The Challenge
What Ozzie Was Facing
Ozzie's AI generated personalised savings nudges, but engagement with recommendations dropped sharply after the first week of use. Survey data was inconclusive; only diary studies and experience sampling revealed that nudges were arriving at moments of high cognitive load — commutes, meal times — when users had no capacity to act on financial decisions.
The Solution
What We Built
We designed a contextual timing research protocol using ESM (experience sampling method) with 60 participants over six weeks, mapping receptivity windows and emotional states to notification timing. The findings informed a redesigned nudge engine and a calmer IA that surfaced savings progress as ambient feedback rather than interruptive prompts.
Results
Measurable Outcomes
✓Weekly active engagement with savings recommendations increased by 54%
✓Average monthly savings per active user grew by £180 within three months
✓Push notification opt-out rate fell from 41% to 9%