Negotiating Constraints
Gerry Gaffney, Director, Information & Design, Australia

Arbitrary constraints can be particularly frustrating. For example, we may be told at the start of a project that the entry screen must be blue, that everything must be no more than three clicks from some starting point, or that a book can have no more than seven sections.

A client asked me to help redesign a form, but a key constraint was that it needed to be faxable and precisely two pages long. This requirement was embedded in the brief, but an examination of the form quickly revealed that although it used the mandatory two pages, the appearance and flow were so compromised by this physical constraint that it was unlikely that a usable version could be created within the same footprint.

It seemed appropriate to push back against this constraint. The first thing to find out was where it originated. There were two drivers – first, that the form could be easily copied and faxed, and second, that it would not appear daunting to users. It transpired that the first factor was less relevant than it had been in previous years, so it could essentially be put aside.

On the second factor, the project team I worked with was gracious enough to allow me to proceed with design proposals that stepped outside the two-page limit, but with a clear expectation that if the form was considered more daunting than the original version, I could expect to be in some trouble.

Another constraint was that the form should be designed to match the data entry system, which was used to store the information gathered. This requirement was problematic because the system was database-centric rather than user-centric, and a form designed to support this model would be inherently difficult to use. Luckily, the project team was willing to be swayed by an argument that an inconvenience suffered by a small number of data entry staff was a lesser evil than all users of the form suffering.

When the draft form emerged at six pages, I undertook usability testing with a good deal of trepidation. However, users invariably failed to comment on the length of the form and found that the simpler layout and improved flow (many questions applied to few circumstances) was easier. Error rates were also much lower than with the existing form.

Although it’s not always easy, I’ve found that in many cases a well-reasoned open discussion with clients can help remove arbitrary constraints.

What About the Answers?
Moshe Ingel, Usability Expert, Biosense Webster, Israel

During the process of questionnaire design, much of the effort is usually dedicated to the questions. No wonder: question design is a complicated effort that involves decisions regarding the right number, order, and terminology of each question. When thinking about the answers, most people might claim that the best method would be to give respondents a small, even number of options. Our long development process of a complicated medical device includes an external evaluation phase (EE). In an EE, we ask physicians to work with the new product on real cases (treat heart arrhythmia patients) to get feedback and inputs. After each procedure, we ask physicians to fill out a questionnaire to capture their attitudes toward the new product and its features.

In our last version, we had a long debate between the usability expert (me) and the marketing department regarding the number of points on the answer scale. Marketing insisted that four should be the right number; I claimed that nine points are adequate for that kind of questionnaire and users. Marketing had two arguments: one was regulatory (FDA) – we have always used a four-point scale for this kind of questionnaires, and if we change it now, we might need to provide explanations to the FDA about this change. We are not sure we can easily explain the reason. The second argument was about user comprehension and complexity level. Marketing argued that they do not want to confuse respondents with too many options. Fewer options are easier to understand and choose from. The international (United States/Israel) interaction with Marketing went back and forth over both email and phone conferences. Eventually, Marketing “won” this debate (mainly because traditionally EE is a marketing event) and the questionnaire had four-point scales.

During the EE, we noticed that many physicians were marking their answers between the points. For instance, 2.5, 3.5, or even 5. Because Marketing team members observed the physicians fill out the questionnaire, the message was loud and clear: four points are not enough to express the exact attitude toward the new product.

Following these EE events, Marketing (not happily) agreed that a four-point scale was too limited. In the three years since this incident, we have been using seven-point (or nine-point) scales and everybody is happy.

Choosing the Right Design
Eva kaniasty, UX Principal, Red Pill UX, LLC, United States

It was an exciting time in the project. We were about to start building a new website. The project involved several highly vocal stakeholders who wanted to be involved at every step and had many opinions about design. In the past, this situation had resulted in tension with the design team, as disagreements over design direction were not an uncommon occurrence. Starting work on a new project gave us the opportunity to start fresh.

The design sessions yielded three viable design approaches and three sketches illustrating them. The approaches differed fundamentally: we called them “the Traditional,” “the Fuzzy,” and “the Get to the Point.” If we let stakeholders decide, we knew they were likely to go with the Traditional – it was the familiar option that would appeal to business-oriented stakeholders. But maybe the more approachable Fuzzy was the answer, or maybe users just wanted to “Get to the Point”? It was clear the users should choose, but who has time to build three prototypes and test them? We certainly didn’t.

