Company: Blackbird.ai

Product: B2B, SaaS, Web application

Areas: Business Intelligence, AI/ML, Data analysis

Year: 2024

This case study aims to illustrate my end-to-end design process, emphasizing the importance and value of UX Research in risk reduction. This approach allows for identifying pain points and shaping solutions based on evidence that enhances the user experience and aligns with business objectives.

Challenge

I was the Product Designer responsible for this project, partnering with a PM and a Tech Lead at Blackbird.ai. Blackbird's main platform is Constellation, which for this particular project the focus were on two pages: Analyze and Network Graph.

Users struggled to understand why the Network Graph didn't display the same Twitter metrics as the Analyze page, which caused confusion and frustration.

Goal

The goal was to reduce user confusion and frustration related to the data mismatch between the Analyze and Network pages and reduce the number of customer support tickets created within this topic.

Research

Translating data structure in Constellation

My first step was understanding how the data is structured in Constellation as a whole and the differences between these two pages.

I discovered that:

• Users specify Data Collection rules when adding data to a project in Constellation.
• At the Data Processing phase, this data added is considered 'Local Data', and to provide additional context to it, Blackbird also add into the project 'Extended Data' and 'Global Data'.
• The Analyze page, displays booth Local and Global data. However, the Network Graph uses Local Data exclusively to create the visualization, formed by Nodes and Edges.

• Users expect to see the same data between booth pages, but this does not happen, leading to unmet expectations and user frustration.