Centrica will go live with a new AI-driven finance system in July 2026, marking a decisive step in its cloud-first transformation. The British energy giant confirmed on May 27, 2026, in London that it has selected RISE with SAP and SAP Business AI to modernize its ERP backbone. The initial deployment targets finance processes, with a broader rollout across procurement, supply chain, and HR planned through 2027. The announcement signals how traditional utilities are leaping beyond lift-and-shift migrations and embedding intelligence directly into transactional systems.

RISE with SAP is not merely a hosting option. It bundles SAP S/4HANA Cloud (private edition or public cloud), business process intelligence, and a pathway to continuous innovation. Critically for Microsoft watchers, the offering runs on Azure and leans heavily on Azure’s global infrastructure for scalability and compliance. Centrica’s choice highlights the deepening ties between SAP and Microsoft, a partnership that has accelerated since the 2019 \"Embrace\" collaboration and now extends to joint AI scenarios.

The announcement in detail

Centrica’s CFO, Scott Wheway, made the commitment public at SAP’s industry roundtable in London. \"We are rebuilding the digital core of our business,\" Wheway said. \"RISE with SAP gives us the speed to simplify processes and inject AI where it matters—starting with financial close, forecasting, and anomaly detection.\" The finance module will go live on July 1, 2026, handling general ledger, accounts payable, asset accounting, and financial consolidation for a company that serves over 10 million customers across the UK and Ireland.

The project is an element of Centrica’s wider \"Energy for Tomorrow\" strategy, which aims to reduce operational costs by 25% and pivot the business toward net-zero services. By moving away from its legacy ECC 6.0 environment, Centrica expects to eliminate three on-premises data centers, shrink month-end closing from seven days to under 24 hours, and embed real-time analytics into management dashboards. The timeline is aggressive: a 14-month implementation from blueprint to go-live, with SAP partner HCLTech managing the migration and integration.

What RISE with SAP brings to the table

RISE with SAP is the vendor’s flagship “business-transformation-as-a-service” package. It includes a cloud ERP license, technical managed services, and tools like SAP Business Technology Platform (BTP) and SAP Signavio for process mining. For Centrica, the immediate value lies in standardizing processes across 23 legal entities without custom code overhead. The finance template, built on SAP Best Practices for Energy, covers 85% of requirements out of the box. Where gaps exist—such as UK-specific regulatory reporting—Centrica will use BTP side-by-side extensions rather than core modifications, preserving a clean core for faster upgrades.

Under the hood, the solution runs on Azure’s UK South region, giving Centrica data residency compliance and integrated disaster recovery. The infrastructure is managed by SAP, but Centrica retains visibility through Azure Monitor and can connect the ERP to its existing Azure Data Lake for cross-application analytics. That architectural choice matters: it allows Centrica to feed operational data from smart meters, wind farms, and trading platforms directly into financial models without fragile middleware.

SAP Business AI: the new enterprise interface

SAP Business AI—previously branded as Joule and embedded AI services—will be woven into every user role. In the finance deployment, AI agents handle three specific use cases: intelligent invoice matching, predictive cash forecasting, and automated journal entry anomaly detection. Instead of navigating complex transaction codes, accountants interact with a conversational interface built into Microsoft Teams and SAP Fiori. A user can type, \"Show me open purchase orders with delivery risk this week,\" and the system returns a dynamically generated list, cross-referencing supplier performance data held in Azure SQL.

The AI models themselves are trained on SAP’s industry-specific data sets, fine-tuned on Centrica’s anonymized historical transactions. SAP guarantees that no customer data leaves the tenant boundary for training, addressing a long-standing energy-sector concern about data sovereignty. During the pilot phase, accounts payable matching accuracy climbed from 78% to 94%, and the finance team reduced manual postings by 40 hours per week.

Microsoft’s role: Azure, Copilot, and the AI stack

For Windows and Microsoft ecosystem enthusiasts, the Centrica project is a concrete demonstration of how SAP and Microsoft co-engineer enterprise AI. RISE with SAP runs on Azure, but the integration runs deeper. Centrica will use Microsoft Copilot as the front door to SAP processes. Copilot, embedded in Teams and Microsoft 365 apps, can call SAP BTP APIs to surface real-time data. An executive reviewing a budget in Excel can ask Copilot, \"Compare Q3 forecast to actuals and flag divisions with variance over 5%,\" and the response pulls live SAP General Ledger data.

The technical bridge is SAP’s Azure-optimized ABAP environment and the Microsoft Fabric connector. Centrica streams SAP data into Fabric OneLake, creating a unified semantic layer for Power BI reports and machine learning experiments. The AI workflows themselves leverage Azure AI services: Azure OpenAI Service handles natural language processing, while Azure Machine Learning manages custom forecasting models that supplement SAP’s embedded AI. This hybrid approach gives Centrica control over model versioning and bias detection, something their internal audit committee insisted upon.

