, reference number: HIM200911014-03.

Today, this interface is Siemens’ standard and it still looks modern and timeless. In 1999 though, it had been a bit of a challenge. Its flat design made it difficult to see what was clickable and what not. Novel widgets, such as a “dog ear” to browse quickly through stacks of images, were elegant but often unprecedented. Classic usability testing would reveal a lot of problems, think for example lack of conformity with user expectations, while at the same time aspects, such as the beauty and innovativeness of the interface, would remain unacknowledged.

1.2 Understanding, Measuring and Designing the Hedonic

Unhappy with this, we sat together. We first collected attributes of a usable system, such as “simple versus complicated” or “predictable versus unpredictable”. We then added all the aspects we found important but that were not yet covered by ISO norms and standard questionnaires, such as “stylish versus tacky” or “innovative versus conservative”. The result was the first AttrakDiff questionnaire, an instrument still quite widely used even nowadays (see [18]). Its idea was simple: there is pragmatic quality, a system’s instrumentality, that is, its perceived ability to support task-fulfillment. And there is hedonic quality, a system’s perceived ability to fulfill self-oriented goals, such as being stimulated (i. e., through novelty) or impressing others. An initial study showed that both quality perceptions substantially differ and that both equally contribute to the appeal of the interactive system [23], [38]. In hindsight, this was quite a breakthrough. We now had a simple model that more or less clarified “beyond usability” in terms of a product quality and an instrument to actually measure it. This was not only helpful for consumer-oriented applications, but also in more serious contexts, such as the already mentioned medical systems, or in automation (see [29]).

Yet, something was missing. While we had a model and a first draft of a measurement instrument, we lacked design competencies. While we knew quite well how to make a system usable, we knew less about how to make it hedonic. To learn more, we started an internal research project at Siemens called ATTRACTIVE. The idea was to identify approaches to design for the hedonic to make interactive systems more appealing. One approach was to take a look at game design. At that time, we all had been inspired by a cool experiment: Dennis Chao [12] designed a UNIX system administration tool, which made use of the first-person shooter game Doom. In UNIX, system processes are spawned automatically and may need to be stopped manually, that is, they need to be “killed” by the administrator. Instead of using a simple list as an interface to control the processes, psDOOM presented them as monsters (Figure 3).

Figure 3

psDoom. Source: [12].

Shooting a process monster lowered its priority and eventually “killed” the process. Administrators loved this game-like approach to their daily work. Only their managers complained that the system blurs the difference between work and play. Oh my, this is really bad.

Inspired by this, Alard Weisscher designed the interface of an energy distribution tool at Siemens based on the metaphors of the game StarCraft (see Figure 4).

Figure 4

(left) StarCraft; (right) Energy distribution. Source: [74].

It featured a control panel to direct troops (maintenance staff) on a battle field (energy distribution area) with possibilities to communicate via radio equipment (telephone connection with the maintenance staff) [5]. While its visual style was adapted to the serious tasks of process control, Alard’s design borrowed many interactions from the game. Marc later summarized what can be learned from computer games into three categories: how to support learning in an unobtrusive way, how to support immersion through details, and how to motivate use through providing a compelling story [24]. Especially the latter became crucial to the notion of experience as a meaningful and enjoyable “story of use”.

This research into alternative approaches to the design of interactive systems was badly needed. Marc still remembers a discussion with a manager of Siemens Mobile about the design of their mobile phones. Of course, a phone needed to look cool (industrial design) and to be packed with technological innovations (cameras, more memory, WAP etc.). But what would be an example of hedonic quality in interaction or in the interface? Challenged by this question, the manager showed a splash screen featuring a clown, if Marc remembers correctly, absorbed in – yes – juggling. Admittedly, this is different, this is new, and maybe even something to boast about to your friends (at least once) – it might qualify as hedonic. Yet, it felt wrong, light years away from the designs, for example, Axel Platz was providing or Alard Weisscher was exploring.

