Our client was developing an open balancing programme to transform the UK’s power grid management — to build a future-ready system that could meet the UK’s net-zero greenhouse gas emissions targets.

But a heap of technical and organisational issues meant that project was going off-track. And in danger of missing critical OFGEM deadlines.

ClearRoute was brought in to help re-energise the project.

‍

Before:

A balancing system… unbalanced.

‍

Our client manages the grid using a balancing system, which ensures electricity production is evenly matched to consumption, the frequency is constant and there’s sufficient reserve capacity.

There’s no room for error: if the grid isn’t perfectly balanced on a moment-by-moment basis, electricity provision is disrupted.

But the country’s power system has become more volatile over the past decade, with reliance on a greater range of energy sources. The system — built in 1990 — wasn’t equipped to meet the demands of modern energy generation.

Our client needed a new balancing platform that could:

• Manage the large market increase
• Adapt more quickly to new requirements, innovations, and services
• Create a level playing field for all providers
• Reduce costs to consumers through optimising balancing costs
• Handle the clean energy to meet the UK’s net-zero emissions targets

That’s where the open balancing platform came in. But the project was off-track, and the team risked meeting their delivery deadline.

The transformation project was stalling, thanks to a host of technical and organisational challenges:

• Distributed monolith issues: Although the platform was designed as microservice architecture, components were being deployed with breaking changes, creating complex dependencies. This effectively created a distributed monolith with cascading failures that were derailing progress.

• Testing anti-patterns: The project relied on an inverted test pyramid that relied heavily on excessive end-to-end tests while neglecting unit and integration tests. This error-prone feedback loop couldn’t catch errors early — slowing progress and making troubleshooting complex and slow.

• Pipeline inefficiencies: Long-running and unreliable continuous delivery pipelines created bottlenecks and slowed deployment.

• Environment inconsistency: Manual verification was required before promoting to test environments, introducing inconsistency, human error and delays – destabilising the release process.

• Data dependencies: Time-sensitive test data forced sequential test execution rather than allowing parallelism, dramatically reducing testing efficiency and slowing the pipeline.

• Siloed teams: Poor collaboration between engineering squads and centralised test teams meant critical issues weren’t communicated or resolved efficiently, fragmenting and slowing progress.

• Unclear quality ownership: Without a standardised framework for assessing whether deployed software met system requirements, there was no clear accountability for quality and performance.

• Inconsistent standards: The absence of enforced coding standards, security scans, and code coverage checks meant that quality was uneven, causing rework and additional delays.

‍

Together, these issues were creating a perfect storm of compounded technical complexity and organisational misalignment — creating systematic friction that threatened reliability and stalled delivery.

That’s where ClearRoute came in

‍

How we helped:

Re-energising the platform in two phases  

Our client had the right foundations for a modern, future-ready platform — but fundamental issues across the delivery pipeline, quality processes, and team structures were stalling progress.

We delivered transformation in two phases.

Phase 1: Strategic re-architecture

ClearRoute embedded two multi-skilled QCE pods into the engineering team, to address both the underlying technical hurdles and the cultural barriers stalling progress.

Comprehensive test strategy redesign

• Designed sophisticated test data factory to unlock rapid, realistic, complex, production-like data that could be provisioned on-demand. Better data earlier, for fewer errors later.  

• Refined microservice architecture to support asynchronous and parallel test automation, reducing interdependencies and empowering teams to release small, independent changes confidently.

• Transformed testing paradigm into a modern testing trophy approach that’s better suited to microservice architectures, providing better coverage, faster feedback, more reliable testing, and fewer bottlenecks.

‍

Engineering integration

• Implemented DevSecOps practices like automated code checks, security scans, and dependency validations baked into the CI/CD pipeline, ensuring changes met rigorous standards from the start.

• Established clear service contracts between microservices, to stop breaking changes and keep everyone on the same page.

• Established rigorous pull request review processes focussing on testability, maintainability, and adherence to architectural standards. More alignment and accountability; less failures and finger-pointing.

These interventions transformed a sluggish, risk-laden process into an agile, high-confidence pipeline, paving the way for the platform’s successful and timely launch.

Phase 2: Continuous testing platform

Following the initial success of our interventions, ClearRoute then designed and built a comprehensive continuous testing platform to embed this new, faster, proactive model of testing as the new status quo.

• Health evaluation: Real-time monitoring during microservice deployments meant teams could spot errors as they emerged, preventing small glitches from spiralling into system-wide failures.

• Performance testing: Testing under heavier-than-production loads exposed bottlenecks early. This pre-emptive approach allowed our client to scale smoothly, avoiding performance surprises once the system went live.

• Chaos engineering: Injecting controlled failures validated system resilience under unexpected conditions. It’s not just about surviving chaos—it's about learning lessons to bolster the platform against real-world future disruptions.

• Zero-downtime operations: Support for routine patching without service disruption reduced costly downtime, underscored reliability, and protected user confidence.

• Production readiness verification: Automated checks ensured the system complied with 99.9% availability SLO, ensuring new features, changes, or fixes never jeopardised uptime.

‍

After:

A future-ready balancing platform

‍

ClearRoute’s intervention helped deliver a dynamic, agile platform with quality at its core – and facilitated major cultural and organisational changes that would stand the energy provider in good stead for the future. The platform can now deliver improvements faster, more safely, and with far greater stability—cementing the success of the overall transformation.

• Accelerated deployment: Deployment frequency increased 3x

• Successful delivery timeline: The platform went live one week ahead of schedule

• Improved quality: Unit test code coverage soared to 75%, all dependency scans became 100% compliant, and integration tests were fully automated

• Faster testing: Test runtimes dropped 96% from 8 hours to 5 minutes

• Operational stability: Achieved 99.9% availability at launch

• Quality ownership: Shifted from siloed QA teams to making quality everyone's responsibility

• Increased team efficiency: Restructured teams to align with service boundaries, reducing handoffs and context switching
  ‍‍
• Empowered engineering teams: Enabled development squads to fully own services from design through production support: better for everyone

‍