Policy : Prioritise Safety Before Productivity

Commitment to Safety as a First Principle
We believe that nothing is more important than the safety of our systems, our customers, and our teams. Productivity must never come at the cost of reliability or wellbeing. Safe systems are the foundation for sustainable delivery, and safe teams are the foundation for resilient culture.

What This Means
We give engineers the authority and support to stop work if safety is at risk—whether that’s a broken build, a failing test, or a risky deployment. We build systems that surface issues early, guide safe behaviour by default, and prioritise learning over blame when risks are flagged.

Our commitment to prioritising safety over productivity is built on:

• Release Gates and Rollbacks – Releases can be halted or rolled back automatically or manually when quality thresholds are not met. Shipping is conditional on safety, not pressure.
• Stop-the-Line Culture – Engineers are supported in stopping work (e.g. pausing a release, raising an incident) without fear of blame. Safety trumps deadlines.
• Proactive Observability – Observability tools are used to detect and investigate anomalies early. We look for unknown unknowns and address early warning signs before they escalate.
• Service Level Objectives (SLOs) – SLOs guide delivery decisions and pace, making it clear when to slow down, fix issues, or defer features to protect system integrity.
• Safe-by-Default Platform Services – Identity, encryption, alerting, and logging are built into platforms as defaults—not optional add-ons—making it easy to do the right thing.

Why This Matters
Pushing for delivery at the expense of safety creates brittle systems, anxious teams, and avoidable outages. When teams don’t feel safe to raise concerns, quality erodes and risks compound. By prioritising safety, we reduce long-term cost, protect trust, and create healthier engineering environments.

Our Expectation
All teams must treat safety as non-negotiable. This includes honouring quality thresholds, responding to SLO breaches, and creating space for safety conversations. Leaders must model this by supporting decisions that favour safety—even under pressure.

To support this policy, teams will be guided by standards for observability, deployment safety, SLO usage, and platform defaults. By prioritising safety before productivity, we create systems and teams that are built to thrive, not just survive.