Leveraging UX Research and Applied Data Science Models to Unlock Insights
Company: Easynvest | Role: Product Design & Research Manager | Year: 2020
Easynvest, the leading digital investment broker in Brazil, had been relying mainly on quantitative data to make decisions about their product. We understood that we needed to go beyond the standard persona creation process and get a more accurate understanding of the needs and pain points of our customers. We also recognized the importance of measuring the size of each of these user profiles within our customer base to direct our business strategy.
To address these challenges, we launched a user profiling project that involved in-depth research with users and the mapping of behavior variables that were used in data science models to identify these profiles within our user base. This project was a joint effort between our UX research and data science teams.
The UX research team conducted extensive interviews and surveys with our customers to gain insight into their motivations, challenges, and
decision-making processes when it came to investing. These insights were then mapped to behavior variables such as investment frequency, investment amount, and risk tolerance.
The data science team then used machine learning models to analyze the data and identify clusters of users with similar behavior variables, creating user profiles that went far beyond traditional personas. The models were trained on a large dataset of user behavior and applied to the customer base to segment users into specific profiles. This allowed us to identify the size and characteristics of each user profile in our customer base, providing us with valuable information to direct our business strategy and product development efforts.
We were able to identify and quantify the most common pains and needs of their users and develop strategies to address them. The data also allowed us to make more informed product decisions and prioritize development
us to make more informed product decisions and prioritize development efforts based on the needs of their users.
Additionally, the project had a significant impact on the culture of the organization. The company developed a deeper understanding of the users, and integrated that information into daily routines.
The project allowed us to gain a much deeper understanding of our customers and their needs. By combining UX research and data science, we were able to create accurate user profiles and use them to direct our business strategy and product development efforts. This project has been also used as one of the foundations for decision making when Nubank, one of the world's largest digital banks, acquired Easynvest operation.
Collaboration roadmap between UX Research and Data Science teams
Main variables used to define the profiles
Final clusterized data science model used to identifiy profiles on our customer base
A total of 7 differentes profiles were identified
Our one-page description of a model profile for easy access and reading