3.4 - People metrics: Tracking team health and turnover risks
Engineering Managers love metrics. We track CPU usage, database latency, and API error rates with extreme precision. However, when it comes to the humans building the software, we often rely entirely on "gut feelings".
As a Tech-Led Coach, you must apply engineering rigor to human operations. You can use quantitative metrics and operational sensors to measure qualitative human traits, allowing you to catch burnout, frustration, and turnover risks long before someone hands in their resignation.
Sensor 1: Code review turnaround time
Most teams measure how many Pull Requests (PRs) are merged. That is a vanity metric. Instead, measure Code Review Turnaround Time - the hours or days a PR sits open waiting for approval.
- The Human Signal: If PRs sit open for days, it is rarely a tooling problem. It signals a breakdown in team culture. It means engineers are working in isolated silos, do not feel responsible for collective goals, or are too burned out to help their peers.
- The Fix: Shift the team's definition of success from "I finished my ticket" to "We merged our code".
Sensor 2: eNPS (Employee Net Promoter Score)
eNPS asks a simple, anonymous question on a regular cadence (e.g., quarterly): "On a scale of 0-10, how likely are you to recommend working on this engineering team to a friend?"
- The Human Signal: eNPS gives you a macro-level pulse of psychological safety and morale. A sudden drop in eNPS usually correlates directly with a recent shift in company policy, a forced top - down deadline, or accumulating technical debt that is causing daily friction.
- The Fix: Do not just collect the data; act on it publicly. Say, "Our eNPS dropped this month because of the legacy database migration. Let's discuss how we can make the next phase less painful".
Sensor 3: 1:1 psychology logs
Your 1:1 meetings are your micro-level sensors. Track the sentiment of these meetings over time.
- The Human Signal: Watch for the "Apathy Shift". If a previously vocal, passionate engineer suddenly stops complaining about bad code or broken processes, they are not happy - they have given up. Apathy is the final stage before resignation.
- The Fix: Confront the shift immediately. "I noticed you used to champion our refactoring efforts, but you have been quiet lately. Do you feel like your voice isn't being heard?"
By combining system data (PR turnaround) with human data (eNPS and 1:1 sentiment), you build a comprehensive dashboard for team health.
🎯 Self-Reflection for You
- If your best engineer resigned tomorrow, would you be completely surprised, or were there subtle metrics and signals you ignored?
- How long does a PR stay open on your team, and what does that tell you about your team's collaborative culture?
Learning to read human metrics is a critical management upgrade. If you want to set up an operational dashboard for your team's health, let's map out your sensors in a virtual coffee chat.