1.3 And Finally an Emotional Turn

While our work on the hedonic certainly progressed, the field, and especially the more computer science-oriented HCI people, remained skeptical about the role of fun, enjoyment and beauty in human-computer interaction. For example, in a seminal paper Noam Tractinsky and colleagues [71] claimed that “What is beautiful is usable”. They found a correlation between subjectively perceived usability and judgments of beauty. In fact, this is not surprising. Some attributes, such as symmetry or adhering to basic Gestalt laws, let an interface appear more beautiful and more usable at the same time. But the stance that this is now it – problem of beauty solved – had been symptomatic for the time. In the same vein, bearers of the ISO 9241 part 11 standard never grew tired of pointing out that enjoyment already was a part of the respective norm, they just chose to call it satisfaction. These tendencies to disregard the need for an expanded concept of usability and to oversimplify complex phenomena, such as emotions and beauty, shows how constricted the view of many experts had been at that time. Interestingly, lay people would easily agree that there is at least a difference in intensity between enjoying something and just being satisfied with it. In general, they would not hesitate to acknowledge the importance of emotions and hedonics in choosing and using interactive products. Usability Experts, though, had their doubts, despite the mounting evidence.

All this changed tremendously in 2004. It was the year, Don Norman published his new book “Emotional Design. Why we love (or hate) everyday things” [63]. For him, it was certainly a 180 degree turn, from a distinctive cognitive engineering perspective to an emotional design perspective on interactive systems. While most of the book was just borrowed from other people doing the actual groundbreaking work (see the book’s epilogue), it made the hedonic, beauty, emotions, joy acceptable to many. As we all know, gurus rarely err. In a way, this book was a clear sign of the times: the more ephemeral side of HCI was here to stay.

2.1 From “Usability Plus” to Psychological Needs and Wellbeing

As already said, our first approach, more than 20 years ago, was to complement usability with a further quality: the hedonic. In this sense, early approaches followed the logic of “usability plus”. Typical concepts from consumer psychology, such as curiosity and novelty or self-presentation and symbolic value, were just added. In a first model, Marc identified four sources for enjoyment [25], [27]: manipulation, stimulation, identification and evocation. The first mapped onto the pragmatic (i. e., perceived usability), in the sense that good usability supports the fulfillment of a human need to be active in the world, to change and control it through tools. The other three mapped onto the hedonic. While a first good start, this simple model seemed limited, mainly because it was formulated from a perspective, where usability was prime and other aspects just complemented it. Something was wrong with its priorities.

Help came from “old friends”, namely psychological theories of action, such as Activity Theory [60] or Self-Regulation Theory [11]. In line with usability, those theories assume that action is driven by goals. However, these goals can be on different hierarchical levels. For example, when preparing a kaffeeklatsch for friends, one needs to do a number of things, such as baking cake, making coffee, or laying the table. In Carver and Scheier’s terminology, these are do-goals. Each do-goal in turn affords a number of concrete operations, such as opening tins with coffee powder, switching on coffee machines, or stacking plates. These are motor-goals, which result from the higher-level do-goals. In a way, their importance can only be understood by putting them in the context of the higher-level do-goals. Marc often calls these levels the What and the How of human-computer interaction [26].

However, baking a cake, putting it on a nice plate, and placing it in the center of a beautifully laid table in turn can only be understood by referring to further higher-order goals – the Why. Carver and Scheier assume that do-goals are motivated by higher-level be-goals. In the example, this might be a need to feel related to other significant people, to experience social exchange, to try out a new cake recipe, or to impress friends or relatives. Following the same logic as before, the meaning of given do-goals cannot be understood without knowing the higher-level be-goals, which motivate them. To give a further example: In a recent study, Eva Lenz, Sarah Diefenbach and Marc Hassenzahl [58] made detailed hierarchical tasks analyses (HTA, Stanton [70]) of people preparing coffee. The result of a HTA are hierarchies of do-goals and sub-ordinate motor-goals. Eva, Sarah and Marc extended this, however, by an interview about the overarching meaning of the activity (i. e., “Why preparing coffee? Why is it meaningful?”, be-goals) and by letting participants identify the personally most meaningful do- and motor-goals (i. e., “Which parts of the activity would significantly impact its personal meaning, if they would be excluded?”). In line with our reasoning, the meaningful do- and motor-goals differed depending on the prominent be-goals. For example, participants, who enjoy preparing coffee as a ritual, consider all steps meaningful, while participants, who enjoy preparing the “best” coffee, focus on all steps important for controlling the quality of the coffee. Participants, who enjoy preparing coffee as a social event, focus on steps, such as selecting beautiful cups and pouring the coffee.

