David Siegel, Susan Dray
A company that we know has a small but active user-centered design (UCD) community and does a fair amount of both laboratory usability research and field research. Over the years, the company has increased the number of user-centered design professionals it employs, and it is a rare product that does not receive at least some attention from usability. Even during the technology downturn, staffing has been maintained, a far cry from the many companies that have significantly reduced staffing for UCD functions.
Sounds like UCD heaven, doesn't it? To an outsider, it certainly might seem like an example of corporate commitment that should be validating and gratifying to the UCD community at that company. Yet, the UCD staff had some serious concerns. These people felt that they had too little influence in their teams and that their career paths were limited. They complained about being paid less than developers who had less education and experience.
At a meeting with upper management they were told that management valued their contributions and that nothing inherently blocked their opportunities for advancement. However, they were also told that influence came from initiative. They were encouraged to act in a more empowered manner and to take the initiative in getting involved in projects.
UCD professionals who focus on doing "studies" as opposed to generating designs and products, will always be perceived as peripheral.
Should the UCD professionals see this as a sign of management commitment, endorsement, or mandate? Should they be reassured by management's expressed openness to considering them for promotional opportunities? Is it reasonable to expect that simply taking more initiative will make them more central?
In our opinion, initiative is only part of the solution. First, the initiative has to be truly directed toward contributions that will be perceived as of real value. But even beyond this, we argue that some of the perceived peripheral status of the UCD professions comes from the inherent dynamics of specialization within a corporate context. These dynamics can be exacerbated by paradoxical effects of the profession's own attempts to increase its influence. Frankly, we think that UCD professionals too often shoot themselves in their own feet, inadvertently behaving in ways that limit their influence or keep them on the periphery. The good news is that an understanding of the dynamics of specialization within organizations, and the ability to employ this understanding strategically, can help us be more effective in achieving increased influence for the concepts and approaches of UCD.
Unfortunately, the kinds of issues highlighted by our story are not rare these days. Usability and UCD professionals frequently struggle with what they perceive as their marginalization, although they rarely use this term. It exists even when demand for services like usability evaluation is increasing. Although this is certainly a type of endorsement, UCD professionals may still experience their role as limited and lacking in influence.
The classic form of this complaint is that UCD is too peripheral in the design process because teams request input too late for it to be effectively incorporated. A variation is the complaint that usability activities are not given a high enough priority in the competition for resources. After all, when was the last time that executives asked for cost justification for running yet another survey or for producing a television commercial? Also, UCD professionals often feel that they do not have the ear of upper management in the same way that, say, marketing does.
The complaints about career path limitation and about perceived pay inequity we described are yet another variation on the marginalization theme. Certainly, we do know of many people who have moved into UCD professions from backgrounds in development, technical writing, or fields like anthropology, sociology, and psychology, but we know of far fewer UCD professionals who have moved out of UCD into the mainstream management career ladders in their companies.
The concern about being marginalized and pigeonholed was very much in the air at CHI 2002, where several speakers addressed it during panel discussions.
We feel that we heard less of it at CHI 2003, but we suspect that may be because, in the continuing technology downturn, the issue had evolved to an even more drastic formnamely concern about having a job at all.
Most UCD professionals we know already view themselves as working very hard to increase their contribution, as well as to champion UCD, such as by promoting the importance of usability. Most see their efforts to get involved in projects earlier as an example of taking initiative toward making more of a contribution. However, this is probably not enough. UCD professionals who focus on (or are perceived as focusing on) doing "studies" and on critiquing others' designs, as opposed to generating designs and products, will always be perceived as peripheral or adjunctive. Even striving for iterative evaluation of early prototypes is probably not enough to let us emerge from this adjunctive role. There will still be little support for involving UCD earlier and according it more and broader influence until it is perceived as pointing toward solutions, or even better, new product ideas, rather than just identifying problems with existing solutions. Does this mean that UCD people will be less peripheral only when we can assume overall responsibility for designthat we essentially need to transform ourselves into designers and engineers? Not completely. However, we do need to function more like designers or engineers.
