Major Housing Association
45,000 homes • West Midlands
How AI-powered triage transformed housing repairs and complaints handling, saving £180,000 annually while improving tenant satisfaction.
Key Results
The Challenge
The housing association was receiving over 2,000 repair requests, complaints, and service enquiries per week across email, web forms, and phone. Manual classification was time-consuming, inconsistent, and often led to incorrect routing—causing delays and tenant frustration.
Senior housing officers were spending up to 40% of their time on initial triage rather than complex casework. During peak periods, backlogs would build, leading to missed SLA targets and increased complaints.
Key challenges identified:
The Solution
We built an intelligent triage system using Azure AI Document Intelligence, trained on 50,000+ historical cases. The system automatically classifies incoming requests, assigns priority scores, and routes to the appropriate team—with human review for edge cases.
14 weeks
Discovery to go-live
Azure AI
Microsoft ecosystem
UK hosted
Data sovereignty
Project Timeline
Data audit, process mapping, success criteria definition
Model training, integration build, testing environment setup
Limited rollout, accuracy monitoring, threshold tuning
Organisation-wide rollout, staff training, go-live support
The Results
“The AI triage system has transformed how we handle incoming requests. Our housing officers now spend their time on complex cases that need human judgement, not routine classification.”Director of Customer Services
Get Your AI Roadmap to discuss how AI-powered triage could work for your organisation.