Starling Bank Powers AI Banking Revolution with Google Cloud

Starling Bank is leveraging its longstanding partnership with Google Cloud to revolutionise how customers interact with their finances, with the challenger bank's latest AI-powered tools demonstrating the transformative potential of enterprise cloud technology in financial services.
In an interview at the Google Cloud Summit London, Harriet Rees, Chief Information Officer at Starling Bank, outlined how the eight-year partnership has enabled the bank to scale from startup to serving over 4.5 million customers whilst maintaining its mission to change people's relationship with money.
AI-Powered Customer Experience
The partnership has culminated in Spending Intelligence, a groundbreaking AI tool that allows customers to interrogate their financial habits using natural language.
Powered by Google's Gemini models, the feature enables users to ask simple questions about their spending patterns and receive immediate, actionable insights.
"We wanted to develop a tool to help customers to be able to ask really simple questions about their finances and get really simple answers," Rees explains.
"Spending Intelligence, powered by Gemini, allows customers to use natural language – text or voice – to ask really simple questions like: 'How much did I spend on transport last month?' or 'What did I spend on holidays last year?'"
The tool addresses a critical challenge in UK financial services, where low financial literacy leaves many customers struggling to understand their spending habits.
Rees acknowledges that while the insights provided might sometimes be uncomfortable – she noted "jaw-dropping looks when people ask about takeaways last month" – the tool is designed to foster healthy financial habits.
Seven-Year AI Journey
Starling's AI implementation represents a maturation of its technological approach. The bank has been utilising machine learning across its operations for seven years, but 2024 marked a strategic shift towards customer-facing AI applications.
"Our AI journey is not a new one," Rees notes. "We've been using machine learning as we've scaled across our operations for about seven years, but this year for us, we had one aim, and that was to bring AI into the hands of our customers."
This customer-centric approach distinguishes Starling from competitors who focus primarily on operational efficiency.
Rees emphasises the bank's commitment to demonstrating AI's transformational potential for consumers, drawing parallels with how AI has revolutionised search experiences.
Comprehensive Google Cloud Integration
The partnership extends far beyond AI capabilities. Starling has built its entire technological infrastructure around Google Cloud services, creating an integrated ecosystem that supports both current operations and future expansion.
The bank utilises Google Workspace for daily operations, with Gemini's integration providing staff with organic AI experiences within their existing workflows.
Its data warehouse operates on BigQuery, whilst engineering and developer pipelines are built around Vertex AI, enabling rapid and secure model development and deployment.
"From the developer experience perspective, it's been received by our teams as a great tool set to allow our staff to build models and push them to production with speed and safely," Rees explains.
"We've been using machine learning as we've scaled across our operations for about seven years, but this year for us, we had one aim, and that was to bring AI into the hands of our customers"
The robust technological foundation has enabled Starling to expand beyond the UK market through Engine, its software-as-a-service subsidiary.
This platform delivers Starling's proprietary tech stack and core banking platform to international clients, with operations currently live in Romania and Australia.
The expansion demonstrates how cloud-native architecture can support rapid scaling whilst maintaining the innovative features that differentiated Starling in the UK market.
Many mainstream banking features, including card controls and gambling blocks, were pioneered by Starling before becoming industry standards.
Security and Data Sovereignty
For Rees, Google Cloud's security assurances proved decisive in the partnership selection. As CIO, she required comprehensive guarantees about data handling and storage to ensure both regulatory compliance and customer trust.
"With my CIO hat on, of course, it was most important to me that I was able to provide both myself and our customers with the assurances around where that data would be held, how that data would be used, and to be comfortable that we were able to articulate that clearly to our customers," she says.
Google Cloud's ability to provide these assurances ahead of other providers made it "a natural choice" for Starling's continued development.
The success of Spending Intelligence has paved the way for additional AI-powered customer tools. Starling is developing a similar application focused on scam detection and prevention, with plans to integrate AI capabilities across its entire application suite.
"We'll be taking that further and looking to see what else we can do with generative AI in other spaces," Rees reveals.
"Starling customers will be able to look out in the next few weeks for a similar tool, which will help customers to perhaps avoid and detect scams."
This expansion reflects Starling's broader ambition to utilise AI for creating "intuitive, helpful tools" that support its core mission of helping customers "be good with money."

