TELUS unifies 15 data sources with Adobe Customer Journey Analytics to proactively reduce call volumes
By stitching together 15 siloed data sources — including web, app, billing, chatbot, and call center — into a single customer journey view, TELUS identified that call spikes occur 3–4 days after bills are sent and implemented targeted push notifications to address billing confusion before customers call.
Background
High call volumes at telecom contact centers represent a significant operational cost. TELUS recognized that many calls were predictable and potentially preventable — the challenge was identifying the patterns that predicted inbound contact. Without a unified customer journey view, this analysis was not possible.
What Was Implemented
- Implemented Adobe Customer Journey Analytics (CJA) as the unified analytics platform
- Configured 8 datasets from 15 online and offline data sources (web, app, voicebot, chatbot, billing, call center, product activation, MyTELUS app)
- Built custom SQL user stitching to create unified customer identifiers across sources with no common ID
- Created a translation layer for business users to access journey analysis without engineering dependency
- Identified key journey patterns: bill spike call timing (3–4 days after billing), post-activation contact behavior, and same-day digital-to-phone escalation patterns
- Deployed proactive push notifications and in-app messages for customers flagged as billing anomaly risks
- Extended analytics capability across additional TELUS business units
Results
TELUS gained visibility into previously invisible customer journey patterns. The most operationally impactful insight: call volumes spike 3–4 days after bills are sent , especially when bills exceed the prior 3-month average — enabling a proactive push notification intervention. A substantial share of annual phone calls also originates from customers who visited the app or website within the prior 24 hours, pointing to addressable digital self-service gaps. Customers are significantly more likely to contact support on their first day after wireless product activation, enabling targeted onboarding improvements. The case study does not quantify the reduction in call volume, cost savings, or NPS change resulting from these interventions.
Lessons
- Unifying online and offline customer data reveals non-obvious behavioral patterns (e.g., bill spike timing) that cannot be identified from any single system alone
- Proactive push notifications triggered by journey analytics events convert reactive call center interactions into preventive digital engagements — reducing cost while improving experience
- Custom user stitching is required when data sources lack a common identifier; SQL-level customization above the platform layer is often essential in telecom environments
- Enabling business users to perform journey-level analysis without engineering dependency dramatically accelerates insight-to-action cycles