Standards

This standard ensures that technical debt is explicitly identified, tracked, and prioritised in parallel with feature work to maintain system health and delivery agility. Making tech debt visible prevents it from being neglected, reducing long-term risks and costs.

It supports the policy “Balance Speed with Sustainability” by embedding technical debt management into regular planning and delivery cycles. Without this standard, hidden debt can accumulate, undermining system stability and slowing future innovation.

This standard ensures that teams use Service Level Objectives (SLOs) to balance delivery velocity with system reliability and user experience. SLOs provide a measurable way to define what ‘good enough’ means for availability, latency, and other performance criteria, enabling teams to manage change safely and make informed trade-offs between speed and stability.

By anchoring delivery decisions in SLO thresholds and error budgets, this standard supports the policy “Prioritise Safety Before Productivity” by embedding resilience into planning and fostering responsible innovation. Without this standard, teams risk over-optimising for speed while degrading service quality and user trust.

This standard ensures that engineers are empowered and supported to stop the line whenever there is a risk to service quality, reliability, or safety. Inspired by Lean manufacturing principles, it embeds a culture where raising concerns and halting flawed work is not just tolerated—it is expected and protected.

It advances the policy “Prioritise Safety Before Productivity” by creating psychological safety and organisational backing for those who intervene. Without this standard, risks go unchallenged, trust erodes, and teams fall into reactive cycles that normalise failure and burnout.

This standard ensures that any engineer, regardless of seniority, is empowered to halt production deployments or trigger a rollback in the face of service degradation, critical defects, or emerging incidents. By removing barriers to decisive action, this standard protects customer experience and enables faster incident containment.

It supports the policy “Run to Stop When Problems Arise” by reinforcing a culture of safety, trust, and shared responsibility. Allowing anyone to stop the line, without fear or delay, is foundational to operational resilience. Without this safeguard, teams risk compounding outages, delayed response times, and a culture of hesitation when decisive intervention is most needed.

This standard ensures that platform services and infrastructure components come with built-in, opinionated defaults for identity, security, observability, and compliance. By embedding safe defaults into the platform, engineering teams can move quickly without compromising on essential safeguards.

It directly supports the policy “Prioritise Safety Before Productivity” by preventing unsafe configurations from entering production and reducing the cognitive load on developers to secure every component manually. Without this standard, teams risk inconsistent practices, fragile configurations, and security or reliability vulnerabilities.

This standard ensures that incidents lead to meaningful improvements in services, platforms, or practices—not just retrospective documentation. The purpose of an incident review is not only to understand what happened, but to make sure it does not happen again.

It supports the policy “Run to Stop When Problems Arise” by shifting the focus from post-mortem reporting to actual learning and engineering action. When teams apply what they’ve learned, they reduce recurrence risk, strengthen systemic resilience, and build a culture of continuous improvement.

This standard ensures that all code, data, and infrastructure pipelines are designed and continuously improved to support seamless, automated flow from initial idea through to production deployment. Optimising pipelines end-to-end reduces handoffs, eliminates bottlenecks, and accelerates feedback loops.

It supports the policy “Design for Flow Efficiency” by enabling rapid, reliable delivery and ensuring that value streams remain unblocked by manual or disjointed processes. Without this, teams face delays, increased risk, and reduced ability to respond to change quickly.

This standard ensures that the size of change batches in development and deployment pipelines is actively monitored and controlled to maintain a sustainable flow and reduce risk. Managing batch sizes helps prevent integration issues, reduces the impact of failures, and supports a steady delivery cadence.

It supports the policy “Respect the Rhythm of Takt Time” by balancing throughput with quality and stability. Without this standard, large or unpredictable batch sizes can cause bottlenecks, increase defect rates, and disrupt the delivery rhythm.

This standard ensures that value stream mapping techniques are applied to visualise and analyse the end-to-end flow of work through CI/CD pipelines, data pipelines, and infrastructure provisioning. Identifying points of friction enables targeted improvements that reduce delays, handoffs, and waste.

It supports the policy “Design for Flow Efficiency” by providing teams and leaders with clear, actionable insights into flow impediments and bottlenecks. Without this, organisations risk inefficient delivery, missed feedback opportunities, and fragmented ownership of process inefficiencies.