What follows from this? In fact, we argued that the hedonic is not on the same level as the pragmatic or simply complementing it. Instead, there is a hierarchical relationship with the hedonic actually motivating the action, that is being the Why of technology use, and the pragmatic simply making sure that the overarching hedonic goals can be met. In this view, usability (i. e., the pragmatic) is merely a hygiene factor. It makes sure that the system is in principle able to fulfill certain do- and motor-goals. The fun, meaning and enjoyment in turn stems from the be-goals and actually motivates use. In other words, an interactive system, which does not address any be-goals will not be used, no matter how great the usability. In contrast, an interactive system, which addresses important be-goals may become really frustrating only, if bad usability creates an unsurmountable barrier to its use. A user experience perspective (as opposed to a mere usability perspective) takes both into account, but places its emphasis on the Why. We even argued that we should design experiences first, and only if the intended experience is meaningful and/or enjoyable, start to think about how to shape it through interaction with technology [28].

If one takes this claim to design “experiences before things” seriously, we need to better understand different sources of enjoyment and meaning. In fact, most action theories only describe processes not content. For example, Carver and Scheier defined the function of be-goals, but remained relatively vague about the different types of be-goals that may exist. Here we used the work by Sheldon and colleagues [68] to fill this gap. They collected 10 different psychological needs, such as a need for competence or for autonomy, and asked people to indicate how intensely the need had been fulfilled in a positive life experience. It turned out that the more positive the experience, the more intense was the reported need fulfillment. In addition, some needs showed stronger relationships to positivity than others. Marc Hassenzahl, Sarah Diefenbach and Anja Göritz [33] replicated this study for experiences with interactive systems. Not surprisingly, the results had been comparable: the fulfillment of certain psychological needs was strongly related to the intensity of positive experiences. In fact, a particular set of needs proved to be a good starting point for thinking about what makes an experience positive: autonomy, competence, relatedness, stimulation, popularity, security and (added later) physical striving.

It is beyond this paper to further detail this approach to User Experience (see [26], [27], [34] for more details). What should be highlighted though is the close link between this notion of User Experience and general ideas of psychological wellbeing. Already in 2010, Marc emphasized that a focus on experience will make people happier [26]. In general, the fulfillment of psychological needs is related to wellbeing [65]. In this sense, designing experiences, which fulfill psychological needs can be understood as design for wellbeing [32]. Nowadays, the focus on the positive, the possibilities of technology to add to wellbeing [15], is part of a number of progressive approaches to the design of technology, such as Positive Design [16] or Positive Computing [10].

2.2 Usability Versus User Experience in a Nutshell

With the initial distinction between pragmatic and hedonic quality, its extension into a needs fulfillment model, and its explicit linking to wellbeing, the concept and, thus, our understanding of positive user experience evolved significantly. It still included issues of usability, but only as sub-element of successful user experience. However, while the theory evolved, and studies and many design cases demonstrated the viability of this approach, making it fruitful for practitioners remained a challenge. To overcome this, Michael started the project Design4Xperience in 2014. An important first step was to better understand the degree of user experience-related knowledge and skills in software developing small and medium enterprises (SMEs). In one study, Michael and colleagues asked the management staff of 172 SMEs about their knowledge of usability and user experience [52]. Eighty-seven percent of the managers had heard the term usability and 73 % the term user experience. They were then asked to define usability and user experience as they know it. The definitions were assessed by experts against established definitions of usability and user experience based on ISO standard and the literature. The expert rating revealed an average knowledge of 1.7 points (on a five-point scale) for usability and .8 for user experience. Many given definitions for user experience were just synonyms for “user experience” or the German translation “Benutzererlebnis” (while “Nutzungserlebnis” might have been more adequate). Emotions were rarely part of the definition and psychological needs or comparable concepts were never mentioned. In the same year, Kaisa Väänänen-Vainio-Mattila and colleagues [73] published a literature analysis of scientific studies on ubiquitous computing and the understanding of user experience in those papers. The striking result was that user experience was still understood as usability, general acceptance or preference. If hedonic aspects were mentioned at all, their description was rather superficial (like “fun” or “trust”) and only marginally related to available models of hedonics or user experience. These studies and ongoing discussions with practitioners and researchers show that user experience is still mostly understood as “usability plus x” – plus some beauty, plus some “fun”, ..., plus something.

