Centrica is rolling out a comprehensive governance model for Microsoft Power Platform across its UK energy operations, using Managed Environments, a tailored Center of Excellence, and Copilot Studio to turn low-code development into a secure, enterprise-grade operating model. The initiative, detailed in a recent company case study, marks one of the most ambitious implementations of Power Platform governance in the utilities sector and provides a blueprint for organisations balancing agility with control.

The Low-Code Surge in the Energy Sector

Energy companies face mounting pressure to digitise operations, from field service management to customer analytics. Low-code platforms like Power Platform promise rapid innovation, but without governance, they can lead to security risks, data leakage, and unmanageable sprawl. Centrica, which owns British Gas and serves millions of UK households, is tackling this head-on by formalising its approach to citizen development.

Rather than restrict access, Centrica is embracing low-code as a core operating model. The key is a governance framework that empowers business users while ensuring IT oversight. This includes standardising on Managed Environments, a feature set within Power Platform that provides enhanced administrative controls, data policies, and pipeline management.

Managed Environments: The Governance Foundation

At the heart of Centrica’s strategy lies Managed Environments, a suite of capabilities that allows administrators to manage Power Platform assets at scale. These environments enable:

  • Environment routing: Automatically directing new makers to their own personal developer environments, preventing accidental changes to production systems.
  • Data policies: Enforcing strict data loss prevention (DLP) rules that restrict connectors and ensure compliance with GDPR and industry regulations.
  • Solution checker: Validating apps and flows for performance and security issues before deployment.
  • Weekly digest emails: Providing administrators with a summary of new resources, inactive apps, and potential risks.

For Centrica, which operates across multiple regions and business units, Managed Environments provide a unified control plane that was previously missing. The company can now enforce policies consistently while giving local teams the flexibility to build solutions tailored to their needs.

Building a Center of Excellence That Works

Technology alone cannot govern a low-code platform. Centrica established a Center of Excellence (CoE) to drive adoption, provide training, and manage the intake process. Unlike traditional IT governance boards that can bottleneck innovation, Centrica’s CoE uses automated workflows and clear criteria to fast-track app approvals.

The intake process leverages Power Automate and custom model-driven apps to submit, review, and approve new projects. Makers submit their ideas through a structured form, which triggers a review based on pre-defined thresholds—such as number of users, data sensitivity, and integration complexity. This tiered approach means that low-risk apps can be approved within hours, while higher-risk projects receive deeper scrutiny from architects and security teams.

The CoE also maintains a catalogue of reusable components, including AI Builder models, custom connectors, and UI templates, all stored in a centralised component library. This not only accelerates development but also ensures consistency and compliance.

Copilot Studio: The AI-Powered Advisor

Perhaps the most innovative element is Centrica’s use of Microsoft Copilot Studio (formerly Power Virtual Agents) to embed governance intelligence directly into the maker experience. The company is developing a conversational AI that acts as a real-time advisor for citizen developers.

Using Copilot Studio, Centrica’s CoE is building a bot that integrates with Microsoft Teams, where most employees already work. Makers can ask the bot questions such as:

  • “What DLP policy applies to my app if it uses customer data?”
  • “How do I request a premium connector?”
  • “Is there an existing component for invoice processing?”

The bot draws on a knowledge base built from the CoE’s documentation, policy manuals, and even past project reviews. By surfacing answers contextually, the bot reduces the burden on the CoE team and prevents mistakes early in the development cycle. Over time, the plan is to train the bot to proactively flag potential issues—for example, if a maker selects a connector that is restricted for their environment, the bot would alert them before they start building.

This melding of AI and governance is a significant departure from traditional top-down enforcement. It turns governance into a helpful, rather than punitive, experience and can dramatically improve adoption rates.

Real-World Impact at Centrica

While the full implementation is ongoing, early results point to positive outcomes. By moving from an ad hoc approach to a structured operating model, Centrica has already reduced the risk of shadow IT while accelerating the delivery of business solutions. Field engineers now have mobile apps that streamline maintenance scheduling, customer service teams use automated flows to handle common queries, and finance teams leverage AI Builder to process invoices—all built within a governed framework.

The adoption of Managed Environments has given Centrica’s central IT team visibility into thousands of previously unknown assets. This has allowed them to identify unused apps, monitor performance, and ensure that all flows comply with corporate security policies. The weekly digest emails alone have become an essential tool for capacity planning and compliance auditing.

Moreover, the CoE intake system has empowered business units that were previously frustrated by long IT queues. With the tiered approval process, simple automation ideas no longer wait weeks for a review. Instead, they are rapidly prototyped in personal environments and moved to production once basic checks are passed. This agility is critical in an industry where customer expectations are rising and operational efficiency can directly impact energy costs.

Overcoming Cultural and Technical Challenges

Shifting to a low-code operating model is not without hurdles. Centrica had to address cultural resistance from both IT professionals, who feared a loss of control, and business users, who were sceptical of yet another platform. The CoE played a pivotal role in change management, running hackathons, lunch-and-learn sessions, and creating a community of practice where makers can share successes.

On the technical side, integrating Power Platform with Centrica’s existing SAP and legacy systems required careful planning. The CoE worked closely with enterprise architects to define integration patterns using Azure API Management as a gateway, ensuring that low-code apps connect safely to back-end data without exposing sensitive systems.

Data residency and compliance with UK energy regulations added another layer of complexity. Managed Environments allowed Centrica to place data in specific geographic regions and enforce policies that prevent data exfiltration. The Copilot Studio bot is also being designed to comply with privacy regulations, with all queries and responses logged for audit purposes.

Industry-Wide Implications

Centrica’s approach reflects a broader maturation of low-code adoption in large enterprises. According to Microsoft, Managed Environments have seen rapid uptake since their introduction, as companies move beyond “Shadow IT” fears to a “Shadow Engineering” model that harnesses business-led innovation safely. The integration of AI copilots into governance is particularly forward-looking.

Other energy and utility companies are likely to take notice. As similar pressures mount—from the need to achieve net-zero targets, manage smart grids, and engage digitally native customers—the ability to deploy fit-for-purpose apps quickly becomes a competitive differentiator. Centrica’s model demonstrates that with the right governance, even heavily regulated industries can embrace low-code without compromising security.

What’s Next for Centrica?

Looking ahead, Centrica plans to expand the CoE’s scope to include Power Pages for external portals and Power BI for advanced analytics. The Copilot Studio bot will evolve to integrate with Azure OpenAI Service, enabling more sophisticated natural language understanding and the ability to generate code snippets or suggest entire workflows based on plain-English descriptions.

The company is also exploring the use of Managed Environments’ pipeline capabilities to automate application lifecycle management (ALM). This would allow citizen developers to simply request a pipeline for moving their solutions from development to test to production, with all environment variables and connection references automatically updated—reducing manual errors and speeding up deployment.

As Centrica continues its digital transformation journey, its Power Platform governance model stands as a testament to how large organisations can turn the low-code movement from a potential risk into a strategic advantage. By combining technology, process, and AI-driven guidance, the energy giant is building a foundation for sustained innovation.

For enterprises still on the fence about low-code governance, Centrica’s story offers a clear message: start with a CoE, leverage Managed Environments for control, and don’t underestimate the power of an AI sidekick. The result is a democratised development culture where everyone wins—IT gets peace of mind, and business gets speed.