XVIII.2 March + April 2011
Page: 66
Digital Citation

Reflexive design thinking

Barton Friedland, Yutaka Yamauchi

back to top 

Over the years we have been involved in practices of applying methods and insights of human-computer interaction (HCI) for client engagements. Specifically, we emphasize a human-centered approach employing ethnography and participatory design that we collectively refer to as human-centered design practices. Yet we have become increasingly aware of a gap between what we do in our practice and what HCI offers us by inherently delimiting outcomes to technological solutions. Unlike research projects in HCI, where researchers have the luxury of exploring technologies they like, practitioners hired by clients need to focus on providing value to the clients regardless of any innovative technology they develop. Human-centered practices involve spending a lot of time inspiring and guiding members of the client organization, facilitating their reflection on their own practices, and thinking about alternative strategies and visions together. Technology has become only a small part of the story, and to this end, we advocate a research direction that extends human-centered design practice across a variety of domains.

To illustrate what we do in our practice, we describe one case:

One of the authors, Friedland, became involved with biomedical startup Beta, which had implemented an intranet for facilitating information sharing and collaboration. Beta's CTO explained to Friedland that the intranet was used in only a rudimentary way and that he saw a need for potential improvements. In response, Friedland proposed using ethnography so that the CTO could understand people and their needs instead of jumping immediately to developing new technologies. Ethnography revealed that integrated information management for the laboratory was the critical area for the organization. Through more than 30 interviews, Friedland constructed an integrated image of all the information for the lab from various operational groups. Beta then worked with Friedland to develop the software system to achieve the integration. The system reduced the time it took to set up, run, and analyze an experiment from more than 1,000 hours to less than 12 hours. Not only were more experiments conducted, but larger ones with larger data sets were also performed, which has had profound implications for the organization's business strategy. Yet the most significant change was not limited to the above-referenced work-efficiency improvement. Of greater value to the organization was the fact that people started collaborating more and shared more information across groups. This was enabled not by the technology per se, but by enabling people to design their own working environments. Beta thus began an ongoing cultural process in which different groups continuously experimented with alternative work processes together, which were continuously refined and improved.

This case provides a clear example of some of the ways in which human-centered practice can contribute to the development of organizational strategy, work practice, and technology, as well as positively affect the way people relate to one another and their workplace. The practices we employ affect not only the design of technology but also the organization's fundamental business strategies. Our observation is that regardless of the organizational application, the underlying practices of design are equivalent. A phenomenon we have observed is that various disciplines, ranging from organization theory and strategy to HCI, have inadvertently erected domain-specific boundaries while employing similar underlying practices. Researchers and practitioners alike would benefit from a forum in which the broader, more integrated style of practice we are describing here can be researched and discussed. However, the results we describe cannot be readily reported and discussed in the HCI community, where an interesting technology is a prerequisite.

We therefore assert that the notion of human-centered design practice would benefit from extending its scope: by putting more "human" in human-centered practices. These practices aim at transforming the culture of the organization, not just technologies. It is not enough to involve users in design. We have found that one key to transforming culture is learning on the part of an organization's members. One kind of learning in particular, we realized, was crucial: design thinking. This refers to a specific approach to problem solving and incorporates a unique epistemology recognizable in the practice of designers. Once individuals acquire design thinking, they take a more reflective stance in relation to their own work, experiment with different ways of seeing it, and continuously revise their work practices. We still think that technologies play a central role—not as an end in themselves, but instead as a means to facilitate continued human learning.

back to top  A New Research Direction

To begin, we need to understand what it means to "design" an organization. It is clearly different from what is meant by designing a technological artifact or a work process. Here the distinction between natural science and social science can be helpful. The goal of natural science is to identify the laws that govern nature, like Newton's law. Nature works according to these laws, although they are always open to falsification—e.g., Newton's law does not apply to quantum physics. On the other hand, social science seeks to understand rules that are normatively maintained. If we say to someone, "Hi," he or she should reply, "Hi." These are normative rules that we comply with but can also break. If somebody breaks the rule, he or she gets punished; e.g., a person gets mad and considers the person rude or perhaps reckless.

