Explain why support backlog rebounds after overtime
Build an evidence-qualified feedback hypothesis and compare guarded leverage tests before repeating the fix.
35 minutes
Scenario
A software support team repeatedly uses overtime when backlog rises. Backlog falls for two weeks, rebounds higher, and attrition rises three months later. Leaders plan another overtime campaign.
- Your role
- Service leader responsible for capacity, customer outcomes and workforce safety
- Method
- Causal Loop Diagramming
Evidence pack
Reference mode
Backlog spikes quarterly, falls 18–25% after overtime, then exceeds its prior level within 4–6 weeks.
Quality
Reopened tickets rise after the second overtime week and generate additional demand.
Capacity
Experienced-agent hours fall as sickness and attrition rise with a 6–14 week lag.
Hiring
Approved hiring adds productive capacity after a median four months because training is required.
Demand
Thirty-two percent of new tickets follow one confusing product workflow.
Guardrail
No intervention may exceed the existing safe-hours policy or reduce priority accessibility support.
Constraints
- Start from the reference mode and time horizon.
- Treat worker fatigue and agency as substantive variables.
- Keep correlation and causal evidence distinct.
- Do not use a qualitative CLD as a forecast.
Case steps
Work through each prompt using the evidence pack. These guided cases support self-directed practice; server-scored attempts are not available yet.
Define variables and causal links that explain the decline and rebound.
Close and label the important reinforcing and balancing loops, including alternative accounts.
Design a portfolio of leverage tests with safeguards and rejection signals.