Opportunity
Explore how AI could help Swissmilk improve recipe discovery and customer engagement while validating the long-term viability of conversational AI within their digital ecosystem. As the project evolved, the focus shifted from experimentation to building a scalable, high-performing AI product capable of delivering measurable business value.
Challenge
Following an initial chatbot launch, the project faced challenges around performance, scalability, user adoption, and long-term sustainability. The client needed a clear understanding of whether the solution could evolve into a viable product and how to maximize the value of their existing investment.
Solution
Rather than focusing on incremental fixes, we evaluated the chatbot as a product. The solution combined prompt optimization, multi-agent workflows, performance analytics, and a strategic infrastructure assessment to improve quality while defining a scalable roadmap for future growth.
Impact
Stabilized a project facing performance concerns
Maintained stakeholder confidence and avoided escalation
Preserved approximately €120K in client investment
Secured buy-in for a new scaling strategy
Established a roadmap for improved performance and lower operating costs
Repositioned the chatbot from a pilot experiment into a long-term product opportunity
Led the evolution of Swissmilk's AI chatbot from proof-of-concept to strategic product evaluation, combining AI optimization, analytics, and executive consulting to improve performance, protect client investment, and define a scalable path forward.
What I did
Facilitated discovery workshops and defined chatbot use cases
Designed prompt architecture and conversational logic
Optimized chatbot performance through prompt refinement and multi-agent workflows
Built a tracking and analytics dashboard to monitor adoption, performance, and user behavior
Evaluated technical, UX, and operational limitations of the existing solution
Analyzed chatbot data and identified opportunities for scaling and improvement
Developed and presented executive-level recommendations for future infrastructure and product direction
Acted as the primary bridge between client needs, business goals, and technical implementation teams