In the past, usability testing had proven effective in persuading stakeholders to fix usability problems in existing applications. But with a new product, we wanted to harness the persuasive power of usability now, not after time had been wasted fleshing out a potentially erroneous design direction. To the rescue: paper and Balsamiq. Balsamiq, a low-fidelity wireframing tool, allowed us to create three paper prototypes and test them with users in a matter of days. Our fears that a low-fidelity method might not be convincing to stakeholders proved unjustified. It was clear that users were able to engage with the prototypes and give realistic and useful feedback. Stakeholders could see for themselves that for this particular user scenario, the “Get to the Point” option was right.

Because we were able to bring the users’ voice into the process at an earlier date, the stakeholders felt convinced that the design direction was rooted in evidence, not just the opinion of the design team. The result? A less contentious design process overall and more stakeholder comfort with design decisions made by the team.

Watch my interview with Meena Kothandaraman, Adjunct Professor at Bentley University and a usability consultant. Meena argues that talking about research in the quantitative language makes stakeholders listen. Use QR code 127 to access the video, a quick summary of the interview, and Meena’s biography.

Watch my interview with Jeff Sauro, at Measuring Usability. Jeff says that when stakeholders tell him that five participants are not enough, he tells them they are right. He explains that these will be good only for identifying the obvious problems and that usually, there are many of those to work on. Use QR code 115 to access the video, a quick summary of the interview, and Jeff’s biography.

Persuading Large Companies that Five Is Enough
Hiroshi Ushioda, Head of User Experience; Susumu kuriyama, Interaction Designer; Reva Hamon, User Experience International Business manager, mitsue-links, japan

Our clients and potential clients, especially big companies, often ask us whether five users are sufficient. We have tried explaining the rationale and referencing Jakob Nielsen, but it usually is not persuasive to clients. In general, we believe we lack persuasive materials to convince customers of the value of lab usability testing with few users. But this year, we tried a new method of persuasion and were able to convince a large electronics distributor in Japan to do usability testing for their e-commerce site.

When making the bid for the client, we felt sure that the client wouldn’t be persuaded to do lab usability testing with a small number of participants, due to the large number of stakeholders that would be involved in the decision. So we decided to recommend a combination of quantitative research, in the form of remote unmoderated usability testing, and qualitative research, in the form of lab usability testing. Actually, we didn’t believe this to be the best plan for the client, because it was much more expensive than they had budgeted for. In spite of that, the project manager told us she loved the bid, due to the quantitative feature. In the end we won the bid, even though the cost of our estimate was by far the highest amongst the bids the client received.

When we finished reporting the results of both tests, the client was so highly satisfied that they decided to continue to do both remote usability testing, and lab usability testing periodically. Many stakeholders, including the company CEO, came to observe lab usability testing, and after watching, they understood the value of qualitative research, even with only a few users. Through the process, the client became a more sophisticated customer and came to understand the advantages and disadvantages of both methods.

In this case, we learned that combining quantitative and qualitative research is a good strategy to use with large customers. As of yet, combining methods is still a lot more expensive than just lab usability testing, so this strategy might not work for clients with smaller budgets. In addition, with smaller clients, there are often fewer decision makers, so it may not be necessary. However, the bigger the company, the more internal decision makers there are in the company. So it is quite difficult to persuade all of them of the value of qualitative research.

Managing Focus, Scope, and Participation
Gregg Almquist, Principal, Experient Interactive & Design LLC, United States

While I was at H&R Block working on digital consumer tax products, we created multiple cross-functional teams that were responsible for improving an existing feature or creating a new feature. The cross-functional teams gave us a diverse perspective, and team members across the development organization were able to participate in the design of the product, which built strong team dynamics.

These teams always included a QA tester. Most of them had worked in the H&R Block offices, and they’d seen every tax scenario imaginable. This experience could be really useful as an anecdotal source of information. Occasionally, some of the QA members would propose extensive coverage of just about every scenario they’d ever encountered in the office – even those that were relatively obscure. Once when we were designing a topic, a QA person brought up a scenario we didn’t cover extensively. It turned out it was a scenario that had occurred once in this person’s many years at H&R Block.

