Metro Bank · UK Retail Banking · AI · 2018
Partnered with Personetics to design and ship the first AI-powered feature in UK retail banking, predictive transactional insights that analysed customer spending patterns and nudged better financial behaviour, shifting the average customer balance from negative to positive.
The Approach
We began with a discovery phase, analysing 12 months of anonymised transaction data to identify the highest-value insight categories. I introduced a tiered insight model: passive insights, actionable insights, and predictive alerts, each with its own visual treatment and interaction pattern.
AI partnership and integration architecture, Led the design integration with Personetics' behavioural analytics AI platform. Curated which of 30+ insight types mattered most and defined how each would manifest in the customer experience.
Tiered insight model, Passive insights (spend observations), actionable insights (cancel subscriptions, set budgets), and predictive alerts (upcoming bills, projected overdraft warnings). Each tier with distinct visual treatment and urgency signals.
Fraud prevention layer, Designed duplicate transaction detection flow where the AI flagged potentially duplicated charges and surfaced confirmation prompts, giving customers real-time fraud prevention without friction.
200+ participant validation programme, Led three rounds of usability testing. Discovered customers responded better to observations ("You spent 23% more on dining") than directives ("Reduce your dining spend"). Ran a live pilot with 5,000 customers before full rollout.
Executive stakeholder management, Presented to the CEO and board, translating complex AI capabilities into business outcomes. Managed quarterly reviews and maintained alignment across product, engineering, compliance, and marketing.
"Implementing the UK's first AI feature in retail banking wasn't just a design problem, it was a trust problem. The experience had to make novel technology feel like a knowledgeable friend, not a credit score warning."
The Outcomes
The AI-powered insights became the most-used feature in Metro Bank's mobile app within 6 months of launch, fundamentally shifting the bank's digital strategy from transactional utility to intelligent financial guidance.
The feature became the app's most-engaged within 6 months, driven by the tiered insight model and personalisation engine.
Retention far exceeded typical benchmarks for new features. Customers who engaged continued returning month over month.
Average customer balance shifted from negative to positive for the first time, a direct result of AI-nudged financial behaviour changes.
73% of duplicate transaction disputes resolved automatically via AI-flagged confirmations, without requiring human intervention.
Want in?
Score 100% on the Design Eye test to earn access. Or enter the password if you have it.
4 rounds. No partial credit.