HEALTH & WELL-BEING
On-Call Load Distribution
The practice of measuring and equalizing the amount of time and effort each team member spends on on-call duties.
On-Call Load Distribution
The practice of measuring and equalizing the amount of time and effort each team member spends on on-call duties.
Measure the Pain
Use data to balance the load. If Person A dealt with 10 incidents and Person B dealt with 0, assign Person A fewer shifts next quarter.
Metric: Pages per shift.
ExLoad Balancing by Incident Severity
"Team tracked "incident burden score" (SEV0=10pts, SEV1=5pts, etc.). Found senior engineer had 2x the average burden due to complexity. Adjusted schedule: gave senior engineer lighter weeks after high-burden rotations. Result: even distribution within 2 quarters."
Impact
Reduced single point of knowledge dependency
Resolution
Load balancing improved knowledge sharing and reduced bus factor risk
Why On-Call Load Distribution Matters
Data-driven approach to fairness.
Identifies when the team is understaffed.
Common Pitfalls
Only measuring time on-call, not incident burden
A week with 10 SEV0 incidents is not equal to a week with 0 incidents. Track incident volume and severity, not just calendar time.
Ignoring the "on-call tax" on career growth
Account for on-call time in performance reviews. Don't penalize people for having fewer side projects due to on-call recovery time.
Related Terms
Frequently Asked Questions
What metric should I use for on-call load distribution?
Use multiple metrics: total incidents, incident severity score, nights awoken, total on-call hours, and holidays covered. Combine into a single "burden score" for comparison.
How do I balance load unevenly across the team?
Not everyone can do equal on-call. Balance by total contribution: code reviews, documentation, mentoring, and project work count alongside on-call. Recognize different ways people contribute.
What if some people are better at on-call than others?
Skill imbalance is a coaching issue, not a scheduling issue. Train weaker responders, don't just load the strong ones. Over-relying on heroes creates single points of failure.