So, what is the difference between usability and user experience [7]? Table 1 summarizes differences from two perspectives: design intention and design process. The intention of usability is to design interactive systems in a way so that users can achieve their do-goals with effectiveness and efficiency. The focus is on intuitive use, which can be defined as the unconscious application of the user’s existing knowledge (e. g. [61], [72]). The intention to design for positive user experience is quite different. Here, the focus is to design for the fulfillment of be-goals, that is, psychological needs. Some say that designing for needs is also the intention of usability. But here “need” is meant as a functional demand. In this sense, feeling close to one’s family is a psychological need, while the “need” to share family pictures in a cloud is rather a functional demand – a way to fulfill the psychological need. As mentioned above, the fulfillment of psychological needs provides experiences with a character, leads to positive emotions, and ultimately wellbeing. Thus, the ideal is that technology enables and shapes positive experiences. Note, that this not only includes enjoyable and outright pleasurable experiences, but also more complicated, so-called eudemonic experiences, which focus on self-actualization, long-term goals, as well as meaning in life.

The process of designing interactive systems can be roughly distinguished into three iterative phases [45], [57]. The first phase is the analysis of the design challenge. Based on this, first design solutions are created. Finally, designs are evaluated to reduce problems and to find further possibilities. Designing for good usability requires to analyze the context of use. Do-goals are strongly tied to the specific situation and tasks are very prominent components of the context of use. The main objective of the evaluation is to identify remaining barriers to task fulfillment (i. e., usability problems). This is because problems cause negative emotions and by reducing usability problems the negative emotions can be avoided. Very often usability professionals claim that reduction of negative emotions automatically leads to positive emotion and by that to positive user experience, but this is not the case. Reduction of negative experience foremost leads to neutral experiences. An exception may be relief (see [14]). Relief occurs when stress and discomfort is removed. But the cause for this positive emotion is the end of a negatively experienced situation. To use this in design would be like deliberately delaying trains just because it feels so good to be finally home after an odyssey. The analysis phase of the design process in experience design is different. It consists of understanding which psychological needs, positive practices [50] or positive experience categories [76] are and could be relevant in certain situations and contexts. This is necessary to create possible technologies to enable need fulfillment and by that positive experiences. In the evaluation phase it is still important to uncover design attributes that cause negative experiences, because their reduction is of course relevant. The central idea, however, is to find additional options to enhance the positive experience.

Differences between usability and user experience are plenty and substantial. All in all, the experiential approach focuses on wellbeing, enjoyable and meaningful experiences, as well as how technologies can create and shape these experiences.

3.1 Designing for Meaningful Work

Let’s start with the opportunities. Twenty years ago, user experience was a response to the fact that interactive systems became everyday objects. Obviously, consumer-oriented interactive systems, such as mobile phones, followed the path of many other products, be it clothing, food, cars or furniture. Mere usability would not be enough to dazzle consumers, in the same way, mere pragmatics are not sufficient to explain how cars became beloved consumer items.

The close connection between consumer technology and user experience already had been a valid observation and point of critique almost 10 years ago [1]. At that time, there was only limited research on the user experience of workplaces available [22], [51], [67]. In fact, this changes only slowly, although our notion of user experience is perfectly applicable to the work domain as well.

For example, Matthias Laschke and colleagues [54] provided a compelling design case in the domain of computed tomography (CT) in cooperation with Siemens Healthineers. Instead of asking radiologists about their problems with current CT scanners, Matthias focused on moments, when they really enjoy their job. For the radiologist, meaningful work was closely related to social exchange, for example, in form of feedback from the referring physicians, who they report to, or in form of master-apprenticeship relations with younger colleagues. They were also quite proud of outstanding and interesting cases and diagnoses and collected them in scrap books or excel lists. While these practices were the gist of an enjoyable work day, the actual work system, that is, the software used to diagnose and report, did not in any way address this. To Siemens this was rather “informal work organization” and not so much CT scanning; yet, the radiologists did not separate both. As a consequence, an opportunity to fulfill the radiologists’ psychological needs was not taken up. A frequently used argument for missing out on this opportunity was that radiologists do not make the buying decision, but institutional buyers, who care more about efficiency and costs. In Matthias Laschke’s case this turned out to be a false argument. In fact, employers cared about the wellbeing of their valuable staff and seemed quite open to new ways of improving it. Thus, while meaningful work has many sources, the work equipment itself can contribute its share. This requires to think about technology in broad experiential terms rather than in narrow terms of tools to fulfill defined work goals.

Besides successful design cases, though, the notion of positive user experience at work appears difficult to implement. There seems a broad agreement that work is about productivity, effectiveness, and efficiency. We are often confronted with opinions, such as “workers should just do their job”. Meaning or enjoyment does not feature strongly. Even in the very early phase of research on emotions and human-computer interaction, there was the plain rejection of the notion of emotional experiences at work. For example, Erik Hollnagel, a researcher of the design of control rooms for power plants, claimed “Keep cool” at the HCl International conference 1999 in Munich. Work should be done in a rational way; emotions will just get in the way. This assumes a contradiction between “rational” work and positive emotional experiences at work.