Product and system design is the result of a complex integration of many perspectives and considerations: business and market factors, technological possibilities and constraints, design constraints and trade-offs, and deep understanding of users and their interaction with the technology. No one can be equally expert in understanding all of these factors and representing them within the team. Furthermore, user research really is a separate set of skills and a distinct knowledge base. In short, there is plenty of work for many contributors to design, and there must be some structured division of labor. It is reasonable to ask why UCD should be of lower status simply because it is only one contribution to the final product.
However, we do think that we, as UCD professionals, can and should do much more to generate solutions. To do so requires thinking and acting more like designers and engineers. Successful solutions or viable innovations must take into account not only the user issues that we are the experts in identifying, but also the constraints that members of other disciplines on the team struggle with. Wrestling with trade-offs and constraints in the process of developing a solution is at the core of design and engineering, and if we want to increase our influence, we need to demonstrate that we can participate in this activity. This is something that the mere critic does not do.
We know many UCD professionals who do indeed take part at this level in generating solutions, but this is by no means the norm. In our conversations with fellow professionals, it does indeed often sound as though they primarily provide a critique. Many usability professionals tell us that their input to the team consists of cataloguing the things that "did not work." As for getting involved in solution generation, they often do not go beyond facilitating a team discussion to elicit possible solutions from others. Of course, it is important for the team to take ownership of the solutions, but it is risky for the UCD person to be perceived as identifying problems and then handing off responsibility to others to generate potential solutions and weigh their trade-offs. This is not a way to increase influence.
We also hear frustration about the self-limited role of usability professionals from new and potential clients. One of the most common sources of skepticism they tell us about is that they have had some prior experience with usability services, and all they got was a list of problems, with vague recommendations like "Fix X," or "Change Y," or "Find another way to let users do Z," but no guidance about how to do so.
We have also heard fellow UCD professionals objecting to calls for them to get involved in the broader range of activities that feed into design. Studying users to find out their real requirements? "That's not my job. That's the job of a business analyst." Working out a new layout for a Web page? "I'm not a designer." Yes, of course, there are differences in skills, and there are heavy demands on people's time, which can make the idea of getting involved more broadly seem overwhelming. It is understandable if someone chooses to focus on being an advanced specialist in usability evaluation and wants to be acknowledged for that skill. However, that person may have to accept that the consequence is a role as an individual contributor.
Much of what UCD professionals experience as pigeonholing is not unique to them; rather, it is a common phenomenon for specialists and professionals in organizations. Specialization is unavoidable as knowledge expands and as the complexity of work increases. Since this tendency always brings a risk of fragmentation, it increases the need for one specialty that focuses on coordination, that is, management. Almost all aspects of mainline managers' work lives differs from those of specialists, including their vantage points, mindsets, training, and incentives. Managers know this and tend to hear specialists as speaking from a local, rather than a big picture, perspective.
UCD professionals, can and should do much more to generate solutions. To do so requires thinking and acting more like designers and engineers.
Managers, especially at the executive level, can accomplish detailed work only through delegation, so when a new issue or priority arises, it is natural for executives to act on it by delegating responsibility. Delegation can take the form of creating a temporary assignment, setting up a task force or committee, adding a new position or even a specialized department, with a new mandate to take ownership of the issue.
Although this type of action often initially indicates corporate commitment to some issue, it has pitfalls. Frequently the issues that are delegated are "meta-issues," in that they have broad implications for the work of a large cross-section of the company and can be addressed only through coordinated changes in practice by many workers at multiple levels. UCD is clearly one such meta-issue. However, specialist departments typically have limited authority over anyone outside the department. They rarely are seen as being as central as those who do (or manage) the "actual" work. This is especially true when the "issue" is seen as influencing the bottom line but not directly determining it. Usability falls into this category, especially in the context of our continuing efforts to show the return on investment of our discipline.