I can’t remember the exact scenario, but it was something really obscure – like the child had been raised in a forest by wolves for half the year. I needed to acknowledge the issue – and the contribution – but explain at the same time that if we handled every scenario like this in depth, not only would the majority of the users be heavily overburdened with a more complex experience, but also we would never be able to release an annual tax product. We had to make trade-offs and choices that ultimately helped the majority of our users, the product, and the business. Bringing the focus to these objective success criteria helped move the conversation forward.

The user experience specialists would often use this strategy, and slowly team members from the other cross-functional groups internalized this point and would raise it when issues of scope creep arose.

In Search of the Usable Creative Director
Michael Summers, Principal, SUMMERS Consulting LLC, United States

Earlier in my career, I worked at a large agency whose UX team was led by a very talented creative director (CD). Although he enjoyed mentoring younger team members, we always knew that when it came down to a deadline, if the team he’d put together didn’t have a really solid comp to bring to the client meeting, he could go into his office for a couple of hours and come out with something great.

I was often in conflict with this guy, but I have to admit that he had fantastic instincts for information design and creating layouts that were scannable and concise. His designs made it easy for users to visually discover which elements of the page or control were needed, and in what sequence – all while juggling the limitations of making sure relevant information was contiguous within a single scroll. He’d walk into meetings with a first draft that was often 80 percent of the way there.

However, the guy just could not deal with the idea that we needed to run real users on designs and then iterate based on the data.

The first few months we worked together, we did a business financing website. It had some complex forms and equities research tools. He put out a fantastic first draft and was then frustrated when I insisted on testing it – with the client observing. We discovered some serious problems involving how users selected individual equities, and pages where users did complex customizations, only to fail to save or enter the changes they had meant to apply.

This CD – we’ll call him Greg – went to great lengths not to see what everyone else was seeing through the one-way mirror. On day one, he said the users were the wrong age. So a hasty call to the recruiter resulted in younger traders for day two – same problems. The blocking and denial went on through additional days of research.

I wish Greg’s story had a happy ending. Despite more than 40 users in that first study and very tightly edited video highlights of more than 30 users having the exact same problems with that first design, he only grudgingly made any revisions and remained defensive.

He refused to attend the actual findings presentations with the client. As a result, the client had to act as the mediator, taking our findings, working with his team to demand revisions, ferrying them back to us to see if we felt they addressed the problems, and so on. Even when we did follow-up testing on revised designs, he sent his junior folks – if he sent anyone at all.

No doubt you’ll go to usability conferences and hear many tactics that are supposed to woo skeptical stakeholders to the lab, like free pizza, heavy use of video, and others. But in Greg’s case, he never had the big “a-ha!” moment.

His team would continue to crank out designs that were much stronger than industry average – say, 80 percent of the way there. We would continue to force the issue and have real users interact with the designs and discover important things they’d missed. Clients would continue to observe and realize that if the problems hadn’t been caught, they’d lose a lot of money. So they’d lobby for changes. Changes would come grudgingly, and we’d start the process all over again.

I went through it with Greg for over two years. When I moved on to my next job, we shook hands somewhat stiffly and muttered some well wishes. He never really changed his fundamental guiding assumption – which is that UI ought to be created by someone with a fine arts background and that what I was doing was “science” and could potentially kill “the creative process.”

I walked away with a stack of reports, video highlights of major problems his first drafts had missed, and notes from clients saying what I was doing was essential.

I guess you can look at it like a Saturday morning cartoon about the “separation of powers.” We were like the legislative and executive branches – meant to pull at each other and compromise. But I must say the process isn’t super fun. I doubt Greg will ever change. And I know I won’t!

Watch my interview with Rolf Molich, owner and manager at DialogDesign, Denmark. Rolf has a strong principle. He doesn’t take assignments on which he does not agree with what the potential client wants to do. Use QR code 125 to access the video, a quick summary of the interview, and Rolf’s biography.

Watch my interview with Whitney Hess, a user experience designer, writer, and consultant from New York. Whitney says that if stakeholders ask to skip research, she understands that they are interested in user interface design, not product strategy and user experience. In this case, she points them to industry UI standards and chooses not to work with them. Use QR code 128 to access the video, a quick summary of the interview, and Whitney’s biography.

