UX Design Process at IBM Analytics
At IBM, I split my time between UX design and user research. I have been responsible for leading user research for my product team and using those insigts to help drive design decisions on analytics platform. This is a brief walk through of my UX design process from various projects I've worked on in IBM's Analytics Platform.
- Kickoff workshop
- Put faces to names
- Bring together dispersed teams
- Introduce roles
- Share team needs
A kickoff workshop to bring extended teams(dev, design, OM) together and make sure their needs are heard. We go through activities to improve collaboration and understand each teams' priorities.
- All hands questions and assumptions
- Meeting with stakeholders to note their biggest questions
- Clarify target users
- Map out sprint deliverables
- Recruit subject matter experts to understand domain
With every project and design sprint, there are questions and assumptions we have about the users and their work flow. Here I am responsible for organizing and facilitating brainstorming activities with the extended team to draw out these questions.
- Research goals
- Methods and activities
- Estimated Budget
- Number of participants
- Engage with user communities
- Create screeners
- Host online surveys for user recruiting
- External recruiting through UserTesting.com
We build relationships with user communities such as Galvanize, a data science bootcamp in Austin and University of Texas's phD students working on data science and machine learning projects.
- Diary studies
- Journey map
- Collaboration map
- Persona map
- Feature prioritization map
- Involve extended team in primary research
Synthesizing user interviews to uncover and provide clear personas and archetypes for IBM analytic platform’s executives.
- Synthesize interviews
- Share knowledge fast with team
- Design principles
- Brainstorm ideas
- Loop in user feedback
- Lo-fi and Hi-fi wireframes
- Clickable prototypes
- Usability testing
- A/B testing
- Survey report
- Agile collaboration with dev/design/OM
After completing initial research around the users, we have an all-hands brainstorming session to quickly share our ideas/solutions on post-its. Then we cluster similar ideas and individually vote for top 3 ideas. When possible, we invite experts and end-users to participate in these sessions with us.
Before getting into the granular details of the product, we create an end to end experience in an information architecture map.
In preparation for usability interviews, I work with my teammates to create clickable prototypes. Then I write scripts to go along with the tests and create note taking templates. After each interview, we keep track of the feedback we received for interactions on each screen, make note of tasks that passed or failed. When we are done with all the usability interviews, it is most effective to share our new knowledge with our extended team as soon as possible. One way we do this is by sending out a bullet point list of high level actionable insights Taking too much time making precious research reports can be ineffective in a fast moving agile environment.
- 2 week design sprints
- Drive design decisions with lean UX research methods
- Update extended team on research/design asap
- Regular tech feasibility check with dev team
At IBM it is common to have your team's developers, designers, and OMs in three different locations, therefore, it is essential that we keep a regular cadence of keeping everyone updated on our work process. I was responsible for leading research and sharing insights to make sure design decisions were driven by user research during our 2 week design sprints. This cycle was recognized internally within IBM analytics and shared among other researchers and designers on the platform.