A provider of connected digital solutions for service shops offers integrated platforms that combine lead management dashboards (e.g., quoting, service headcount, open leads per distributor) with real-time equipment telemetry from systems such as vehicle lifts, ADAS calibration tools, and A/C stations. These solutions support core shop management functions, including inventory ordering, equipment diagnostics, and performance analytics.
As service shops became more dependent on digital workflows, the platform began to experience performance degradation during core business hours. Key issues included:
- Slow dashboard load times affecting technician efficiency
- Periodic timeouts when retrieving telemetry from hundreds of connected IoT devices
- Sluggish diagnostics and reporting on aggregated sensor data
A comprehensive performance engineering approach was implemented:
- Lead dashboard metrics (open leads, quotes, orders)
- Spike tests during business hours; 24-hour soak to detect memory/resource leaks
- Batching equipment polling with an adaptive pull strategy
- Dashboard metrics rendered in <1.5 sec for >90% of sessions under peak
- Telemetry ingestion throughput rose by 65%, with zero processing backlogs
- Completed a 24-hour soak test with stable memory, CPU, and response latencies
- Error rates dropped from 4% to <0.5% under peak and stress conditions
- Infrastructure costs were reduced due to optimized autoscaling and resource tuning