Customer service organisations operating across multiple countries often deal with fragmented data streams and slow manual processing. Mazda Motor Europe and Datasolut tackled this challenge by creating a solution that centralises the processing of multilingual customer inquiries.
The solution automatically translates feedback collected in 12 Mazda European Customer Relations Centres from any language into a consistent format, ensuring that insights do not depend on the language of the original submission. It applies standardised topic classification across all markets, enabling consistent comparison of inquiry types across countries.
In addition, integrated sentiment analysis helps detect customer mood, emerging patterns, and changing trends. Because the entire workflow runs in near real time, insights become available immediately, reducing manual effort and enabling faster market response. This unified approach supports earlier trend detection, cross‑market comparison, and more informed business decisions.
Technology Meets Cross‑Functional Collaboration
The success of the solution was not driven by technology alone. It resulted from close collaboration between Mazda Motor Europe’s analytics team and Datasolut’s data engineering experts. At the outset, both teams invested time in clearly defining the business problem before approaching automation, ensuring that the solution addressed genuine pain points in customer service operations.
A strong emphasis on data quality and shared definitions allowed the Mazda Motor Europe team to work from the same understanding of key terms and categories. Throughout development, the use of transparent and explainable AI methods helped build trust among business stakeholders. The resulting solution demonstrates that AI enhances, rather than replaces, customer service expertise by removing repetitive tasks and enabling employees to focus on higher‑value, customer‑focused activities.
“AI does not replace human expertise in customer service,” says Thomas Fontaine, Senior Manager Consumer Analytics at Mazda Motor Europe. “It strengthens it – by freeing people from manual work and enabling them to focus on customers and better decisions.”
Designed for Real‑World Enterprise Use
“This project shows what is possible when data engineering and business expertise work hand in hand,” says Susanne Heinrichs, Senior Data & AI Engineer at Datasolut. “Mazda’s clarity on the problem and our experience with scalable AI platforms allowed us to jointly build a solution that transforms complex customer feedback into usable insights within seconds.”
The solution developed by Mazda Motor Europe and Datasolut will be in production soon and was designed from the start with real‑world enterprise requirements in mind. It is fully scalable across different markets, teams, and data sources, allowing new languages or customer service channels to be added seamlessly as business needs evolve.
The project was showcased at the Data Unplugged Conference in Münster, Germany on March 26, 2026, where Thomas Fontaine and Datasolut’s Senior Data & AI Engineer, Susanne Heinrichs, presented the tool’s architecture, lessons learned, and practical impact.
