On-Call Health

On-Call Health is a free, open-source tool that gives engineering managers a data-driven way to detect and prevent burnout in on-call teams. Developed by Rootly, it pulls signals from PagerDuty, GitHub, Linear, Jira, and Rootly itself to build a continuous picture of workload, incident volume, and after-hours activity. But what makes it distinct is that it doesn’t stop at technical metrics—it also collects subjective sentiment via automated Slack surveys, then combines both into an individualized risk score from 0 to 100. This hybrid approach is its real strength: it acknowledges that burnout is not purely a function of pagers or pull requests, but of how engineers feel about their workload. For engineering managers who have struggled to move beyond gut feelings and anecdotal reports, On-Call Health offers a structured, repeatable method for spotting trouble before it becomes a crisis.

The tool fits naturally into organizations that already use modern incident management and development platforms. If your team runs on PagerDuty, manages tasks in Jira or Linear, and communicates in Slack, On-Call Health can be deployed with minimal friction. It’s especially valuable for SRE and DevOps teams where on-call burden is high and rotations are frequent. The open-source license (Apache 2.0) means there’s no licensing cost and the code can be audited or customized—a significant advantage for security-conscious or budget-constrained teams. However, its value depends heavily on the breadth of integrations you have in place. Without a robust toolchain, the signal pool will be thin, and risk scores may lack context. Similarly, the sentiment surveys rely on consistent participation; if engineers ignore or rush through them, the subjective component becomes noise. Managers should also note that On-Call Health is an early-warning system, not a remediation tool—it alerts you to problems but doesn’t automatically rebalance rotations or pause work. That decision still rests with human judgment.

For practical use, the tool excels in three scenarios: proactive burnout prevention, post-incident health checks, and rotation optimization. A manager might see a rising risk score for an engineer who has been handling multiple incidents while also pushing code changes. That signal, combined with a declining sentiment score, justifies a conversation and a rotation adjustment before the engineer reaches exhaustion. After a major incident, team health baselines can reveal elevated stress levels across the board, prompting a temporary pause on non-urgent work. And over time, comparing individual scores helps ensure on-call duties are distributed fairly. The AI-powered trend summaries add another layer: they condense weeks of signals into a sentence or two, helping managers stay informed without manual data crunching. Ultimately, On-Call Health is a pragmatic, low-cost addition to any engineering org that takes developer wellbeing seriously—as long as you’re willing to act on what it tells you.