Figure 5

Video of the construction process. Source: Benjamin Erk, Josephine Hofmann, and Nina Kraus (Hochschule der Medien).

Michael did not believe in such a contradiction [8]. He and his colleagues gathered positive experiences in different work contexts. In face-to-face and online “experience interviews” [77], participants were asked to tell stories about positive experiences at work, irrespective of whether it involved technology or not. From 349 positive experience stories, 17 experience categories could be derived, belonging to six different groups (see [76], for an overview). For example, the category “appreciation” is about moments in work, where the person itself experienced some esteeming feedback from colleagues. Based on this, Michael’s team developed a feature to express gratefulness to others [53]. It is basically a button allowing to send a “thank you” to another person along with a comment. This solution was evaluated in different work contexts. It generated positive experiences for the sender and the receiver alike. The data further showed it to be important to users that this exchange takes place exclusively between sender and recipient and is not made public. This is in stark contrast to concepts, such as social media and ‘likes’.

A further experience category, “finishing a task”, is about celebrating the completion of a task. Together with students, Michaels’ team integrated a respective design solution into a tool for constructing digital twins. With this tool, a machine can be run virtually as a digital copy of the real machine and be tested before it even exists. After completing this task or parts of it, the users can start a video of the construction phase of the machine model. The inspiration for this idea came from computer games, such Grand Theft Auto or Driver, which automatically create video summaries of especially great moments [24]. This makes it possible to re-experience all the challenges and successes of the construction process. This concept had been developed as a video prototype (Figure 5) and was presented to 12 engineering students in a study. With the valence method [4], 36 experiences could be recorded; 32 (89 %) were positive and four (11 %) negative. The negative experiences were minor usability problems, while the positive experiences were – as designed for – moments of pride and reflection. The summary was seen as a celebration of work done and found quite motivating for further tasks. Interestingly, the 3D construction tool Autodesk Fusion 360 has a feature called timeline. It is a list of all 3D operations done during a construction process. In principle, it would be possible to playback the construction process, but this feature is introduced as a tool to facilitate corrections in the 3D construction (a history) and not to celebrate a successful completion of a task.

In summary, positive experiences are an important part of the work and digital technologies can be designed to enable positive and meaningful experiences.

3.2 Designing for Wellbeing

Besides a stronger focus on work as an application domain for user experience, other opportunities emerge from our particular understanding of user experience. For example, the focus on wellbeing provides an interesting twist. While wellbeing – or let’s say happiness – is certainly the desire of most humans, research show that not many know how to actually realize it. For example, people engage in a multitude of short-term pleasurable, often self-abusive activities, many of them technology-mediated. They spend hours consuming social networks, just to feel depressed about their own, not so perfect lives. People lose themselves in games, stay awake in front of the TV, procrastinate or overwork, use cars instead of the bike. In early notions of user experience the general stance towards these issues had been quite clear: enjoyment is what people find enjoyable. Whether this particular type of enjoyment is good for them in the long run or does negatively affect others was out of scope. Even more: it was not in the place of the designer to decide, whether an experience is good or bad, moral or immoral.

In our approach to user experience this is different. While the sources of enjoyment and meaning are clear (i. e., the different needs), the ways to fulfill theses needs, the everyday practices, remain subject to design. In this sense, we design technology-mediated ways to fulfill psychological needs. This almost always implies change. To give an example, some years ago Marc was involved in a case study about commuting with the car [37]. Instead of asking commuters about problems of their daily commute, we gathered papers about when commuting is actually meaningful and enjoyable. Interestingly, many papers studied the negative effects of commuting, while only a few focused on potential positive sides. One positive aspect was that commuters, who enjoy their commute, frame the time in the car explicitly as me-time. Secluded from the world, they avoid social contact, enjoy the contemplating effect of driving, being idle and often just day dream. This can be thought of as a pattern of a wellbeing-driven commute. However, it is a pattern often challenged in daily life. Employers expect commuter to participate in meetings while in the car, spouses believe that the idle time of the commuter can be perfectly made use of by discussing the latest school problems of the children, and commuters themselves hook up their phones immediately when being in the car to remain reachable. In fact, none of those practices showed up in studies as especially happiness-inducing, yet people practice them regularly. It seems no coincidence that commuting is one of the least positive daily activities [48]. We can now simply design for the practices people already engage in, such as working in the car, being always reachable, using the time in the car to discuss difficult private issues, or we may imply a “better” practice, namely to explicitly enjoy the solitary, contemplative, and idle time offered by the commute. While of course we cannot force people into contemplation, an interactive commuting system can induce and support it, for example, by not showing the work calendar first thing in the car or by offering communication buffers instead of hyper-connectivity. What if the car takes your calls while you drive? All this would offer the commuter a possibility to change current practices, even if they appear already enjoyable. Maybe the car knows better, what its driver needs?

