What I Learned from Watching People Clean

The process is critical, and a core tenet is bringing together cross-functional expertise from start to finish. This approach was pioneered by IDEO, the firm that designed the first usable mouse, for Apple’s Lisa computer, in 1980. At that time, innovation was typically siloed. Someone in the research-and-development or the engineering department would come up with an idea and hand it off to someone in the design department, who would in turn hand it on to someone to build it, and then to somebody else to market and sell it. IDEO used an entirely different approach, however, bringing all these capabilities together at the same time. In doing this, it helped popularize the notion of cross-functional teams in which people across multiple disciplines, with distinct expertise, work together throughout the life cycle of a project. Design thinking depended on building teams that comprised researchers (including people with anthropology backgrounds), as well as people in strategy, R&D, marketing, manufacturing, industrial design, finance, and so on. While this approach is now more common, it is remarkable how many organizations still deploy something closer to the traditional siloed model. 

A design-thinking approach proceeds along the following lines:

Step #1: Empathy

Once you’ve assembled a cross-functional team, the first step is really one of empathy. Just as it was in the bathroom-cleaning research, the goal is to learn about the people for whom you’re designing, to gain a deep understanding of their lives, needs, problems, and frustrations. 

This stage begins by uncovering a lot of qualitative customer data, with the aim of defining meaningful problems to solve. Come up with core themes or insights, and then develop those insights into problem statements that can be used to spark ideas. Look for customer needs that are currently unmet or only partially met by existing products. Once you define customer problems, developing solutions can be much more targeted. 

To identify those problems, it helps to have good questions, which typically start broad. When interviewing people about bathroom cleaners, we didn’t start by asking about cleaning products. Instead, we asked people about their lives and overall cleaning routines, and to show us what they actually did. The questions were less product specific and more journey or process specific. For example, we’d say, “Tell us about the last time you did this. What was the process you went through? What worked best? What was worst?” We eventually drilled down to “Why did you do that?” and “Show us the products you use, and how you clean your bathroom.”

This sort of in-context research process is standard practice for many innovative organizations today. For example, in his 2017 book Hit Refresh, Microsoft CEO Satya Nadella observes, “The business we’re in at Microsoft is to meet the unmet, unarticulated needs of customers. That’s what innovation is all about. There’s no way you’re going to do that without having empathy and curiosity.”

It’s important to consider which end users will give you the most useful responses: people who are at the extremes are often more articulate about their needs. Say you are trying to create a new pet-care product. The best subjects might be people who own multiple pets, or people who run pet-walking or pet-sitting businesses. Tapping into groups such as these, that experience the problems in the extreme, is more likely to uncover unique insights and inspire novel innovations that could be relevant for the broader market. It’s not that you design solely for people with extreme needs; rather, people in the mainstream may not have developed the same language to express their needs. In the cleaning-products example, extreme users included people with very large families, those with allergies to harsh chemicals, people who clean very frequently or very infrequently, and people with brand-new versus very old bathrooms.

Step #2: Ideation

If the first phase of design thinking is about building empathy to inspire innovation, the second is about creating new ideas. Because you’ve been immersed in the end consumers and their needs, you’re able to come up with more-effective solutions, and to do so efficiently. 

Take my bathroom-cleaner example. There was a clear need that wasn’t being met: people wanted a protective barrier on their shower tiles to reduce the cleaning effort. Because the market for bathroom-cleaning supplies didn’t fill that need, they instead turned to a work-around by using a product designed for cars. Through interviews, we realized people were willing to take an extra step up front to avoid a future pain, to invest more time initially in applying a car wax to avoid a greater amount of time, energy, and effort later. They wanted to make future cleaning easier, therefore they wanted something that would prevent soap scum, mold, and mildew from sticking to the tile. When you clearly identify the problem, you know what you need to solve. 

In terms of effective ideation, more than 50 years of research on the subject has demonstrated that traditional, in-person group brainstorming suffers from significant flaws. Research by Radford University’s Gary R. Schirr on ideation suggests that in-person group brainstorming tends to hinder both the quality and quantity of ideas. In a group setting, not everybody is fully engaged. Only one person can share his or her ideas at a time, which at least temporarily blocks everyone else from sharing their ideas. A third problem is conformity: when we hear others’ ideas, we’re influenced by them, tend to conform to them, and generate more ideas like them. 

A different approach, modeled on research by Columbia’s Olivier Toubia, is to start generating ideas first as individuals, before using the power of the group. Individual and group brainstorming can then come together using a simple online discussion-board format, where individuals anonymously post their own ideas, and then other group members build upon those ideas. This can help keep separate the process of generating ideas from that of evaluating them, which is an important distinction. One advantage of using an online method is that people can contribute in their own time, rather than having to be together in the same room at the same time. Another is that their comments can be kept anonymous, eliminating any bias toward ideas that come from senior executives. Toubia’s research suggests this approach is more effective than traditional in-person group brainstorming and generates a higher quantity and quality of ideas. 

Step #3: Prototyping and testing

Ultimately, the evaluation process should essentially lead us back into the bathroom, to ask the customers which of the solutions we’ve come up with best meets their needs. The final step of design thinking is to continue to deepen our understanding of the need, and to test and refine initial solutions. It’s important to assume up front that you don’t have the ideal solution, so it will be necessary to test and iterate, retest and reiterate, and learn along the way. 

The team creates prototypes of the most promising ideas, and then tests these with users to get their feedback and to refine the prototypes. A prototype is part of a feedback loop that helps us understand the problem more in-depth, to ensure our solutions are appropriate.

The simplest prototype is a new product description and visual, what some people call paper prototypes, that you can take to customers for feedback. What do they like and dislike about the product? How might they improve it? How likely would they be to buy it? After this stage, you can move to testing physical prototypes that may or may not be functional. You’re looking for quick feedback to refine the idea and test it again, without investing a lot of time and money. This stage is ultimately about failing and adapting. Expect to fail along the way—fail fast, fail cheaply, learn, and figure out the next best thing to do.

The most promising prototypes will move forward into development, and eventually one will launch as a product or service that meets the three tests of design thinking: it is desired by customers, technically feasible, and viable from a business standpoint.

This may all sound very linear, as if the process simply moves from one stage to the next. But design thinking is messier than that. It’s a process of repeated divergence and convergence: the empathy phase uncovers diverse needs, and teams diverge on what’s important until they converge on a few key themes. They diverge again at the ideation stage, coming up with a variety of ways to solve the problem that’s been identified, until they converge on a small set of high-potential solutions. Also, design thinking is flexible and agile, so teams move back and forth rather than only in one direction. You might test a few prototypes and learn that some just don’t work, which would prompt you to go back and ideate more. You’d then develop more prototypes, test those, and perhaps repeat the cycle. The result is an iterative approach to innovation.

Design thinking eventually led to recommending a new bathroom-cleaning product: a daily shower cleaner. This now-well-known product comes in a spray bottle and quickly creates the barrier consumers sought for keeping mold, mildew, and soap scum from sticking to shower tile. The success of the daily shower cleaner is one example of the power of design thinking. Meet customers in the environment in which they use the product, ask good questions, listen and observe closely, and iterate solutions to solve important, unmet customer needs. The organizations that do this will innovate more systematically, and have higher success rates.

Arthur Middlebrooks is clinical professor of marketing at Chicago Booth and executive director of Booth’s Kilts Center for Marketing. This essay is adapted from a presentation delivered at the Kilts Center’s Marketing Summit 2019.