A challenging problem space...
As a product designer, I enjoyed working on the ambiguous and complex problem space presented to me. To deal with ambiguity, I believe in taking a step forward and iterating as necessary, as product design is an iterative process. I started my research by understanding the target persona, which was data engineers. I discovered that enterprises needed a seamless end-to-end platform that reduced friction for each data engineer persona to work with the tools of their choice. This was an opportunity that I identified.
Deep dive into customer insights
I conducted further research by interviewing 25+ customers. I discovered issues with performance time, user experience, and data dispersion among competitors in the market. To form a product strategy, I broke down larger chunks and scrutinized smaller tasks. Later, I looked at it holistically as an advocate for user experience.
Performance time
Competitors took 40% more time than the ideal performance time to ingest, manage and transform data.
User experience
Numerous errors and clicks
Unalignment between user's mental model and the application
Data dispersion
On an average customers use 5 or more different platforms with a rough split of 50-50 on-premises and cloud to store data according to EMA research
What I shipped..
I designed the solution which is a unified data administration tool for our target personas. Focusing on data engineers as a persona alone, I designed and shipped the workflow for their tasks i.e ingesting data from various sources into our platform to clean and transform for the other personas.
Some screenshots of the high-fidelity mockups
The Process
We were building a consumer-grade experience for enterprise users. I followed a collaborative, data-driven, and decision-heavy process involving various teams to create a unified hybrid data product. I relied on research data and collaboration to iterate on complexities and problems. I presented all findings to key stakeholders, and each phase informed the next phase of the design process.
Strategize
"Handling feedback was a particularly difficult task and seemed like a constantly shifting balance of divergent perspectives"
To tackle this, I relied on the data and used design principles to inform decisions. From my research, I created user stories based on pain points in the industry such as performance time and data dispersion. These stories served as the basis for project planning, estimation, design, and development work. I led the creation of personas, a user journey, and information architecture.
High level user journey map of a data engineer
Design principles
Before designing, I set success metrics and design principles that tied back to the business goal of increasing NPS score and customer adoption. I designed low and high-fidelity mockups in Figma and accounted for internal stakeholder feedback to refine the solution. During the design process, I considered design rationales such as why cards were better than traditional tables.
1
Implement Plain Language
Focus on use of Plain Language for clear and concise writing to increase usability
2
Increase Readability & Efficiency
Focus more on readability, findability, and quick-to-perform actions where demanded
3
Ensure Intuitiveness & consistency
Focus more on readability, findability, and quick-to-perform actions where demanded
High level information architecture
"I considered prototyping as the most effective means of receiving substantial feedback"
Iterations
Usability Testing
I conducted usability testing with 20+ participants, resulting in an 96% task completion rate, 84 SUS score, and 3.5 min time on task, compared to 15 mins for competitors. I involved every key stakeholder in my design process beforehand to create a smoother design hand-off.
Example of user testing results with a data engineer
The Design System
I discovered a need for a card component in the HPE Design System while working on the Unified Analytics product. After investigating, I found no specific component for it, and designers were detaching existing components and seeking help without proper guidance or documentation. The HPE Design System was still maturing at that time which led to my contribution of the card component.
To address this, I collaborated with other designers, developers, and product owners to research use cases and analyze other design systems. I defined the needs, guidelines, variants, accessibility, and responsiveness with an eye for detail.
As a full-stack designer with a technical background, I tested the components with designers and handed off my work to the developers. Later on, I worked with developers as well to look at how things are in the code base. The card component I created increased productivity and efficiency for designers, engineers, and product managers, reducing 20 hours of work per month per employee. It increased end-user adoption by 10% and user retention for over 300k customers across multiple products, while also improving the designer-developer handoff by eliminating broken design tokens and documentation. I will be going into detail on this in a separate section.