Watch my interview with Jared Spool, CEO and founding principal at UIE (User Interface Engineering). Jared claims that there’s no point in working with unengaged stakeholders. He suggests that you find better ones. Use QR code 116 to access the video, a quick summary of the interview, and Jared’s biography.

More Research as a Persuasion Technique
Richard Buttiglieri, Director of Design, Pearson, United States

Years ago, I was designing a system for call center reps to make referrals to sales reps in other departments within a company. These reps were to get monetary incentives for making the initial referrals as well as if the referral turned into a sale. While conducting field research in order to study the user’s workflow, I discovered a number of key factors that would lead to a successful referral, one of which was to support an ongoing dialog between referrer and recipient in the context of a referral. The referrer could learn over time what makes a higher-quality referral likely to purchase and thus receive more money with less effort. This ongoing dialog was the cornerstone of long-term usage of the system because call center reps had very little time between calls for this “other” work.

When designing the interface, I laid out the dialog area in reverse chronological order, as that best supported each user’s workflow – to focus only on the most recent response. This was also consistent with other applications the users worked with. The product manager, however, had a different idea: to put the dialog in chronological order because that was how they were most used to seeing a dialog. They fought for “consistency” without appreciating a higher design principle – use the natural order of objects appropriate for the target user.

I tried convincing them with the data I discovered in field research and with a usability test with internal participants that were a close analog for the actual target user. Unfortunately, this did not convince the product manager to overturn his ruling. Because this seemingly small part of the system was so pivotal in encouraging long-term usage, I decided I needed to go bigger – essentially over engineering UX research with several focus groups plus individual usability tests of actual end users comparing both designs (in a counterbalanced way, of course). I collected feedback from nearly 50 end users in total and made sure the product manager attended some of the sessions. Only after witnessing the head-to-head comparison with actual end users did the product manager agree that my first design was correct for this situation.

In our field, UCD can mean “user-compromised design,” as designers must make design compromises in the name of progress. But when that compromise could potentially put the product’s success at risk, it is worth the effort to collect extra data and involve your stakeholders to help make your case.

Focus on Projects that Produce Tangible Results
Gregg Almquist, Principal, Experient Interactive & Design LLC, United States

In one of my earliest roles as a user experience specialist, I was often asked to weigh in on projects that were outside the scope of my responsibilities. I believed these opportunities were good for our group’s exposure, so I would accept them when time permitted.

One day I got a call to attend a meeting about redesigning the customer support website. Call volume was climbing every year, and the added support costs were negatively impacting the profit and loss (P&L). The company decided to redesign the site in hopes that customers would use this self-service option versus making a call.

The team redesigning the site had no user experience background, but they had a firm plan to move forward, including an information architecture schema. I immediately recognized some basic information architecture and navigation issues, so I weighed in. However, it was clear that my ideas met some resistance because it didn’t align with the group’s direction.

I decided that our team could knock this project out of the park if we were allowed to lead the design. I contacted an ally of user experience design in the company and explained what I believed we could accomplish if given the opportunity. I laid out a plan – two card sorts to understand how users think about and categorize their problems, a usability study to validate a prototype design, and a review and rewrite of the content. Together we hammered out a budget to hire some contract UX designers, and she got us in front of the project sponsor. The sponsor gave approval, and we were off.

The project was successful beyond our expectations. At the end of the first three months, call volume had decreased significantly, as visits to the support site increased dramatically. Additionally, page views per visit were down, and customer satisfaction with the support website increased nearly 5 percent, which indicated that users were finding answers to their questions with fewer clicks.

This project cemented the credibility of the user experience team’s activities and its reputation as a valuable resource for the business.

Show, Don’t Tell
Lior Yair, CEO, and vitaly mijiritsky, UX Designer, Netcraft, Israel

Our agency was hired to redesign the “My Account” area on the website of one of the largest service providers in Israel, serving well over a third of the local population. Business intelligence data indicated that a very large portion of the requests handled by representatives at overcrowded service centers and overloaded call centers could be solved quickly and efficiently via well-designed self-service online. Our job was to make the interface efficient and enjoyable to use, maximizing the usage of available online services.