The problem is that dealing with issues by delegating ownership of them to a separate category of specialists makes the issue the concern of the defined specialists, so everyone else can go on with business as usual. The specialized identity of the professional becomes a pigeonhole, defining that person or group as a separate part of the organization, even though the issue supposedly concerns everyone.
A good example is quality. When U.S. businesses jumped on the "total quality" bandwagon, many of them created total quality management (TQM) or continuous quality improvement (CQI) positions or departments. Their jobs were to establish quality programs and initiatives in their companies. The departments were staffed with specialized "quality consultants." They certainly understood that the new model called for all activities throughout the company to be organized according to a new paradigm, and their official role was to promote TQM throughout the company. However, unavoidably, delegating TQM to a specialized department resulted in the segregation of quality from everything else. Other functional departments participated in programs facilitated by the quality department, but these programs were generally adjuncts to their basic work. In order to gain some cooperation, the quality initiatives were determined not by what would make the most profound differences, but by what would be the easiest and create the least additional work. Variables were chosen on the basis of which would be the easiest to influence and measure, so that annual goals of completing a certain number of quality projects could be met. In short, paradoxically, the programs that were supposed to transform the companies had to be constructed in such a way that they would create the least interference with "business as usual." This is not a problem that TQM staff could solve no matter how effective the training classes they offered, no matter how successful isolated quality initiatives were, and no matter how vociferously they might complain about their insufficient influence over the basic processes of the company. It is a structural problem associated with specialization.
Complaining about lack of influence is the surest way to invite marginalization.
Not all specializations are equally likely to be pigeonholed, and this depends on the core identity of the company. In engineering companies, engineers are less likely to be pigeonholed; in health care organizations, doctors and nurses; in legal services firms, lawyers; in financial services firms, accounting and investment people. The perception is that they are steeped in the core issues, concerns, and basic processes and domain of the organization, as opposed to being the "owners" of specialized issues. However, we might be surprised by how often even these specialists express concerns similar to ours about limited influence.
The dynamics of specialization are pernicious. The way people respond to you can be determined automatically and unconsciously by your identification as a specialist. The limiting identification tends to attach to you automatically. Once it does, all your communication tends to be seen as reflecting your particular (and, by definition, limited) vantage point. Thus, anything that evokes or reinforces your specialist identity can exacerbate your pigeonholing.
Luckily, the problem is not hopeless. Conceptualizing the self-perceived marginalization of UCD professionals as simply another case of the basic dynamics of professional specialization within companies suggests more effective strategies for dealing with it. It is possible to learn to communicate and operate in ways that combat this limiting effect of specialization.
Don't Think It's a Personal Affront
First of all, we need to abandon the illusion that someone else has done this to us, that some powerful person has set up this marginalization by design, through ignorance, political motivation, or malevolence. Contrary to the beliefs of many corporate workers, managers tend not to experience themselves as powerful, but as tangled up in the many constraints they have to juggle. Overtly complaining about lack of influence is the surest way to invite marginalization. Managers are used to hearing this complaint from many interest groups. When you complain, you sound as if you are just competing with all the other groups managers have to coordinate, all of which want more influence. In doing so, you confirm how different your world view is from that of management and how incapable you are of seeing the big picture as clearly as they do. It is an essential management skill to be able to hear to complaints like this (and to respond diplomatically), without feeling a need to leap into action to take care of them.
Guard Against Appearing to Act in Parochial Ways
Paradoxically, attempting to promote the value of your discipline using verbal argument can reinforce the pigeonholing effect of your discipline label. It is all too easy to come across as though you are motivated by parochial self-interest or by allegiance to your discipline rather than to the larger goals of the company. Appearing to give primacy to your professional identification can unconsciously raise questions about your identification with the company, its processes, its culture, and so on. This is not just a matter of how you label yourself. It also comes up when the rationales for your arguments seem to come from a set of principles specific to your discipline.
