Optimizing Workflow
Run Configuration
Surfacing batch submission and engine settings
from CLI to UI, for a more user-friendly Experience
⬆ 3x Feature Adoption
⬆ 85% prefer UI over CLI
Context
DNAstack: a cloud-based company, that aims at advancing precision medicine through secure data sharing and analysis
Workbench: a product enables bioinformaticians to analyze multi-omics data, build workflows, and collaborate across teams in a secure cloud environment
Problem
Limited User Adoption
The product's reliance on command-line operations (CLI) limited its audience to users with coding skills.
Role
Product Designer: User Interview, Competitive Analysis, User Flow, Wireframes, Hi-Fi Design, Interactive Prototype, Developer Handoff, User Testing, Heuristic Analysis
Collaborated in a cross-functional team of 8
Solution
In two months, I successfully transitioned the command-line interface (CLI) functionality into the user interface (UI). The redesign resulted in:
✦ 3X increase in feature adoption, as non-technical users found the tool much easier to use
✦ 85% of users preferred the UI over the CLI during testing, confirming the improved user experience and accessibility
Research
User Interview
I conducted interviews with 3 bioinformaticians, to gain insights into their workflow submission process. My focus was on identifying repetitive tasks they performed using CLI during each submission. This helped prioritize which tasks to transition to the UI for the initial redesign.
Key Insights:
• Users rely on the CLI for workflow batch submission
• Users also use the CLI to configure engine settings
Competitive Analysis
I researched on some leading bioinformatics tools commonly used by our target users, and examined how these products integrate batch submission and engine setting features into their UIs.
Technical Constrains
Transitioning the CLI features to the UI required ensuring compatibility with our existing framework. I presented my research and findings during team meetings, and actively sought input from developers to assess the feasibility of implementing these changes.
Feedbacks:
• Some engines have a large number of parameters. To simplify the interface, suggested displaying only the most essential parameters by default.
• Recommended incorporating a JSON editor for users needing access to advanced settings
Prototype
User Flow
Integrating engine setup into the current interface was complex. To determine the most user-friendly approach, I mapped out several user flows and conducted A/B testing to identify the optimal solution for our users.
By collaborating with the team and incorporating their suggestions, I refined the user flow to enhance the experience. After thorough evaluation and testing, I finalized the flow that best aligned with user expectations and needs.
Interactive Prototype
This is an interactive prototype I created that enables users to create multiple workflows and submit them for batch processing.
I designed this interactive prototype to demonstrate how users can configure engine settings prior to submitting workflows.
Prototype Testing
I shared the prototype link with our data science team to gather feedback. The response was positive; they appreciated the user interaction and recognized the potential impact of having such functionality accessible entirely through the UI.
Implement & Test
User Testing
After implementing the new interface, we invited users from our partner organization, who were interacting with our product for the first time, to test the feature. However, the feedback we received wasn’t as positive as expected. During the user testing session, we closely observed their interactions and identified key areas where the feature could be improved.
Key Insights:
• Users had difficulty discovering that they could open the sidebar to edit the table
• Users tried to use keyboard shortcuts for data editing, but the feature was not supported
• Users would need a link to documentation on the page
Heuristic Analysis
After conducting user testing, it became clear that the current UI needed usability improvements. To further evaluate the interface, I applied Jakob Nielsen’s 10 Usability Heuristics as a framework to systematically identify specific areas where the design was causing friction.
.svg)
.png)
.png)