Normative rules are different from scientific laws because anybody who is socialized to the culture acts on them. In other words, people have reflexive understanding of the rules. To engage in the "design" of an organization means to address and reflect on these rules, whereas designing a technological artifact is done according to laws that correspond to the natural sciences. Because normative rules are reflexively understood and produced by people, any "design" of the rules necessarily involves the people they affect. Thus, no one can design an organization for someone else. The only thing that someone can do for another is to design representations and constraints such as formal roles, staffing, processes, and hierarchical structures. Designing an organization therefore involves other activities, such as supporting members of the organization in reflecting and bringing attention to the normative rules of their own organization.

Despite this important distinction, these two applications of design have much in common. There appears to be a basic way of framing the important elements of what it means to design. Scholars have named this design thinking and design attitude. Nigel Cross also used the phrase "designerly ways of knowing."

First of all, design embodies a distinct epistemology. Schön contrasted it with the dominant epistemology of technical rationality, which emphasizes rational problem solving based on theory and science [1]. On the other hand, actual professional practices in general and design practices in particular embody reflection-in-action. This means designers face each unique situation and, at the same time, see it as something familiar. Through this, they frame the situation and test successive frames by taking actions experimentally and reframe the situation based on what they learn. To many in management, this kind of practice appears unstable and subjective.

Second, in many modern workplaces, design is less emphasized. People are expected to analyze a problem and derive a solution to solve it. Solutions are often framed in terms of decision making, where an option among given options is chosen. Design is form-giving, the creation of what has not existed. As such, designers often take a solution-focused or abductive approach, by which they explore solutions and problems together and often use a potential solution to better understand the problem [2, 3]. Resulting designs can vary from situation to situation, from designer to designer. In contrast, an emphasis on analyzing a problem and rationally choosing the best solution occludes design possibilities.

Third, design is also a moral matter. One indication of this is that we label designs "good" and "bad". "Good" designs often have a coherent thought, idea, or concept running throughout the parts. Designers simultaneously pay attention to the whole while designing the parts. To guide the design, designers typically develop a key simple concept or metaphor that guides the design throughout. This concept is not given from the beginning but discovered and created as the design proceeds. When members of the organization design, this core concept becomes even more important because multiple people have varied goals, ideas, and personalities. In order for an organization to achieve a coherent design, its business strategy needs to be examined, revised, and implemented. Unfortunately, this is not what always happens in practice. Regardless, strategies are neither developed nor executed in abstract. In each instance of detailed parts (e.g., technical features, roles, communication, work processes, language, documents), strategies become salient, emerging over time, often in ways that are not intended, ultimately impacting interpersonal behavior.

Technologies play a crucial role in this learning, although, as we've emphasized, technologies are only a means, not an end. We therefore need to broaden the role of technology. Technology can be a tool for learning design thinking and facilitating cultural change.

Fourth, design involves power relations. Buchanan claimed that design is rhetorical, as "the art of conceiving and planning products," which are "vehicles of argument and persuasion about the desirable qualities of private and public life" [4]. We have experienced the process of design in an organization as a political process. One group tries to take more control by designing a process that favors them; other groups refuse. The power conflict renders the design inconsistent and therefore costly to implement and maintain. Much of this struggle around the placement of power and responsibility comes from the fact that people are often resistant to change, viewing the change from their existing personal perspective. Once members of the organization acquire design thinking and develop the discipline to see problems from fresh angles, many of these conflicts can be anticipated and addressed more directly.

Nonetheless, non-designers cannot acquire design thinking easily or without applicable real-world opportunities for practice. The biggest challenge for people in the acquisition of design thinking is their ingrained knowledge. As they acclimate to life in the organization, they may take many aspects of the work and the organization for granted. They typically do not think about alternative ways in which the work could be organized. For users, their work practices become a reality, and alternative work practices, particularly ones that contradict theirs, can represent a threat to that reality. For this reason, when we engage clients as practitioners, we emphasize members' self-reflection. We often use ethnography to represent people's own practices and "hold up a mirror" for them—which is a key reason that we've come to use video as a reflection device. We create short video podcasts to report ethnographic findings and help people to start reflecting on their own reality. Video conveys the details of the phenomena, e.g., work, interactions, conflicts, strategies and learning.

