Leading telecommunications provider VodafoneZiggo (VFZ) has partnered with Xomnia to transform its customer service using state-of-the-art Generative AI (GenAI) applications across several use cases. Our GenAI Lead, Fried Schölvinck, brought his expertise to VFZ, stepping in as a technically-focused Product Owner and introducing Xomnia’s approach to align their work with clear business objectives.
Through our collaboration, VFZ successfully deployed multiple GenAI applications, including a customer-facing RAG-based search functionality that is currently live on their website. This solution aims to improve user experience and accessibility, ultimately boosting customer satisfaction while reducing the workload on support teams.
"Xomnia helped us to strengthen our team with the essential knowledge on scaling our proof-of-concepts further to capture benefits that GenAI brings for our agents and customers." - Mustapha Abbach, Cluster Lead COPS (Customer Operations) for Data & AI
Challenge
To continue modernizing its customer service capabilities, VFZ sought ways to support its agents with Large Language Models, aiming to reduce customer wait times and enhance accuracy and efficiency. VFZ’s Data & AI department set out to build two initiatives: Customer Assist and Agent Assist—targeting both direct support to customers and assistance for call center agents.
The Agent Assist team focused on developing AI products that provide real-time insights to agents, help during and after calls, and generate automatic call summaries. The Customer Assist team began by building a chatbot using Retrieval Augmented Generation (RAG) with public data from the Ziggo website, community forums, and related subdomains.
Initially, the broad scope and fast pace of development presented several distinct challenges. One major hurdle was the adoption of GenAI in customer service, which is inherently complex due to the need for business alignment, integration with both front-end and back-end tooling, managing numerous stakeholders, and meeting high customer expectations. Furthermore, previous attempts at implementing customer service chatbots with older NLP techniques had been unsuccessful, leading to a decline in NPS (Net Promoter Score, which measures customer loyalty and their likelihood of recommending the business). In addition, the old search functionality on VFZ's website was ineffective, often failing to rank the most relevant web pages accurately. To address these challenges, VFZ partnered with Xomnia, bringing in Fried Schölvinck to provide GenAI expertise and direction.
Solution
Fried joined VFZ as a GenAI Product Owner, acting as a bridge between the business and technical teams. He helped align all stakeholders towards the shared goal of delivering effective GenAI products for Customer Operations.
With his deep technical understanding, Fried worked closely with the developers, challenging them to design the best solutions to meet customer service needs. He facilitated collaboration between different teams and drove a more focused approach.
After contributing to the AI assistant tools for call center agents, Fried joined the Customer Assist team. “We worked a lot on scoping and alignment to understand what was needed to launch the application,” Fried explained. “This process led us to pivot from the original interactive assistant concept to an AI search function.”
The team then focused on quickly releasing a minimum viable product (MVP) to gather user feedback, iterate based on real-world data, and progressively add advanced features. This agile approach enabled a flexible, responsive development process that adapted to user needs.
Impact
With Xomnia’s support, VFZ is building a strong foundation for customer-facing GenAI products. Over the past year, VFZ’s Data & AI department gained valuable insights and is now equipped with the right tools and knowledge for future innovation.
After overcoming initial adoption challenges, VFZ’s technical team launched an AI search product and is motivated to continue improving it. The new functionality captures the semantic meaning of customer queries, providing more relevant information and reducing pressure on customer service teams.
With smarter AI tools, VFZ aims to enhance customer support while streamlining operations for their customer service teams.