This emphasis on change opens up interesting application domains. One paramount domain is health. Whether in private or in work life, healthy practices are very important to avoid stress and burn-out. Thus, instead of designing technology in a way to further accelerate self-exploitation, just because people demand it, we may rather do the opposite: shape healthy work. An example are work calendars [31], [46]. In one project, we designed a calendar based on chronobiology. This calendar roughly “knows” the inner clock of people and is, thus, aware of suitable times for relaxation, as well as optimal phases for problem-solving, being creative, or regeneration. If a user wants to insert an appointment, s/he has to categorize the activity first. If the time doesn’t match the type of activity, the calendar protests and suggest to move the activity to a more “biologically suitable” time. The calendar knows, what is good for its user.

3.3 Designing for Sustainability and Transformation

Another interesting and quite relevant domain with a focus on change is sustainability. While climate change progresses, it becomes more and more obvious that we cannot solve the problem of resource consumption and social aspects of sustainability on an efficiency basis alone. While we try to make cars as efficient as possible, excessive mobility is likely to be the consequence (a so-called rebound effect), leading to even more resource consumption. Interactive technology can be of use here in at least two different ways. First, it can provide viable alternatives to replace current, resource-intensive practices with more sustainable ones. Video-conferencing is an example of a technology able to save many resource-intensive plane and car journeys. However, to be accepted as a true alternative, the technology needs to be designed with experience in mind, not as a mere infrastructure to distribute video. This is not really done yet. For example, many people giving talks or lectures via a conferencing software, such as Zoom, complain about speaking into a black hole. And in fact, there is no real way to get an impression about an audience and its responses in any way comparable to a face-to-face situation. While in presence, a speaker easily gets an almost ambient impression of the size and the mood of the audience, Zoom speaks in numbers and participant lists to the speaker, often hovering in front of her slides. However, this is not a given. Zoom could be designed differently as a viable experiential alternative. One could even think about how to provide speakers and the audience with experiences better than a real talk or a real lecture. In a way, each technology offers its own superpowers in wait to be used in experience design [66], [75]. Instead, video conferencing is approached as a primary technical problem, an infrastructure, without any clear understanding of desirable experiential outcomes of its use. As a consequence, we will keep on flying and driving to distant places, just to have the experience Zoom fails to deliver.

Second, interactive technology can directly help to change behavior and attitude towards more sustainability. Eco-feedback systems are just one example of interactive technology with the main purpose to instill change. Here, the interactivity is especially potent. Systems can proactively provide users with new perspectives on their resource consumption, make them re-think or even break habits. An example is Matthias Laschke’s Key Moment [36], [55]. This is an interactive key holder mounted to the wall, maybe, in the hallway. It features two hooks: one for the car key and one for the bike key. If one takes the bike key, nothing much happens; if one takes the car key, though, the key holder drops the bike key to the floor. This is an unmistakable prompt to reconsider the current choice of mobility, delivered by quite a simple, yet surprising design. While interesting, this type of interactive technology still challenges most people’s understanding of what technology should be: practical, serving, and if possible beautiful. The experiential view substantially broadens the potential roles technology may play in the future.

3.4 Designing Otherware

Besides opportunities, there are future challenges for an experiential perspective. Recently, a particular type of technology is on the rise, such as robots, self-driving cars, voice assistants, chatbots, or other artificial intelligence-infused systems. These interactive systems are quite different from the systems, we normally use in everyday life, such as computers or smartphones. Voice assistants and alike present themselves as rather autonomous, proactive, complex and sometimes even anthropomorphized (or zoomorphized) counterparts. They listen, they speak and they answer to names, such as Alexa or Pepper. Let’s call such systems “otherware” [30].

