Designers
Stefan Ecker
Year
2025
Category
Product
Country
Austria
Design Studio / Department
George Labs GmbH

Three questions to the project team
What was the particular challenge of the project from a UX point of view?
Users are charred by bad chatbot service implementations, which are all around us. So our main challenge in terms of UX was to regain their trust. This can only be achieved by replacing those bad memories with memorable moments, which make our users think “Wow, that was helpful” and return to use “Hey George” again. Also, where do you even start designing for this plain playing field, where a conversation between a user and AI can go in any direction? We had to put a lot of effort into exploratory research to find the topics and questions which users would ask, and what data sources we need, in order to answer those questions to a level that exceeds their expectations.
What was your personal highlight in the development process? Was there an aha!-moment, was there a low point?
When we reached 1 million users in the beta rollout, and we saw more reliable tracking data coming in, it became clear that the topics and questions from the exploratory user research matched up with what users want to know in the real world. Even though the ROI of UX has been proven many times, that feeling that we focused on the right things never gets old. For example: We knew that cost and spending analysis will be the most asked AI skill, so we focused on training our models with the transaction data of our users, and how to give the best answer and visualisation for a myriad of possible questions. It was very satisfying that this effort paid off and to see the positive trends in our answer rating.
Where do you see yourself and the project in the next five years?
AI systems are evolving so fast, so it is difficult to predict the next 5 years. For us as a bank compliant advisory via a fully AI driven chat bot is not possible right now, but we aim to make that happen in the future. Also in 5 years time we would like to have conversational AI much more integrated into all workflows, and not as a separate area within the app, where users can talk, but conversations and AI in general as a fully integrated, cohesive experience across various aspects of our user’s financial lives.