Don't Expect Hearts and Minds to Change by Mandate
Most UCD professionals are passionate about the vision of UCD, but it is important to be careful about overtly pushing for a paradigm shift. Anyone who is seen as pushing for a fundamental change in the company's way of operating has an uphill battle and is likely to elicit resistance or even defensiveness. Most managers don't have the power or influence to produce fundamental change by fiat, even if you were able to convince them. Furthermore, paradigm shifts by fiat are rarely effective. Don't make this mistake of thinking that the mandate that created your specialized role should serve as a mandate to any other people to change their behavior. The best way to promote the value of your discipline is to demonstrate the beneficial impact on the end product your approaches allow you to make.
Identify with Larger Goals
Actively demonstrate your identification with concerns that exceed your discipline and with the company's larger goals. Whenever possible, speak a common language rather than the language of your discipline. All your recommendations should be clearly supported not by reference to the principles or belief system of your discipline, but by clear evidence of what is best for the company in a practical way. Consider reframing what UCD is about, in ways that show its links to larger, shared concernsdue diligence, product planning, quality, reduced risk, efficiency, and innovationand to marketing and business goals like customer satisfaction and retention, and, ultimately, profitability.
Share Your Turf
It is common for members of a specialty group, especially if they feel embattled, to band together defensively and try to protect their turf. This rarely helps to increase influence. If we want to combat the circumscribing effect of delegating UCD to a specialized subgroup, we need to support and encourage others in taking ownership of some of the practices and attitudes of UCD. We need to show them how they can do some of what we do and how that will help them achieve their goals. After all, aren't many people in companies supposed to be contributing in one way or another to a positive user experience? We also need to respect the methodologies of others that are complementary with ours and be prepared to recommend these methodologies when appropriate.
Support UCD Professionals Who Move into Mainline Management
Members of some disciplines tend to look favorably on those who endorse the field by making a career move into it but feel a sense of betrayal toward those who move out. Do not attack those who have made such career moves. When a colleague moves into some other part of the organization, it is important to support them and maintain alliances with them. Recognize that these people may have to go through a process of establishing credibility in their new reference groups, and this means that they can't show a bias toward their old group. If you complain that they "have gone over to the other side," you increase barriers and display your own sense of disempowerment.
Consider Carefully the Potential Risks of a Centralized Group
Organizationally, consider how you can move from a centralized usability or user-experience unit to a more decentralized model. In our experience, UCD people whose primary organizational "home" is within project or product teams are more likely to be able to join other disciplines in the difficult work of weighing trade-offs to create solutions. In one company that we know, the exception to the rule that usability people just identify problems is the one who is not primarily attached to the laboratory but to a business unit.
Find Ways to Grow Outside UCD
In terms of personal professional development, recognize that your expertise in UCD may not be sufficient for you to move into management. You may need to obtain additional training or education in finance, organizational design, project management, or product development. Consider making lateral moves to broaden your experience base. Do not give up if you apply for positions that represent advancement and are not selected. Sometimes going through this process repeatedly is a way to learn more about the "big picture" in the company, to educate others about how your skills apply to other functions, and to create the perception that you are a potential contender.
Of course, these suggestions won't work for someone who isn't willing to change. Sometimes when people complain about being marginalized they don't really want a more central role and the responsibility that goes with it. They want to stick to their familiar turf but at the same time want more recognition, power, and unlimited raises. They want to be loved more for who they are; they don't want to have to do anything different.
Unfortunately, the real world isn't like this. If you want to have more influence in your company, you will likely have to make some changes in how you act. Not everyone wants thisand that's fine. Just realize that you can't have it both ways. In most companies, it is very difficult to be a master of a particular craft and be considered part of the mainstream leadership. If you choose this route, be aware that the master craftsperson typically has a "niche" role and also that you may need to keep your resumé well tuned in case you find your specialty on the "downsize" list.
Susan Dray & David A. Siegel
Dray & Associates, Inc.
2007 Kenwood Parkway Minneapolis, MN 55405, USA
612-377-1980 fax: 617-377-0363
©2003 ACM 1072-5220/03/0900 $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 © 2003 ACM, Inc.