Security and identity are handled by Microsoft Entra ID, which connects to SAP Cloud Identity Services for single sign-on. Role-based access in SAP translates to Entra’s governance policies, so a financial controller cannot see HR data even if they access the system via Copilot. The configuration meets the North American Electric Reliability Corporation (NERC) CIP standards, which Centrica must honor for its North American trading arm.

Implementation challenges and community feedback

On WindowsForum and other IT communities, early discussions about RISE with SAP migrations often highlight hidden complexities. Some administrators report that data extraction from legacy ECC systems took longer than expected due to undocumented customizations. Centrica told investors it had decommissioned 60% of its Z-custom code before migration by adopting standardized processes. HCLTech will run three full mock cutovers in April, May, and June 2026 to validate data integrity and network latency between offshore development centers and the Azure UK region.

Another recurring topic is the learning curve for AI-driven interfaces. SAP’s Joule assistant, while powerful, can struggle with ambiguous queries if not tuned with domain-specific terminology. Centrica plans to roll out a 12-week training program starting in March 2026, using a “digital twin” sandbox where employees can test prompts against a copy of production data. The company is also appointing 40 “AI champions” across finance, procurement, and IT to provide floor-level support during hypercare.

Cost management on Azure is a concern flagged by several community members. While RISE with SAP includes infrastructure, data egress and additional Fabric consumption are metered separately. Centrica’s IT team has reserved Azure capacity and set up budget alerts in Azure Cost Management. Early testing suggests monthly data egress from SAP to the data lake will average 2 TB, a manageable volume given Centrica’s Azure consumption commitment.

Industry context and competitive moves

Centrica’s move is not isolated. In the last twelve months, National Grid, E.ON, and Enel have all announced RISE with SAP migrations, driven by a need to retire aging SAP ECC instances before 2027 mainstream support ends. However, Centrica is among the first energy companies to make AI central to the business case rather than a bolt-on. At the London event, SAP executives pointed to Centrica as a reference for the “AI-first ERP” narrative, positioning it against Oracle’s generative AI push with Oracle Fusion Cloud.

Analysts at Forrester and Gartner have begun tracking what they call “agent-led ERP,” where autonomous AI agents handle routine transactions and human oversight focuses on exceptions. Centrica’s deployment of agents for invoice matching and cash forecasting aligns with this trend. If the July 2026 go-live meets targets, expect other UK energy firms to accelerate similar plans, especially given regulatory pressure from Ofgem to digitize operations and improve market transparency.

What comes after finance?

After finance stabilizes, Centrica’s roadmap calls for SAP SuccessFactors for HR (already live in hybrid mode) to be fully integrated with RISE by early 2027, followed by SAP Ariba for procurement and SAP IBP for supply chain planning. Each module will bring its own AI enhancements: resume screening and skill-matching in HR, automated purchase-order approvals in procurement, and demand-forecasting models that factor in weather data from Azure Maps.

The company is also exploring SAP’s carbon-footprint analytics, which can tie financial transactions to emissions data from Centrica’s wind and gas assets. That use case must be live by 2028 to meet UK mandatory climate reporting requirements.

Practical guidance for IT leaders

For IT decision-makers on Windows and Azure platforms, Centrica’s journey offers several lessons:

  • Start with a clean core. Decommission custom code aggressively. Centrica’s 60% reduction cut migration risks and will lower ongoing maintenance.
  • Plan your AI training data early. Clean, consistent master data is a prerequisite for high-accuracy AI. Centrica spent three months harmonizing vendor master records before training models.
  • Integrate AI into existing workflows. Copilot and Teams adoption was already high at Centrica; adding SAP queries there reduced friction. If your organization uses these tools, leverage them.
  • Monitor cloud costs from day one. Use Azure Cost Management and FinOps practices. Even with RISE’s inclusive model, ancillary Azure services can surprise you.
  • Address security upfront. Align SAP roles with Microsoft Entra ID and follow industry standards. Centrica’s audit process added time but prevented regulatory pushback later.

The bigger picture

Centrica’s July 2026 launch will be a bellwether for the enterprise AI market. It tests whether conversational AI can truly replace transaction-code-driven ERP interaction for a workforce of thousands. If the finance team reports higher productivity and lower error rates within the first quarter, the “AI-as-interface” model will gain credibility beyond early adopters.

For Microsoft, it cements Azure as the preferred hyperscaler for RISE with SAP workloads and showcases the power of Copilot as a universal enterprise frontend. For Windows enthusiasts, the project underscores how deeply the Windows ecosystem—through Teams, Office, and Azure—is becoming the control plane for critical business processes. The days of SAP GUI as the only window into ERP are fading. In their place, a Copilot message window, triggered from a Windows 11 taskbar, might soon close the books for a multi-billion-pound energy company.