...
In conclusion, the user testing went very well and really showed the necessity for the loading screen animation. We might show two tiles being pinned in the animation and a page transition in order to show users how to use the app. It would basically be an automatic demo in the loading screen, and that is probably the next step in the implementation. User testing was very beneficial to us during this project's implementation.
Reflection
Through the iterative design process, our team learned a great deal about how to design for different user populations. Although we feel that we produced a great product that solves a real problem in a novel way, there are a few things we would have done differently.
First, we would have spent more time thinking about and developing our idea from the beginning. In retrospect, we felt that we could have done more to solidify our idea earlier on in the design process. We had an idea of the problem we wanted to solve, but did not have any strong idea of exactly how we thought it should look. This sometimes prevented others, such as our TA, from really understanding what we wanted to build. If we had been able to explain more precisely, we probably could have gotten even better feedback.
We wish we would have done a better job of taking bigger risks earlier on. Our final product is one that steps outside the box in many cool ways. This final design only developed after we had gone through multiple iterations, and we could have gotten there easier if we had been less risk-averse early on.
One of the hardest parts to paper prototype was the fact that we built our app for mobile. It is hard to simulate actions such as swipe on a paper prototype. It would have been more effective if we had used a Wizard of Oz prototype or something similar on an actual handheld device instead of the paper.
We wish that we would have investigated the possibility of building an Android native application earlier in the design process. We realized that some of the features we were trying to implement would be more natural on Android than on Mobile Web, but not until after we were working on GR5. We made it work on Mobile Web just fine, but we would have liked to have explored other options sooner.
We wish we had prototyped the more complex edge cases of our home screen a bit better. For example, when a user "pins" a certain restaurant that he wants, we don't allow him to swipe that option off unless he unpins the restaurant first. This provides safety for the user, as well as is consistent with our theme. We didn't really prototype this feature though, and occasionally our users would try to swipe pinned items, which would cause problems, because we never learned if users could learn from those mistakes. In other words, we didn't do a great job accounting for edge cases in prototyping, because we thought they were rare and less beneficial. In retrospect, some of those edge cases could have been more prevalent than we realized, and we could have learned more about the correct way to handle them.
We sometimes had difficulty handling situations where two users would provide opposite feedback about a certain feature. We sometimes didn't know what to do in these situations. Usually we would find a "tiebreaker" user to decide which way made more sense. If we could do it again, we really should have gotten even more users to test features that seemed controversial. Doing this would have provided us with two benefits. First, we would have had a larger sample size so we could have made our decision with even more confidence. Second, it is possible that one of our users could have helped us identify a third solution that would have been acceptable to both sets of users that were arguing from the beginning.
Throughout the project, we felt that we made a lot of great decisions and did a great job. The preceding commentary is focused on things we could have done better if we had to do the project again. We chose to omit commentary in this section about things we did well because we felt there was less of an opportunity to learn from those. Going forward, we expect that we could make an even better UI Designed product if we had another chance, and we would learn from the aforementioned cases specifically.