All projects should include research.
That’s the current thinking in design research and user-centered design. Indeed, many of my Adaptive Path colleagues won’t do a project unless it includes some research to uncover the goals, motivations and needs of potential users. More and more, however, I’ve found my views about the importance of research have become less dogmatic. On several recent projects, I’ve conducted no research at all—or at least very little of it—and those products seem to have turned out fine and are well liked by users. Luck? I’m not sure.
What I am sure of is that there’s only a loose correlation between research and the final outcome of a product. Microsoft spent at least two years researching Vista; Apple did no research that I know of on Mac OS X. Now, obviously there are many factors (technology, business, marketing, etc.) that go into the creation of a product or service, and it’s probably unfair to judge products in this way, but there are few other ways for designers to evaluate the value of research except through the success of the final product. Brilliant insights into users’ needs are effectively useless—just proverbial trees falling in an empty forest—unless they reach those users in the form of a successful product.
And what about projects that build upon other projects—which is to say, most projects? Is it necessary to conduct research simply to add a feature to an existing piece of software, or a new section to a website? Perhaps. Or, just as likely, perhaps not.
In Jesse James Garrett’s seminal essay ia/recon, he admits that, in the end, he follows his hunches: “Guesswork is an inescapable part of our work. More importantly, the quality of guesswork is what differentiates a good architect from a bad one.” And Michael Bierut reveals the same in a recent essay: “Somewhere along the way an idea for the design pops into my head from out of the blue. I can’t really explain that part; it’s like magic.”
One of the reasons designers are hired is their expertise—those “good guesses”—part of which comes from knowing what works in most situations, and what doesn’t. It could be argued that this expertise (which is made up of intuition, experience, understanding and taste) is more important than an understanding of users. I’m not sure I want to go that far, but I have decided that there is a more reasonable approach than the dogma that research has to be included on every project. Evidence that is all around us, from the humble fork to the lauded iPod, proves that this dogma simply is not true.
Most experienced designers have enough expertise to get many products 80% designed without ever doing research, and sometimes that 80% is all that’s needed. Research can be a useful tool, but it can also be an ineffective waste of time. Good designers make good designs, not research. Even with good research, you can follow users (and time and money) down some serious rabbit holes, never to return.
When to Use Research
Here are some research guidelines that I use for my projects. Only use design research when:
1. You don’t know the subject area well. I’m not an expert in investment banking, so if I had to design a product for investment bankers, I’d need to learn about what they do and why they do it.
2. The project is based in a culture different to your own. Chinese culture isn’t the same thing as the culture of the United States. Or India. Or Western Europe. Cultural differences can cause differences in behavior and expectations for a product.
3. You don’t know who the users are. This should be self-explanatory, but amazingly enough, many companies don’t know who uses their products or why. If you find that your view of the users is different from the stakeholders’, you might want to establish a consensus around that—the type of clarity that only research can provide.
4. The product is one you’d never use yourself. Luckily, as an affluent white male in my 30s, I have a lot of products directed at me. But I’m not a doctor or nurse, and I’m not likely to use medical devices, so if I was working on a medical device project, I’d have to rely on research to teach me how the device would be used. Note, however, that this approach can make for some narrowly focused products, which only work well for a small group of people.
5. The product contains features and functionality that are for specific types of users, who are doing specific types of work, work you don’t necessarily do yourself. MS Office contains a bunch of features that I would never use, but if they were removed, key power users would scream bloody murder. Sometimes you have to conduct research to understand the nuances of a specific feature, as well as its importance to a specific group of users.
6. You need inspiration. Sometimes you get stuck and an afternoon away from your computer screen can spark ideas and provide unexpected directions to take a product.
7. You need empathy. Some types of people and groups are harder to identify with than others. Illinois Neo-Nazis for example—not that I’d ever do a project for them. But what about the elderly or infirm? It’s difficult to understand their situation unless you know about it.
8. You don’t have much expertise. Admitting this is humbling, but necessary. Research might not make you a good designer, but it might make you a better designer by exposing you to new things and preventing you from making simple mistakes that a more experienced designer would avoid as a matter of course.
A Tool, Not an Approach
It could be argued that these “Only When” guidelines I just outlined apply to every design project and every designer. Which is true, to a degree. Who, for instance, doesn’t need inspiration and empathy?
But I want to shift assumptions about research: Namely, stop thinking of it as a necessary approach to design and start thinking of it as just a helpful tool. Saying that research is required for every project would be like saying that all projects need wireframes or content analyses, which just isn’t the case. Yes, research is good for many types of projects (as outlined above), but it isn’t always a necessity.
As Jesse pointed out, “Research can help us improve our hunches. But research should inform our professional judgment, not substitute for it.” Like other tools in the designer’s toolbox, research should be used only when necessary, not applied to every project unthinkingly.