Those people who are doing the work are generally not professional designers. Their goal is not to design the work but to do the work. Furthermore, design thinking cannot be explained such that people can then be ready to practice design; practicing design differs from the theoretical understanding of what design is. Learning design thinking therefore requires actual participation in designing in which experts can "coach, not teach, learners," to use Schön's phrase. On one hand, design thinking is a general skill that everybody possesses—many people can write coherent, well-crafted prose, cook a creative new dish, or arrange a living room nicely. Yet not everyone can learn to become a competent designer.

Technologies play a crucial role in this learning, although, as we've emphasized, technologies are only a means, not an end. We therefore need to broaden the role of technology. Technology can be a tool for learning design thinking and facilitating cultural change. This is one area in which we see HCI as offering a somewhat limited purview. As an outgrowth of engineering, HCI tends to treat technologies as an end. In the case we presented earlier, members of the organization acquired design thinking experientially, initially by working with the practitioner and designing with simple intranet tools. Gradually, they started proposing new tools and new ways to organize the work. Instead of designing abstractions such as roles, communication paths, and strategies, they designed tangible artifacts. We believe this played an important role in helping them acquire design thinking.

Therefore, to help acquire design thinking, as practitioners we need to constantly challenge people's assumptions. In this relationship, we are always conscious about our relationship with the client. We do not take our relationship with them for granted and instead try to design the relationship as part of our project. Inherently we become part of the situation that we seek to change.

Power is also an issue here. Because we are hired by the organization to deliver some value, members of the organization tend to think of design as our job. They believe they can simply say what they want and we will design it. The new research program should take this reflexivity seriously. In most studies in HCI, computer-supported cooperative work (CSCW), and even participatory design, practitioners are seen as observers outside the situation. Much research is needed in this aspect of practice.

back to top  Research Agenda

While practitioners like us can talk about many folk theories as to how to help clients transform their organizations, we feel that more grounded theories to guide us in practice are needed. There is a lack of existing theories that could guide us through the clients' learning process. It is not clear in what process clients acquire such skills as design thinking while they initially have no idea and become defensive. Furthermore, while theories emphasize the political nature of organizational change, little guidance is available as to what specifically practitioners can do. We require theoretical insights more substantive than participatory design methods, such as future workshops, card sorting, and games.

For this reason we would like to call for research that is sensitive to the issues practitioners really face. We propose a research agenda that incorporates a broader organizational frame, taking into fuller account how organizations actually change (e.g., cultural change and power relations), and one that still emphasizes the valuable roles of technology, communication, and coordination. Just as HCI was highly successful in its goal of improving the relationship of a user to a computer by integrating multiple disciplines, this new research will also require many disciplines. Social scientists have a larger role to play here, one that's much more important than finding ways to make technology more useful: We also need to better understand the broader role of technology and its design. We need to understand not only usability and the usefulness of a technology for a certain work practice, but also the interactions the technology affords in transforming human conditions and learning.

back to top  References

1. Schön, D. A. The Reflective Practitioner: How Professionals Think in Action. Basic Books, New York, 1983.

2. Cross, N. Designerly Ways of Knowing. Springer, London, 2006.

3. Lawson, B. How Designers Think: The Design Process Demystified. 4th ed. Elsevier, Oxford, UK:, 2006.

4. Buchanan, R. Rhetoric, jumanism, and design. In R. Buchanan and V. Margolin, eds., Discovering Design: Exploration in Design Studies. The University of Chicago Press, Chicago, 1995.

back to top  Authors

Barton Friedland is managing partner of Luminous Group Consulting LLC. He is a veteran of Apple and NeXT, and was director of engineering for Spoken Translation. His research examines how people's beliefs shape their use of tools.

Yutaka Yamauchi is a senior lecturer at Kyoto University's Graduate School of Management. His ethnographic studies of technology-rich workplaces include IS design, troubleshooting complex technologies, and product design.

back to top  Footnotes

DOI: http://doi.acm.org/10.1145/1925820.1925835

back to top 

©2011 ACM  1072-5220/11/0300  $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 © 2011 ACM, Inc.

Post Comment

No Comments Found