Information Architecture Study
Background and Research Goal: A new feature was developed for the Samsung Smart TV that aggregated rows of content from all sources including: streaming video services, live channels and DVR. With the integration of a new “content-first” form of browsing, questions from the team surfaced around how the structure content rows would impact the amount of time and the success for finding content. The main goal of this research was to asses the information architecture for content-first browsing in order to support design in their efforts to refine the Information Architecture (IA) for an exemplary experience.
My Contributions: I led the planning phases including: Designed the methodology, drafted the study plan, secured and initiated onboarding of Optimal Workshop vender, drafted tasks, translated TV UI wireframes into nested categories within optimal workshop, analyzed data, and presented insights to all stakeholders.
Method: A remote tree testing method was conducted (N = 200) using a tool called Optimal Workshop, where participants were provided with a nested list of categories and asked to complete a set of tasks using those categories. Typical TV watching tasks were defined based on findings former research where we investigated how people watch TV. A between subjects design was used where half (N = 100) of participants completed the study using source-based tasks, and the other half (N = 100) of participants completed non-source-based tasks using an identical set of categories. The purpose of using two sets of tasks on an identical set of categories was to compare how people would navigate based on whether or not they knew the source of their desired program. Time on task and success rate was documented.
Outcomes: It was found that source-based tasks took longer to complete and had a lower success rate than non-source based tasks, resulting in the recommendation to increase visibility of “go-to” streaming apps and also visibility of search. It was also found that integrating all different types of TV sources (live channels, streaming videos and DVR) in a single row slowed time on task and reduced success rate for source-based tasks, resulting in the recommendation to keep different sources of TV in their own designated row (e.g., designated Live TV row and DVR row).
Impact: Prior to this research, content from all types of TV sources were mixed within a single row. These findings inspired design to generate a new IA structure that separated the live TV source row with the streaming service rows.
Example of translating TV UI Wireframe to nested categories used to complete tasks in Optimal Workshop:
Disclaimer: Contents displayed in distorted image are for illustration purposes only and this is not an actual depiction of the UI.
