Business Context

NovaCloud is a B2B SaaS company with 55 engineers across 6 squads shipping a workflow automation product. Over the last two quarters, customer-reported bugs increased from 18 to 31 per month, average feature cycle time rose from 9 to 14 days, and voluntary engineering attrition increased from 6% annualized to 11% annualized.

The CTO asks you to define a concise metric framework for engineering team health. The goal is not to create a single vanity metric, but to identify the core KPIs, how they should be balanced, and how to diagnose whether the team is healthy, overloaded, or slowing down. Stakeholders want a monthly dashboard and a clear explanation of which metrics are leading indicators versus lagging indicators.

Requirements

1. Define 5-8 metrics you would use to measure engineering team health, including precise definitions and why each matters.
2. Separate the metrics into productivity, quality, reliability, and people/organizational health.
3. Identify which metrics are leading indicators versus lagging indicators.
4. Propose how you would combine the metrics into an executive view without encouraging gaming.
5. Explain how you would diagnose a situation where output is rising but quality and retention are worsening.

Data Available

• jira_issues: ticket_id, team_id, issue_type, story_points, created_at, started_at, completed_at, reopened_flag
• github_prs: pr_id, repo, author_id, opened_at, merged_at, lines_changed, review_cycles, deployment_id
• incident_log: incident_id, severity, service, opened_at, resolved_at, root_cause_category
• bug_reports: bug_id, source, severity, created_at, resolved_at, linked_release
• employee_hr: employee_id, team_id, start_date, termination_date, engagement_survey_score, manager_id
• oncall_events: engineer_id, alert_count, after_hours_pages, week_start