IMPROVING USER RESEARCH PROCESSES AT SAMSUNG ELECTRONICS CANADA R&D LAB
user research / communication of findings
(To comply with my non-disclosure agreement, confidential information has been omitted/obfuscated. The information here is my own and does not necessarily reflect the views of Samsung Electronics.)
Problem
The product development teams at Samsung Canada’s R&D Lab (SRCA) have valued and relied on user research for years, but there was a strong desire to revamp our research processes to produce more timely and usable insights.
Past user research processes used at SRCA were based on academic Human-Computer Interaction research practices where user researchers alone designed and conducted the research, recommended design fixes and communicated research results (often in detailed reports) to stakeholders. Researchers often needed more than a month from initial planning to run the study and present results.
With month-long software development sprints, the results often could not be consumed. Further, design recommendations often could not be followed because the design had already been finalized or the recommendation wasn’t a suitable fix for the problem.
I was tasked in early 2016 with updating our site’s user research processes to increase the turnaround time and influence of user research on design and product strategy decisions. My UX team lead expressed a desire to test each product more frequently and for designers to be more involved in user research. At that time, I was the sole user researcher on my UX team, which had almost 10 visual/interaction designers, working on 4 major projects and a number of smaller ones.
New Process
Using published best practices (e.g., from Tomer Sharon, Jared Spool) I made the following four changes to our processes, changes that focused on decreasing research turnaround time yet maintaining the quality of the research results and the usefulness of the results to stakeholders:
1) Stakeholders enlisted
Stakeholders were invited to be active members of the research team, which I led. Stakeholders for our user research are usually interaction designers on my UX team, but also include product managers/strategists, technical documentation, executives and others. Stakeholders were invited to be involved in identifying the research goals and questions, as well as helping to analyze the research data. Further, stakeholders were strongly encouraged to attend user research sessions, often helping to take notes.
By watching and talking to users, stakeholders gain rich first-hand experience with them and how they interact with the design. Stakeholders build empathy for the users to help make more user-centered decisions.
2) Participant pool expanded
Recruiting suitable participants can take considerable time and effort. We expanded our participant pool to allow us to recruit suitable target users as quickly as possible.
Given the characteristics of our products’ target users, we recruited more people with particular roles (e.g., IT administrators, software developers, marketing professionals), with certain backgrounds (e.g., technical, non-technical) and/or having specific experiences (e.g., managing deployment of mobile devices, android app development). We recruited colleagues, people working at other companies, and current users of our products.
To decrease the recruiting time for any particular study, we invited people from our pool who we could most quickly meet with and would still be representative of a target user for the product and task being looked at.
3) Existing prototypes tested
Most of our research studies are usability tests of product designs. Testing designs often requires prototypes that users can interact with, but these prototypes take time to create. Fortunately our designers already create, as part of their normal design workflow, interactive medium-fidelity prototypes to communicate the design to developers.
We found that our designers’ existing prototypes could be tested with minimal modification. This gave us many opportunities to test designs before developers started coding them.
4) Results documented leanly
Helping stakeholders easily consume and act on research findings is a key part of the user researcher’s job and can take time to do well. By involving stakeholders more actively in the research, we found that less effort is generally needed to communicate research insights to them (as well as “sell” the value of the research).
Communicating user insights to UX designer stakeholders usually required minimal time and effort. The research deliverable was often simply a spreadsheet of prioritized usability issues. Designers usually took part in research sessions, as well as analyzing research data and documenting findings. Given their active involvement in the research, designers (i.e., not the user researcher) provided design recommendations based on their expertise and understanding of the design space. The researcher offered suggestions when appropriate.
When stakeholders were people outside our team, more comprehensive documents of varying formats (e.g.,slide deck, videos, formal reports) were created.
Outcome After 6 Months
While our research processes are still evolving, we found the above four changes already made a big difference after 6 months:
- Participant pool doubled: Our pool has grown from around 30 people before the process changes to more than 60 people.
- Shorter turnaround: The average timespan for planning and conducting research studies was decreased by around 50%.
- Twice as many studies: In 2015, around 10 studies were completed every 6 months, compared to 20+ studies in a similar time period after the changes.