Why are these systems different? Conventional technology typically feels like an extension of one’s self. For example, the computer mouse comes across as an extension of your hand, directly manipulating “things”, such as folders on a desktop. While the objects remain virtual, they are nevertheless directly manipulated. Direct manipulation [69] is the hallmark of good interaction design. It is a body of design principles, which guides the design of windows-systems, but also tangible computing or interaction in virtual reality. It is certainly responsible for the immense proliferation of computers in everyday life. Direct manipulation succeeded command line interfaces, which required conversing with the computer rather than directly using it. With otherware, we are back to the command line. Instead of using Siri or Alexa, we now have to converse and to delegate. In the future, we might need to cooperate with machines in a way we normally cooperate with other people. And many roboticists work on social robots, meant to become companions, such as a pet, to overcome loneliness or boredom.

Of course, there are many practical, aesthetical, legal and ethical challenges related to this. A full discussion is beyond the scope of the present paper. Focusing on experience, however, these technologies create a problem. Experience, as we understand it, requires activity and agency. To feel competent, requires to master something, not to let it be mastered by “somebody” – even if this somebody is a machine pretending to be a being. Holger Klapperich, for example, showed how fragile experience actually is. In one study, he and his colleagues [49] compared the enjoyment and meaning derived from grounding coffee beans with a standard electric mill, a hand-mill with a crank, and a custom-made electrical mill operated by a crank. Using the standard electric mill was perceived as less enjoyable and meaningful compared to the electric mill fitted with a crank, although both had the same underlying technology. Just the interaction was different. This also works in the context of driving automation [19]. While mills and automated cars are not necessarily otherware, this nevertheless shows how important agency is – there needs to be a feeling of being in control.

For example, Michael was engaged in a project with Siemens to explore how artificial intelligence (AI) could contribute to positive experiences at work. He and his team designed Wizard of Oz prototypes of an AI-extended workplace for knowledge workers [9]. In the study, 40 participants had to plan a creativity workshop. The simulated AI had knowledge of workshop moderation, methods and structures. Two prototypes were developed. One version, named “Hugo” contained social elements to instill feelings of creative cooperation, such as giving feedback on the progress of the planning document. Another version, had no name and no social elements, and remained rather in a “neutral” form. Both systems offered speech interaction and some on-screen information. Interestingly, while Hugo was perceived as more natural, supportive, helpful and positive compared to the neutral prototype, participants wanted more control over the system. In other words, while framing a technology as a counterpart triggers a positive social evaluation, such as finding the other helpful or supportive, it reduces agency and, thus, the potential to experience, for example, competence. People want more control to regain this, which in turn counteracts promises of more autonomy and automation on behalf of the technology. We all know how unnerving it is to meticulously instruct other people to do things. Often, the feeling is that it would have been better to just do it on one’s own. However, if we delegate to or even cooperate with a clever counterpart, it becomes increasingly difficult to claim resulting successes. In an unpublished vignette study, Marc asked people to imagine writing an important report. In all three scenarios, the outcome was positive. The superior explicitly praised the quality of the report. However, in one condition participants had to imagine doing it just with the tools available on an office computer. In another condition, an AI-system made some suggestions, and in a third condition, the system compiled a report to approve. Unsurprisingly, the meaning of and satisfaction gained from report writing differed dramatically. Although participants had always been praised, it dropped from a highly positive experience (when just doing it without AI) to a mere neutral experience (when approving was all there was left). Of course, otherware may offer alternative experiences, such as experiences of relatedness, sociability and closeness, normally reserved for fellow humans or selected animals, or plain stimulation. However, other important needs, such as competence and autonomy, are in danger. From an experiential perspective on technology, we still lack a profound understanding of how to best design practices, which include technology framed as other.

3.5 Becoming Accepted

While the recent years inspired a huge amount of models, methods, case studies and empirical explorations of positive user experience, this work does not seem to have the impact on the way we think about and design technology, it should have. It’s a strange situation. There is a lot of interest in the design for positive user experience. The work group “The Positive X” (established in 2018) of the German Usability and User Experience Professionals Association (German UPA) on design for positive user experience and design for wellbeing features a list of 56 (August 2021) interested in collaboration. This group conducted workshops at the conference “Mensch und Computer” in 2019 and 2020, which were completely booked out. In addition, there is great interest in the offers of the Competence Center for Usability and User Experience (www.kompetenzzentrum-usability.digital) funded by the German Ministry of Economic Affairs and Energy (BMWi). For example, the coordinating partner Stuttgart Media University (HdM) carried out 15 pilot projects between 2018 and 2021; nine of these focused on a positive user experience. Marc ran a funded Leitmarkt project on “Design for Wellbeing” (http://www.design-for-wellbeing.org/), working with partners, such as Braun or Lufthansa, to pilot a wellbeing-driven approach to their product design. Both efforts have been quite successful.

