Utilities2021Machine Learning (classification)NLPB2C
Stadtwerke Düren

German energy utility Stadtwerke Düren automates 55% of customer inquiries with NorBot AI chatbot

Integrated within hours using OMQ's AI chatbot platform, NorBot automatically resolves more than half of Stadtwerke Düren's customer inquiries in real time, reducing agent workload and eliminating peak-period backlogs.

Inquiry Resolution Rate55%
Integration TimeHours
3 min read

Background

Stadtwerke Düren's customer service team was fielding recurring standard inquiries that could be handled without agent expertise. During year-end periods, inquiry volume spiked significantly as all customers submitted meter readings and received annual bills simultaneously. The existing live chat system provided no automation layer to absorb this demand.

What Was Implemented

  • OMQ AI chatbot (NorBot) integrated with existing Lime Connect (Userlike) live chat system within hours
  • Available on website, WhatsApp, and Facebook Messenger; 24/7 availability
  • AI knowledge database enabling intent recognition across varied phrasings of the same question
  • Human escalation available during service hours when chatbot cannot resolve a query
  • Named "NorBot" after the Head of Customer Service to give the bot a human identity

Results

55% of all customer inquiries are resolved automatically by NorBot without agent involvement. This relieves agent workload significantly for recurring standard questions and eliminates the pressure of peak inquiry periods at year-end. Both customer satisfaction (instant answers) and agent satisfaction (freed from repetitive queries) are reported to have improved.

Lessons

  • A chatbot achieves high resolution rates fastest when it starts with a well-maintained, structured knowledge base of the most common questions — not a blank-slate machine learning system
  • An instant, hours-long integration timeline is achievable when the chatbot platform connects to an existing live-chat infrastructure rather than requiring independent deployment
  • Naming and personalizing the bot (NorBot after "Norbert") creates a local identity that can improve customer acceptance
  • Seasonal inquiry peaks are ideal use cases for chatbot absorbing demand spikes without requiring additional seasonal staffing

Ready to implement AI in your commerce operations?

McFadyen Digital helps teams move from case study to live implementation.

Talk to an expert →