Rethinking Application Design

In: Columns > Innovating the Web Experience

By Dirk Knemeyer

Published on November 7, 2005

I’ve learned a lot about application design during my career. I started out working in medium-sized agencies in the Midwest, primarily servicing large corporate clients. From that experience, I developed a conceptual model of how to accomplish Web design based on my observations of both the structure of my agencies and the structure of the internal design and product teams of our clients.

In May 2004, we established our interface design company, Involution Studios, in Sunnyvale, California. Now, with more than a year of experience working in Silicon Valley and seeing inside various leading technology companies—both as a consultant/designer and through my speaking and networking activities—I’ve realized that the basic corporate design model for Web and application design is broken. This article will share some of the conclusions I’ve drawn and propose some better approaches for designing successful applications.

The Parade of Specialists

One plus one can equal eleven; it can also equal negative one. In agencies, team member specialization can be numbing. I’ve led teams that included (hold on tight now) all of these individuals:

• Design director
• Technical lead
• Web strategist
• Project manager
• Information architect
• Interaction designer
• Visual designer
• Interface developer
• Usability specialist
• Software engineer

In retrospect, it makes my skin crawl. There was a tacit assumption that it was necessary to have a long list of specialists. So there was little resistance when we regularly held meetings with six team members, all tracking billable hours, and spending as much time trying to bridge communication gaps as figuring out what each of us did on the project. Sure, sometimes projects involving large teams ran smoothly, but there was still shocking inefficiency.

Not all agency Web teams are bloated. But some smaller teams follow the assembly-line model, where each specialist or expert owns a particular facet of the eventual product. The rather bland quality of the final product is only overshadowed by the thick final invoice, which details so very many billable hours.

Internal client teams are not much better. Large, traditional companies staff their digital design teams with too many layers and far too many participants, and rampant inefficiencies create a virtual logjam when it’s time to actually get something done.

It took me years of observing the dynamics to realize how wrong they were. When I started my own company, I wanted things to be very different.

Silicon Valley: The Land of Milk and Honey?

Despite the shift of digital technology firms to to other places—most notably Asia—Silicon Valley remains the international center of computing technologies. Many of the biggest and most successful technology companies are based here, and the constant buzz from entrepreneurs and venture capitalists is palpable. There is a dynamic mix of industry veterans who have helped pioneer digital technologies, and a regular influx of ambitious and talented young dreamers who are relocating here from around the world. It is the geographic heart of our industry. As Frank Sinatra famously crooned about New York City, if you can make it here, you can make it anywhere.

So I was surprised when I found out how these top technology companies actually approach user interface (UI) design. In general, the process is dreadful. In many cases, engineering drives the product and the UI is an afterthought. Other times, the UI teams are bloated and over-specialized and the expensive, disorganized design process results in mediocre products.

Too often, key people don’t understand their roles or the design process—or what ultimately constitutes quality user experiences. These issues are particularly glaring since much of the work being done in Silicon Valley is application design, not “just” Web design.

What’s the difference? Application design helps users carry out tasks, and can be as simple as Google’s search function or as powerful and complex as Photoshop. Web design is the design of a strictly informational or marketing Web experience that employs a fairly straightforward interaction model. Many of us who came into the industry through the Web do not fully appreciate this distinction or properly recognize the complexity of application design. This makes it all the more surprising that so many technology companies are not taking an effective approach to designing their UI. Their various problems are so clear to the experienced eye. Why do they persist?

Over- and under-staffed UI teams are behind this conundrum. There is no standard process for designing an application, so ineffective methods and structures live on. And not everyone with the same title has the same background and skill-set. This can make staffing a design team like playing Russian roulette: portfolios and case studies might look great, but it usually takes working with someone to truly appreciate where their areas of expertise begin and end.

Our industry generally does not know how to design great user experiences, so the blind are always leading the blind. This is amplified when firms copy the methods of teams that have produced successful products. Unfortunately, successful product design sometimes occurs despite the structure and process of the design team—not because of it.

So What Should We Be Doing?

Here are my recommendations, based on working with and for numerous companies, and seeing success with my own company:

Use the smallest design team you possibly can

The idea of small design teams is in vogue right now for a reason: they just work better. They are faster, their work stays in focus and true to vision, and they avoid many of the political problems and the interpersonal tug-of-war that happens in larger design teams or departments. We have clients who want us to redesign their applications without their internal design team to avoid muddying the process and degrading the quality of the final product—even though it’s their staff! What an amazing example of business-level recognition for the warped nature of corporate design teams today.

Put an interface designer in charge of the interface design team

When design teams are led by engineering, marketing, testing or research, they’re led by someone who is not skilled and experienced in designing to meet business needs, market opportunities, technological capabilities, and user desires. That is what an interface designer is trained to do. After all, would we let the chief information officer make decisions on the financial direction of the company over the protests of the chief financial officer? Of course not. Design is every bit as specialized, professional, and important. Successful design—not surprisingly—is best led by someone with design skills.

At all costs, avoid design by committee

Everyone has an opinion about design. That’s because people see design as form: something sharing more with style and layout than behavior and interaction. But design is about function as much as form, and suggestions to change one aesthetic thing or another can have a huge impact on the cohesion of the design and even undermine the conceptual model. Create an environment that is open to feedback and input, but ultimately let the experts make the decisions and control the design.

Invest in fewer, more senior people

People with years of experience and strong skills across various points on the design continuum are typically more effective than junior employees and a string of specialists. Fewer participants means clearer communication, less room for interpretation from one person to another, and better attention to the task at hand (instead of fractured focus across various projects and a less thoughtful production-oriented mentality).

Let the design team operate outside of the traditional organizational framework

Much of what makes up a company or organization is management structure. That includes processes and policies that are anathema to innovation, change and creativity. That is why design firms are popular: design thrives outside of the staid corporate framework that it ultimately needs to serve. But even if you want to keep your design in-house, at least recognize that the design process is far different from the other things happening in the organization and give it the space and opportunity to succeed. Treating design like just another corporate function will stifle its success.

Treat user testing as a data point, not the driver

I’ve written previously on the insidious nature of a usability culture on design. The results from supposedly objective, quantitative usability testing and evaluations often end up driving design. This not only limits innovation, it also results in design being dictated by unnatural testing environments and analytical observations from non-designers instead of use by real people in real situations. This offense is particularly pointed in application design, where success is less about how “intuitive” a design is and more about the value it provides from extended use. User testing provides valuable insight, but it is not meant to serve as the arbiter of design decisions.

Getting to Work

If you are currently operating on a bloated and oversized design team, providing design in a suffocating corporate environment, or designing products that reflect the analysis of usability experts over the insights of actual designers, you are contributing to products that will be less successful than they could be. Worse, you are reinforcing bad habits in your own mental model and gaining the wrong kind of work experience.

This is a disservice to your career, your employer, and/or clients. As we enter into a new period of growth and opportunity in application design, driven by the evolution of the Web to an increasingly rich interaction environment, we have the opportunity to improve our skills and increase the value of our work and the quality of what we create. Since the demand for talented Web professionals now far exceeds the supply—that is, there are plenty of jobs to choose from—this might be the perfect time to reconsider the path your career is on and move into a professional situation that properly values and empowers great design. The future of our industry is bright, and the better prepared we are the more successful we will eventually be.

And for those of you who aren’t learning anything new from this article because you’re in a good situation, share your experiences as much as you can. The result will be better designed applications for all of us, and a much more enjoyable and usable world.

Related Topics: User Experience, Interaction Design, Programming, Business