Improve Information Architecture and UI of a Conversational Data Analysis Tool

Problem

The analytics tool was failing to meet users' needs. While the back-end capabilities were in place, the tool lacked a user-centered approach in terms of how the data was displayed and which data was most relevant to show.

Skills Used

Research | Stakeholder Management | User Personas | User Journeys | Design/Dev Collaboration

The Story

Cue AI is a web application that analyzes conversational data from live chat and chat bot conversations. Cue AI graphically delivers reports ranging from conversation topic, channels used and conversation sentiment. Cue AI gives the analyst the ability to explore the data according to their needs and analysis goals.

My role: Visual Designer, User Experience Designer
Platform: Web

Summary

After the Cue product team demoed their analytics tool to the analyst group (their primary users). I saw an opportunity to help the company make a tool that could not only improve the quality of the analysis work, reduce time spent on analysis, and provide a tool that gives analysts (and in the future, our customers) a tool they need to get the information the way they want. I offered my time and services to help them, wherever needed.

It started out with a few help requests from the development team here and there and then I convinced a skeptical product owner of the value of talking through some personas and user journeys.

Hindsight observation: I could have improved the questions I asked and should have dug deeper into what problems the analyst users and the development needed solving.

I formed a good relationship with the Cue AI development team on UI improvements and defining their end users. I provided the reasonings behind my decisions and what I look out for so in design and interactions that they were empowered to do the same. They expressed relief in not having to make design decisions on their own.

I took this on as a side project, which meant that I only had a limited amount of time to dedicate to the project; the goal was to make UI improvements that would improve the visual experience and not be time consuming to implement. The team was also looking for user experience feedback and direction. I was most interested in trying to understand what the analyst users wanted and to make sure we were delivering a solution for them.

I created high-fidelity mockups and prototypes and delivered the necessary specs and assets. I worked with the product owner to explore user profiles and user journeys through internal knowledge and internal user research on our primary users. This informed me of the world they were working in and was also useful for team alignment.

Notable

The Cue AI product owner wrote:

“Despite the fact that she has plenty of other tasks on her plate, Renae volunteered to spend time to design the user interface for Cue. The difference in our end result is enormous with her help versus what we are capable of producing on our own. She is making the difference between a useful application that looks somewhat acceptable and one that actually looks compelling and usable at first glance. She’s taken the initiative to work with Verint’s UX team, and if you take a look at the northwest corner of the parking lot her truck is there all the time.

The development team and I are going to look a lot better because of the work that she’s done to contribute to the project... and it’s not even part of what she’s required to do for her day job. She’s spending a lot of time to help us out and improve her skillset in this area, and it shows. I really appreciate everything she’s done and how easy she is to work with, and she’s having a significant positive impact on our team’s output and general morale. I hope that she’s able to continue to help us make progress with regards to usability and UI cleanliness across our portfolio.”