Visualising the user journey within the 'Ceba' chatbot for CommBank

Client: CommBank

Problem: When the Ceba chatbot was launched within the CommBank app, it quickly became a success in terms of user experience. However, optimising the chatbot proved challenging due to the lack of visualisation tools that could map the user journey. Without a clear understanding of how users navigated between different intents and guided flows, it was difficult to identify pain points and areas for improvement.

What I Did: Leveraging my expertise in data and analytics, I developed a comprehensive bot taxonomy using a unique tagging system. This system created data points that made sense in sequence, allowing us to map out and visualise the order of customer events. By doing so, we gained a clearer understanding of customer behaviour.

Building on this foundation, I designed custom dashboards that incorporated Sankey diagrams to illustrate the end-to-end user journey. These visual tools provided invaluable insights into how users interacted with the chatbot.

Results: The visualisation of user journeys revealed significant overlaps between similar intents and highlighted redundant journeys. Based on these insights, we undertook several redesigns that resulted in a 12% increase in containment and an 8% improvement in customer satisfaction (CSAT).

Furthermore, the dashboard became a critical tool for continuous improvement. It helped identify areas where new chatbot intents and flows were needed, thereby expanding Ceba's scope and enhancing its overall effectiveness.

This project not only optimised the performance of the Ceba chatbot but also provided a scalable model for future improvements, ensuring that CommBank could continue to meet and exceed customer expectations.