Redesigning Reports around User Mental Models
MTA is a financial trade regulation and analytics platform. It was released in early 2018, coinciding with new European financial regulations for the 2017 filing deadline. Since then, the platform has expanded to include SEC regulations and trading analytics.
Process: User Research | User Experience Design | User Testing
Role: Researcher | Designer | Test Moderator
Original Query Builder
One of our post-MVP goals for this brand new regulation technology platform was to redesign the custom report builder. In its current iteration, the feature was sold as a “Query Builder” and it was clear the original goal was to create a UI around building a query string and running it against the data in the database.
Research
I started by researching with our client services team to understand who understood how to use the Query Builder internally and who knew how to use it externally. I quickly found that only one user actively used it, a client services lead who writes custom SQL reports for clients.
In my research I discovered three main findings:
Our user base does not know SQL. They do not know what a ‘query’ is. (This was a surprise to internal stakeholders)
Users are not looking to build custom reports often or on the fly, they want to build them once and set a recurring schedule.
Users think of the data in segments and generally follow the same order: Identifier, Date & Price, and then Attributes
Design
I started by exploring different ways of organizing the available data around sub-classes. The current process had two large groups:
Category (data provided by the client)
Metric (Data calculations provided based on our reference data)
As we progressed through the designs, we moved further away from the current distinctions and allowed the sub-categories to become their own sections:
Identifiers
Date
Price & Size
Trade Attributes
As we made these decisions, we continually tested with our Client Services team to make sure we were creating something they could use. They found these new categories were much easier to use and loved that they were guided in an intuitive way through the report building process. We also discovered that filters needed to be redesigned because they were built on the same SQL design and were not designed with any user understanding.
Testing
As I was designing, i decided to do a formal usability test with the basic design direction to make sure the design would work. I built a prototype in Axure and ran three users through creating reports. All three found the process a huge improvement over the current and design and were excited to see it be built.
