Swype
Usability Testing
Project Type: Team, Internal Project at Nuance
My Role: UX Specialist
Skills: Recruiting, Test Planning, Facilitating
Tools: Mobile devices, camcorder, etc.
OVERVIEW
While working at Nuance Communications as a QA Engineer, I was invited to work on a special project with the UX team to conduct Usability Testing for the Swype Korean Keyboard. I led the project from start to finish, and delivered a final proposal to the VP of Mobile at Nuance.
Problem
Without disclosing too much detail, the goal of the project was to compare Swype's implementation of the Korean Keyboard against the competitor's in terms of performance and usability.
Solution
We discovered usability issues regarding the keyboard layout and the special multi-tap timeout algorithm. Then we proposed a number of design modifications, many of which were implemented and released to the market.
1. DISCOVERY Phase
I recruited, screened, and scheduled participants 2 months in advance. I also drafted a Usability Test Plan and acquired devices and equipment needed for the usability study sessions. Then our team traveled to Korea for a week and conducted in-person, moderated usability sessions.
2. Design Phase
We observed users by having them "think aloud". We learned valuable insights from observing actual user behaviors and understanding their motivations. The data we gathered informed the subsequent design decisions for the keyboard’s feature set and layout.
Most importantly, we uncovered our false assumptions regarding how Koreans actually used their 12-key layout keyboard:
Observation:
Used two thumbs to tap really quickly
Only looked at the keyboard, not the WCL
Typing was faster without using WCL on top
Always manually exit multi-tap mode
Assumption:
Uses one finger to tap or Swype
Loks at the Word Choice List for corrections
Utilizes WCL to assist typing
Understands multi-tap timeout feature
3. Outcome
We presented our findings and design recommendations to the Nuance VP of Mobile. Some of these design modifications were implemented and released to the market.
I learned how difficult it is to plan and conduct a full-blown usability testing with participants. But I also learned the value and importance of incorporating real people into designing a good product.