We cooperated closely with our client’s project manager, coming up with a design that both sides believed in. However, when the design was presented to the executives, they didn’t approve it. They felt that it was cumbersome and unclear and that users wouldn’t be able to use it well. They suggested a few key changes of their own, undoing our layout and information architecture principles. Our citing of various references from professional literature and similar websites was to no avail, and the final verdict was that it wasn’t enough of an improvement on the current interface, and they weren’t going to invest significant resources in something that offered “superficial changes” and provided “marginal benefit at best.”

Because we were also the ones who were supposed to develop the new interface, this was a big blow for us. After assessing the situation, we decided to go all-in. At our own expense, we developed fully interactive prototypes of both designs, set up our usability lab, recruited participants, and invited the executives to observe the testing with us behind a one-way mirror. We performed a full eye-tracking study, complete with user interviews and retrospective think-aloud sessions. When we saw the executives completely engaged in the process, watching the users’ progress, rooting for their own design as if this was a ball game, and listening to the users voice their dilemmas; we knew we had made the right call. Normally, we would also perform statistical analysis of the findings, but in this case it wasn’t needed – the executives saw the users handling our design with ease, completing all their tasks quickly and with few mistakes. Their design, on the other hand, brought about a lot of guesswork and frustration, with some participants giving up halfway on several tasks.

A couple of days later, the order came in for the graphical design and the development of our original design.

Watch the fascinating interview with Chris St. Hilaire, author of 27 Powers of Persuasion, from Los Angeles. Chris defines persuasion as the creation of consensus from conflict or indifference. When done appropriately, persuasion can move mountains. Use QR code 134 to access the video, a quick summary of the interview, and Chris’s biography.

Watch my interview with Takashi Sasaki, Partner at Infield Design, Japan. Takashi-san says that their pitch to clients who ask to drop the research part of a design project includes a metaphor of a pilot asking to not use a navigation system. Use QR code 121 to access the video, a quick summary of the interview, and Takashi’s biography.

Usability Is Subtraction through Addition
Bill Albert, Director, Design and Usability Center, Bentley University, United States

Adding usability to the design and development process both saves time and money. This is counterintuitive to most people, specifically for folks who have not gone through usability testing before. How can you possibly save time by adding a step in the design/development process? The analogy I like to give is asking directions. Sometimes stopping to ask for directions takes a few extra minutes at the beginning but saves a lot of time in the long run. This is especially the case when driving around Boston! Usability testing and user experience research is really no different. Conducting user experience research and usability testing, when done well, clarifies the team’s understanding of the users, their current experiences, their needs/desires, and what improvements need to be made to the current design. Without this information, you are designing in the dark. How can a design team possibly know how to design something without this basic information? If they stumble on the right design, it is sheer dumb luck. Much more likely is that the design team comes up with something that isn’t right, and it doesn’t sit well with the rest of the team. There are countless meetings to come up with the right design, with a lot of guesswork and anecdotes being thrown around. Had the design team only done the upfront research and testing, they would have reached the right design much sooner and with a lot less pain. To some people, testing a product that is going to change doesn’t make sense. However, once they see the benefits, they never go back. The old adage “you can pay me now, or pay me later” rings true.

The Art of Selling Experience Design and Design Thinking Approaches
Silvia C. Zimmermann, Managing Director, Usability, Switzerland

When I started working in the field 17 years ago, I had to write business cases to convince potential clients to embrace Design Thinking approaches in their development life cycle. I had to present them with detailed cost and risk models to demonstrate that they most probably will engage in an unbearable risk when they don’t involve their users and clients in their product or service design processes early on. Today, I don’t need to do this any more because my long-term clients have achieved a similar maturity model when it comes to usability, UX, and experience design.

One thing that was always very important for me personally, though, was the promise I gave myself when I started working in the field. I don’t try to persuade someone at any price. I show them the methods we use, the success we had in other projects, and the importance of embracing UX, but if a client remains skeptical about the added value we can bring to their innovation cycles, then they should not embrace our methods and services. It is not the responsibility of our field to persuade someone that what we do could be beneficial for him or her. That would be the same as if we would try to persuade someone to stop smoking, if he or she likes to smoke.