This standard ensures that delivery cadence is intentionally synchronised with business demand, sprint cycles, and release schedules. Aligning engineering rhythm with external factors creates predictable delivery flow, optimises resource utilisation, and enhances stakeholder confidence.

It supports the policy “Respect the Rhythm of Takt Time” by establishing a deliberate tempo that balances customer needs with engineering capacity. Without this standard, mismatched cadences lead to bottlenecks, overloading, or idle resources, reducing overall flow efficiency.

This standard ensures engineers maintain a clear understanding of how their work affects end users and customers, fostering empathy and aligning technical efforts with real-world value. Awareness of user impact drives better decision-making and prioritisation.

It supports the policy “Pursue Value from the Customer’s Viewpoint” by embedding customer focus into engineering culture and practices. Without this standard, development risks becoming disconnected from actual user needs and business outcomes.

This standard ensures that system design favours simplicity and pragmatism by avoiding premature abstraction or over-generalisation. Delivering only necessary complexity reduces development overhead, accelerates delivery, and lowers maintenance burden.

It supports the policy “Eliminate Non-Value-Adding Activities” by preventing over-engineering and wasted effort. Without this standard, projects risk bloated designs, higher technical debt, and slower response to change.

This standard ensures that all code changes, feature updates, and infrastructure modifications have documented and understood links to specific business outcomes or user impacts. Clear traceability strengthens accountability and focuses efforts on delivering measurable value.

It supports the policy “Pursue Value from the Customer’s Viewpoint” by maintaining alignment between engineering work and organisational goals. Without this standard, work risks becoming disconnected from tangible results, reducing effectiveness.

This standard ensures that automation efforts are prioritised based on clear return on investment (ROI), focusing on automating tasks that save significant time or reduce risk. Automation is not pursued for its own sake but as a strategic enabler of faster, safer delivery.

It supports the policy “Deliver What’s Needed, Not Just What’s Possible” by aligning automation with business value and team capacity. Without this standard, teams risk over-investing in automation with minimal impact or introducing unnecessary complexity.

This standard ensures that codebases and infrastructure components are actively evaluated and maintained only when they continue to deliver tangible business value. Retiring or refactoring obsolete elements prevents waste, reduces technical debt, and optimises resource allocation.

It supports the policy “Deliver What’s Needed, Not Just What’s Possible” by promoting disciplined lifecycle management. Without this standard, organisations risk carrying unnecessary maintenance burdens that slow innovation and increase costs.

This standard ensures that feature toggles (flags) are used effectively to separate the deployment of code from the actual release of features to users. This decoupling provides greater flexibility in release timing, supports incremental rollouts, and reduces deployment risks.

It supports the policy “Enable Just-in-Time Production” by enabling teams to deploy frequently while controlling when features become active. Without this standard, deployments and releases are tightly coupled, limiting agility and increasing risk.

This standard ensures that development efforts focus on delivering features validated by clear and current user needs, avoiding speculative or premature work that risks wasting resources and increasing inventory.

It supports the policy “Minimise Inventory and Overproduction” by aligning delivery with validated demand, thereby reducing waste and maximising customer value. Without this standard, teams risk building unnecessary features that add complexity and technical debt.

This standard ensures that data pipelines are designed to minimise the number of intermediate hops and transformation steps between the original data source and the final insight consumers. Reducing data movement and processing layers accelerates delivery of insights and reduces latency and complexity.

It supports the policy “Shorten the Value Stream” by streamlining data flow, improving responsiveness, and lowering operational costs. Without this standard, excessive transformations introduce delays, increase errors, and reduce data freshness.

This standard ensures that teams prioritise delivering Minimum Viable Products (MVPs) and iterative releases that deliver real user value early and often, rather than investing heavily in large upfront designs. This approach reduces risk, accelerates feedback, and supports continuous learning.

It supports the policy “Minimise Inventory and Overproduction” by enabling faster validation of assumptions and avoiding wasted effort on unnecessary features. Without this standard, teams risk delayed feedback, higher costs, and reduced agility.

This standard ensures that organisations actively measure, monitor, and optimise the total lead time from initial code commit through to customer value realisation. Transparent visibility of end-to-end delivery time enables targeted improvements and sustainable flow acceleration.

It supports the policy “Shorten the Value Stream” by making delivery efficiency a key performance focus. Without this standard, delays and bottlenecks remain hidden, reducing responsiveness and increasing risk.