In-Development Internal Tool
Problem
The current state of the software workflow at the company is complex and convoluted, with no single formal path or toolset provided. Due to the nature of the work and interaction of a multitude of differing engineering expertise, an internal tool is being developed to bridge the gap between disciplines.
MY ROLE
UX researcher, UX designer
METHODS
Contextual interviews, usability tests, UX design, mockups, affinity diagram, journey map
CLIENT
Sandia National Laboratories
Role
I operate as both a UX researcher and designer on a UX team of 3, engaging with a larger software development team who operate in an agile environment with a 3-week sprint cycle.
My role includes involvement in:
Assisting the lead researcher in managing ongoing and future studies for the team, including development of research plans, participant selection, scheduling, and presenting results for each sprint to the larger audience of interested engineers and management.
Taking on the role and responsibilities as the UX lead when requested, acting as the main communication line with the dev team for UX needs and responses while balancing and delegating tasks each sprint as needed.
Conducting interviews, usability tests, and wireframe feedback elicitation sessions depending on the needs of the current sprint.
Collecting and synthesizing data to enable informed decision making about the future direction of the internal tool along with visual artifact generation.
Outcome
Refinement of scope and requirements for the in-house tool.
Understanding of the user bases, their current processes, needs, and gaps.
Improved tool functionality and identified areas for improvement.
Data-driven visual artifacts developed to inform future strategic planning and decision making by the team.
Continued funding for UX partnership with the software development team.
Selected and requested to join another development team as the lead UX researcher.
Methodology
Problem 1: Domain Exploration
Several pushes were done to understand the userbases, their needs, and how different userbases were connected via touchpoints, activities or goals. From these pushes into different userbases, several data-driven visual artifacts were generated along with continually-evolving wireframes for early development of a web-based version of the tool.
Participant recruitment
The original list of participants was selected by domain-knowledgeable stakeholders. As interviews were completed, snowball sampling was utilized to identify additional participants.In-depth interviews via Skype
Each interview was 2 hours, either as:1 hour of semi-structured interviewing and 1 hour of process feedback sessions with users (which led to the creation of an experience map depicting two types of users and where their processes intersected)
1 hour of semi-structured interviewing and 1 hour of co-design sessions with users. Co-design sessions presented each user with the current version of wireframes to identify missing features, incorrect language, and what balance of needs would create the most value.
Interpretation sessions & data analysis
After each interview, interview data was translated into interpretation notes and moved into emerging themes within an affinity diagram, with each domain exploration push resulting in different affinity diagrams and themes to help inform our understanding of the space, current issues, and the future state.Wireframes
A barebones minimum viable product (MVP) wireframe was developed in Balsamiq as a joint effort between the UX team and internal development team to use as a discussion point with participants. A more in-depth set of wireframes with additional functionality and clickable features was also developed for additional discussion as more information was gathered.
Problem 2: Software Iteration and Evolution
As new features are implemented into the tool, usability tests allowed us to identify confusing call-to-actions and expected functionality. Each set of tests provided greater insight into the userbase’s expectations and needs that may not be supplied by the current set of tools available within the company.
Participant recruitment
Previously-created personas were leveraged to inform participant selection to reduce sampling bias from a compiled list of suitable participants across the company. 5 participants were recruited for each sprint of usability testing.Usability testing and moderation
Participants completed 10+ tasks per sprint cycle in a 1-hour timeframe along with additional follow-on questions to collect further user sentiment. The internal team would attend sessions when possible as bystanders and participate in the follow-on discussions about what was learned during the usability test. All sessions were done via Skype due to Covid-19 restrictions.Data analysis
After each set of usability tests, I would analyze and synthesize the findings into a presentation to present to the development team to discuss next steps and backlog based on the latest results.Client presentations
A high-level version of the results and recommendations is generated for the larger audience of stakeholders and interested parties to keep the community abreast of the latest changes in the tool development.
Challenges & Limitations
Throughout this project, challenges came up that impacted decisions and project evolution:
Domain complexity
The user base and software cover multiple engineering disciplines, requiring rapid uptake and in-depth understanding of the users, tools, and methods employed to ensure robust analysis and appropriate recommendations.Remote interviewing & testing
Unable to test in-person due to the pandemic, leading to relying on online resources and screen shares which were not always reliable nor allowed for in-person expressions or gestures. Required quick adjustments on the fly with virtual diagnoses and resolution when problems arose with virtual communication.UX tool limitations
Unable to utilize many common web-based UX tools due to security reasons, resulting in the need to adapt to the limit toolset available internally.
Next Steps & Reflections
Domain exploration continues currently due to the complexity of the userbases and interactions within the company with the expectation of the data saturation leading to guidance on the decisions of the future direction of the program.
Along with continuing this work, another development team in the same domain space has requested UX assistance, which I am now leading as the main UX researcher.