Here are my five tips to sell Experience Design and Design Thinking approaches:

Watch my interview with Dana Chisnell, an independent researcher and consultant at UsabilityWorks. Dana says that when the bottom line changes for the worst, stakeholders become more open to UX research. Use QR code 119 to access the video, a quick summary of the interview, and Dana’s biography.

Words Matter
Carol Smith, Lead Consultant, Midwest Research, United States

To gain my client’s trust, I try to use clear language and terms and to incorporate the language that they are most comfortable using. In some cases, the client’s language supersedes the terms that I would normally use – or even those that would be considered proper.

For example, last year I was planning the redesign of a major website for a client. I needed to communicate my recommendations to the people who would create the clickable prototype. I determined that given the limited time available, it would make the most sense for me to create an initial set of wireframes as a communication tool. However, the people who were to develop the clickable prototype also owned the creation of “wireframes” for the organization. I spoke with the team, and they agreed that my suggestion of making initial wireframes made sense. A win! In order to make the process acceptable to the organization, we had to come up with a different term. Wireframes became pictorial representations.

With another client, despite the presence of a strong and successful UX team within the same building, the marketing department was responsible for “research” and “surveys.” The UX team was unable to use those terms to describe any of their work, so common UX methods were hidden with more obscure terminology. A simple focus group became a Joint Application Design (JAD) session, though when we needed to do a JAD session, there was confusion about what to call that to differentiate it as well. The work was interesting and challenging (and most importantly, funded), so the UX team continued to put extra effort into these minor details to avoid major conflicts. I also bring my vocabulary into the client’s realm. For instance, I have had a few clients show me their “personas,” despite not having conducted much (if any) user research. I take the time to educate them on what a persona entails with regard to user research and plan work to transition the existing documents into true personas. Until that effort is complete, I ask them to differentiate the existing documents by referring to them as profiles or a similar term indicating that it is not yet a persona because it is lacking research to support it.

These may seem like UX “losses,” but they don’t have to be seen that way. By making the effort to clarify and in some cases change the language we use, the team takes ownership of the work, issues with other teams are avoided, and we can get work done.

Watch my interview with Professor William Gribbons, Director of the Human Factors in Information Design program and founder of the Design and Usability Center at Bentley University in Waltham, Massachusetts. Professor Gribbons sells the value of research by explaining that it is predicting a problem before it occurs and helps avoid it in the first place. Use QR code 132 to access the video, a quick summary of the interview, and Professor Gribbons’s biography.

Watch my interview with Eric Ries, author of The Lean Startup: How Today’s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses and the father of the Lean Startup movement. According to Eric, the Lean Startup approach calls for testing hypotheses. We all agree that we have a strong point of view about what customers ought to want and that we have a rigorous methodology for testing which elements of our vision are brilliant and which are crazy. Use QR code 129 to access the video, a quick summary of the interview, and Eric’s biography.

Watch my interview with Giles Colborne, author of Simple and Usable, Managing Director of cxpartners, and former president of the UK Usability Professionals’ Association (UPA). According to Giles, stakeholders need to understand why research is happening. We help them understand by framing it as “validation,” which answers to their fear of doing the wrong thing. Use QR code 126 to access the video, a quick summary of the interview, and Giles’s biography.

Watch my interview with Janice Fraser, an Adaptive Path cofounder and principal of LUXr (Lean UX Residency). Janice is one of the more important voices in the Lean UX approach, which follows the Lean Startup movement. Notice how the following key principles of the Lean UX approach relate to those of the lean startup. Pay attention to the language. There’s not a word about usability and user experience.

Use QR code 130 to access the video, a quick summary of the interview, and Janice’s biography.

Watch my interview with Donna Spencer, a freelance information architect, interaction designer, and author of Card Sorting: Designing Usable Categories. If Donna hears from a potential client that they want her to do only a design without research, she might not take that job. When she does take the job, she keeps reminding clients that they are making things up, making mistakes, and that they don’t know which of the things they do is wrong. Use QR code 120 to access the video, a quick summary of the interview, and Donna’s biography.

Watch my interview with Caroline Jarrett, an independent usability consultant from the United Kingdom. Caroline emphasizes that the need to develop products that are great for people is only one of the pressures on stakeholders. Use QR code 122 to access the video, a quick summary of the interview, and Caroline’s biography.