YipitData
Data Vizualization
Context
We observed a high volume of custom data analysis requests that weren’t covered by standard reports. While data-savvy teams downloaded raw files to create their own visualizations, less data-savvy teams relied on us for custom data cuts, often missing out on critical, time-sensitive insights. To address this, there was a clear need to develop dashboards that empowered customers to visualize data in their preferred way while also reducing the workload on our internal teams, who spent countless hours generating custom analyses.
What did I do?
Led user research to gain a deeper understanding of customer needs and collaborated with the product manager to visualize the data. Additionally, developed a long-term strategy to enable teams to create dashboards independently, reducing reliance on design.
Understanding our customers
I reviewed over 100 custom analysis requests from customers and watched user recordings of research analysts speaking with them to understand their questions, challenges, and recurring patterns. Applying the 80-20 principle, I distilled key themes and worked closely with the product manager to define the next steps.
One of the many affinity diagrams I created to identify key themes
Building the Key Metrics dashboard
We designed the visualization around three key user needs
Key Metrics Overview: Providing a clear, high-level snapshot of essential KPIs for quick insights.
Interactive Forecasting: Empowering customers to project future trends by selecting custom growth rates.
Comparative Analysis:
Using visuals to help users easily interpret daily comparisons and historical trends for better decision-making.
Establishing Visualization Principles
The Key Metrics Dashboard quickly gained traction, leading to a 45% increase in user activity as more research teams adopted it to create their own dashboards. Teams saw it as a more efficient way to publish data more frequently and expand coverage across sectors and companies. However, while teams understood the questions they wanted to answer, they lacked clarity on usability best practices and selecting the right visualizations.
Recognizing this gap, I developed the Data Visualization Guidelines—a resource to help research teams design more effective, user-friendly dashboards. These guidelines also supported engineering by introducing streamlined components, reducing the need to build dashboards from scratch and improving overall efficiency.
