Panel Report: OneBusAway and Nordstrom

OneBusAway

I learned that the designing of OneBusAway followed a (albeit more elaborate) similar path to the one we took. They identified a user group, which was bus riding veterans who simply needed to know when the bus was coming, and molded the app after the needs of that group. I was surprised to learn about the politics aspect of making the app, especially concerning literally one line of text. It sort of made me realize how much attention to detail an app got before being published to the public. I was also fascinated by how through the years, OneBusAway adjusted their user group to more broad ranges, and modified their app to accommodate for that. This showed me that the ideas you have for an app starting out are not always going to be the same, and that it’s OK for the app to shift.

Nordstrom Anniversary Sale

I was most impressed with Laura discussing how they “failed” the usability test two times. In my mind, as a professional, she was able to be very upfront about how the previous two usability tests were not successful and how they went about conducting the third one (which was successful!). She talked about how they tested the website and whether or not it was clear to the customers when will there be a paywall, when do they need to sign up of log in, etc. I liked the subtle changes that each testing brought to the website, such as highlighting the mouse-over sections in yellow, and putting the disclaimer in a yellow background to make it stand out more.

What now?

Now that I’ve seen industry professionals, I think I would go about modifying my projects in the following ways. For my interaction design project, I wasn’t really thinking of a target user group. I also wasn’t too particulary bothered with making sure my icons are actually intuitively clear. As OBA showed, even trying to show when location data is available took a long time for them to perfect. I would need to do more testing with my design (i.e show it to more people). For my usability sprint project, I think I would need to try to measure the effectiveness of the test results. There were definitely a lot of noise and bias in the data, and next time I should strive to eliminate those.