By combining physical and digital elements in our innovation experiments, we can often uncover significant opportunities that weren’t apparent before. Dan Keldsen shares a fascinating example from a retail setting.

Nordstrom’s Innovation Lab posted video of an experiment they ran live on the retail floor of one of their stores this fall, to see what would happen if they created a “flash mob” of iPad app developers to quickly solve a sales opportunity in the store. This initiative is part of a series of one-week experiments run by the Lab to uncover new ways to enhance the retail experience in its stores.

The experiment combined the “high-touch” customer experience they’re famous for with professional staff at hand for shopping advice in the store, with an agile development and lean user experience design team coding a sunglasses selection app for the iPad.

For those who haven’t experienced the fast-paced iterations of innovation that most recently have been made famous through the lean startup movement (which is not limited to being applicable to startups, incidentally), the video condenses many aspects of this methodology.

Similar to a flash mob, where a large group of people show up at a pre-arranged location and time to conduct a performance, the Nordstrom development team set up shop on the retail floor, with a suite of Macs networked together on a table. A sign proclaimed that this is “innovation testing in progress” – signaling to Nordstrom customers who encountered this unusual scene that the retailer is committed to improving their shopping experience.

Initially, the team knows what it’s going to build – an iPad app that helps customers to select the best sunglasses for them. But they don’t know what features and functionality it will have. That will all be driven by customer input on the retail floor. Brave people, these developers. The process begins with storyboarding the process of how customers select and buy sunglasses, displayed on a glass panel with Post-it notes. In the video, customers are seen providing the developers with their perspectives and ideas for improving this process map. This tells the team what they need to build in order to support the customer’s decision-making process.

The next step is a paper prototype. A developer draws prototypes of what the iPad screens and navigation tools will look like, and asks customers for feedback. These are simple line drawings, done with a fat-tipped marker. Again, all this is done on the retail floor, in real time. Customers are asked to interact with the paper prototype, just as they would a real iPad app. This enables the team’s user experience specialists to see how customers would use these proposed screen designs and navigation elements, and can tweak the app’s design accordingly. In the video, a designer identified as “Kim” explains how using paper prototypes enables her to quickly iterate through a series of prototypes, based on the feedback that the team receives from customers.

Within a day, the rapid development team has a working prototype running on an iPad, and continues to refine it over the next several days. It uses the iPad’s front-facing camera to take pictures of the customer wearing two pairs of sunglasses, and then displays the photos side-by-side to see which one they like better. The video shows how the team worked its way through some challenges, such as discovering that polarized sunglasses rendered the iPad’s screen (which is also polarized) unviewable in portrait mode. So the development team re-coded the app to force it to use landscape mode, where the two polarized surfaces no longer interfered with each other.

In another case, the team discovers that customers who tried on many pairs of sunglases need some way to keep track of them. So the team adds a feature to the app where images can be named, either by the sequence the customer tried them on, by brand or model, or by some other distinguishing feature. Zooming is also added, so customers can more easily see the features of the glasses in the images.

The end result is a very slick iPad app that Nordstrom’s sunglass salespeople can use with customers to significantly enhance their selection and purchase experience. Consider, for a moment, how unusual this experiment really was. Developers usually code in cubicles and offices that are far removed from customers. Putting them right in the middle of the retail floor really forced them out of their comfort zone. This team seemed to be very energized about the whole experience. The problem, they explained, is that they’re often far removed from what customers want. This information, in a typical development setting, is often filtered through other people, limiting the developer’s ability to innovate.

Giving developers direct contact with end customers speeded the iterative prototyping and development process, and resulted in much better end product. Customer feedback informed not only the initial design of the app, but helped to refine it each step of the way. Customers were happy, and the Nordstrom sunglass buyers were really excited about their new selling tool.

In your collaborative innovation work, are you bringing fast iterations into your processes? Mixing digital and physical worlds, as Nordstrom did? Any snafus that others should be aware of? If yes, please comment below and tell us what you’ve learned or how you’ve approached it. If not, why not?

Dan Keldsen is a partner at Human 1.0, a customer strategy firm that helps companies innovate business programs, practices and organizational culture to realize the benefits of social media and Web 2.0. This article is based on one that appeared on the Collaborative Innovation website.