Our field research, shop-alongs, and user testing illuminated a great number of the features that we knew would resonate with girls. We also knew how we were going to technically build a platform that would work with existing systems within the framework of a modern, high-performance web application. What we didn’t yet know was our potential return on investment – so we made a model.
We calculated the revenue that would be generated with in-store touch screen kiosks given the number of kiosk minutes per day over a year, using success and abandon rates with the application. Then we layered online sales onto our model. That got us the revenue we needed, but we realized that we were hamstringing ourselves. Our budget only allowed us to build for desktop web, ignoring mobile. But our numbers told us a different story. Analytics and research indicated that as many as one-third of our customers would use a mobile device to access the web application. A great source of additional sales and delighted customers was being left out. American Girl agreed, and we began building the responsive web application, tightly integrated across in-store, web, and mobile, that you see today.
Given early sales results, our models look conservative and quite accurate.