However, while interest is high, positive design approaches are implemented only rarely. The Competence Center for Usability and User Experience initiated a study of the practical application of user experience knowledge, methods and implementation in products or services. The study is still ongoing, but initial data shows that there is little application and implementation of “real” user experience. In one of the interviews a manager of a UX department in a medium-sized company stated that he had to meet the requirements of the customers and that they do not request a positive user experience. He further added that there should be a debate about it in the society. The German UPA working group and the competence center analyzed the discussions and initial data of the study mentioned and identified three fields of challenge [41]: companies, UX professionals, and customers.

To design for positive user experience, companies have to adapt their design and development processes. In many cases, design for positive experiences is treated as an additional requirement to already existing requirements, for example, when designing software for workplaces, the task comes first and only then design for meaningfulness is considered. This additional work has to be integrated into the development process and in the financial calculation of a project. If this is the first project integrating positive user experience, the design and development team need an appropriate training. All that requires additional effort and resources, even if the companies are interested. This limits the interest in positive experience to quite saturated markets, where this could make a difference.

The hesitation to change development processes has an impact on UX professionals. They suffer from a lack of knowledge and methods as well as insufficient project budgets and time management. In addition, Michael Burmester and Magdalena Laib [6] found several obstacles in the design practice itself. One central problem is that many designers understand design for positive experience still as an add-on. Especially when designing for workplaces, further benefits are needed as an argument, such as that positive experiences lead to more motivation, more productivity or save money. To design for enjoyable or meaningful work seems to be no value in itself – at least from the perspective of the management. In addition, designers are still very much problem-driven. When running design workshops, for example, we introduce the theoretical background, provide an overview of methods, and present design studies. After this, we ask participants to imagine a possible design for positive experiences in a given domain. Very often, however, the designers restrict themselves to ideas to make a product more efficient or to solve an overly pragmatic problem. There are three reasons for that. First, designers are traditionally trained in finding solutions for problems to reduce negative experiences [15]. Second, there seems to be still the implicit idea, that users will be happy just about removing negative experiences. Third, the analysis of opportunities for positive experiences in specific domains is often quite superficial. For example, when designing for “relatedness”, designers simply suggest the integration of a communication channel, such as a chat. The idea is that given the functionality, people will decide themselves about how to best make use of it to fulfill their needs. Of course, a chat can be a helpful, but is it exactly what is needed here? Think of a couple in a long-distance relationship: feeling close to each other over the distance is a complex and delicate emotional matter, not simply addressed by providing a chat (see [35]). In other words, many designers (and their companies) have no clear understanding about the positive experiences their products are supposed to provide. They focus on form and function and leave the experience to their users.

Finally, there is the question of whether customers/users want experience-oriented design? In this case it is important to differentiate between Business to Consumer (B2C) and Business to Business (B2B). Concerning B2C a study of Sarah Diefenbach and Marc Hassenzahl [17] showed that a decision for a product with pragmatic quality, which “solves” an apparent “problem”, is much easier to justify, compared to one, which offers beauty or novel ways of feeling close to each other. Unfortunately, this preference for the pragmatic in choice does not transfer to later use. Here, hedonic aspects become more important. In a number of unpublished studies, for example, Sarah and Marc showed that when buying a product, people are much more willing to pay for usability compared to beauty. However, when selling the product, they ask for substantially higher prices for beauty compared to usability. The experiential and hedonic creates a bond and makes everyday use much more pleasurable: a quality often only appreciated after product acquisition. To give an example: When asked in general about important criteria to choose the train, customers often mention price, speed and punctuality. On their actual train ride, they care for comfort, an interesting view, a good book and beer in the on-board restaurant.

In B2B context, the argument that positive experiences at work are good for the employee wellbeing does not seem to be widely accepted (but see the radiology case study above). Anyhow, in the survey [52], we asked SME managers which benefits they expect from design for positive user experience. The third most frequent answer was ‘to do something good for users’ (after ‘customer loyalty’ and ‘more turnover’). If we assume that this answer was not only due to social desirability, it could indicate that non-economic benefits can also be an argument in management decisions for design for wellbeing.