From replacing decades-old banking systems to rolling out Microsoft 365 Copilot, Nationwide is pairing skilled people with modern technology, deepening its collaboration with Microsoft to deliver better experiences and outcomes for its members.
Most people never think about the technology behind their bank. They expect payments to arrive, branches to serve them, apps to open, and mortgages to complete.
That expectation makes banking technology unusually hard to change, and it’s why Nationwide Building Society treats modernisation as a partnership between skilled people and the right technology.
The work has meant replacing core systems, rebuilding customer data, moving services to the cloud, and bringing Virgin Money into the group, all while keeping the experience steady for the members who rely on it.
For Paul Ballard, Nationwide’s Chief Technology Officer, the aim is simple to state and difficult to deliver: make the organisation more modern while customers carry on with their lives.
“We have to do it where people don’t even know it’s happened,” he says. “We don’t disrupt your life.”
The scale is substantial. Nationwide, the world’s largest building society, serves around 19 million members and has 605 branches. Ballard says it has already moved 15 million customers off its former core banking engine, replaced a 45-year-old customer data system, and is preparing to integrate around five million Virgin Money customers.
That work covers customer records, products, sort codes, card providers, branches and service journeys.
“You’re doing it through the eyes of the customer,” Ballard says. “How do we move you to Nationwide platforms in a way that you don’t notice and you get better features when it happens?”
Firm foundations
That same logic runs through Nationwide’s data strategy. When Sri Kanisapakkam, Chief Data and Analytics Officer at Nationwide, joined in 2023, the organisation had 19 separate data stores serving different teams and purposes.
Since then, Nationwide has been consolidating those systems onto Microsoft Azure and Azure Databricks. Kanisapakkam says that work is now complete, creating one accessible data platform.
“We wanted one unified platform that serves all of the data to the society in one place, on one technology, on the cloud,” he says.
For a broad audience, the important distinction is between structured and unstructured data. Structured data fits neatly into records and tables, such as customer, account, product and transaction information. Unstructured data includes documents, knowledge articles, policies and notes. Bringing those worlds closer together matters because useful AI depends on context.
“AI is not magic,” Kanisapakkam says. “It needs to understand how your data is organised and how your data is structured. The data has to be trusted first before you can say your AI is trusted.”
That trusted base is already helping Nationwide understand about 120 customer journeys, support Virgin Money customer matching and offer senior leaders new ways to ask questions of data.
A capability called Talk to Data has been rolled out to around 70 to 80 executives and direct reports.
The human factor
A similar shift is happening in finance, where Microsoft 365 Copilot is becoming part of everyday work.
Katy Barker-Cook, Director of Future Capability and Operations in Finance at Nationwide, says finance became one of the first group functions to roll out Copilot across the combined Nationwide and Virgin Money team. Around 1,000 people in finance now have premium licences.
The work it supports is practical: turning rough notes into first drafts, summarising meetings, shaping messages, pulling actions from discussions and helping teams work across two organisations.
“A lot of the work in finance isn’t technically complex, it’s just very fragmented or very time intensive,” Barker-Cook says. “It’s not the thinking that takes the time. It’s all the stitching together of that work.”
In a recent internal poll, 46% of finance respondents said they used Copilot every working day and 36% several times a week. Barker-Cook is careful about overclaiming the impact while adoption is still developing, but says colleagues are already seeing faster drafting, quicker turnaround on routine outputs, and more consistent written materials.
“It doesn’t replace judgement, especially in finance where accuracy really matters,” she says. “But it shortens the path to a good first draft.”
The approach also reflects Nationwide’s long-standing message on responsible AI. In a previous public story, the building society said using GPT-4 within Azure OpenAI to help generate customer letters had reduced average response times from 45 minutes to 10-15 minutes. It also stressed that a human reviews generative AI output before it is used.
“It’s not about AI doing finance,” Barker-Cook says. “It’s about AI helping finance do its job better.”
Agentic future
Nationwide is now extending the same thinking from assistants to agents. Ballard points to Microsoft 365 Copilot for everyday productivity, around 2,500 engineers using GitHub Copilot, Azure OpenAI for generative AI use cases, and Microsoft technologies such as Azure AI Foundry as part of future agent development.
Agentic examples include knowledge assistants for frontline colleagues, agents that help business analysts create requirements, tools for architects, and future agents for anti-money laundering, customer due diligence, and mortgage advice.
“We’re moving to the phase where we’re working with Microsoft technology so we can start to create agents that can talk to our systems,” Ballard says.
Nationwide already has a very strong reputation for customer service. It was recently judged to be the UK’s best-performing brand for customer satisfaction, according to the latest UK Customer Satisfaction Index.
For Ballard, the next few years are about integration and innovation.
“We need to make sure we come out of integration modern, innovative and future fit,” he says.
Nationwide is strategically using cloud, data and AI to keep improving the services customers rely on, while preparing carefully for the agentic era ahead.