Shipped

Payday Automation Redesign

Role
Product Designer, sole designer on this project
Team
1 PM, 1 lead engineer, 1 data analyst, 2 frontend, 2 backend (AI squad)
Timeframe
2 sprints, shipped, results measured over following weeks and months

The problem

35% conversion: worst of all automation rules, despite being the highest-value one

The Payday Rule was the best-performing savings rule in the product. Users saved about £80/month on average once set up, but it was badly underused:

"[Placeholder customer comment. Replace with a real quote about the old setup being confusing or too many steps.]"

[Source: customer support ticket, app review, or research session]

Goal

Target: 45 to 50% of automation-active users adopting Payday. Working bet: the problem was access and friction, not demand.

Process

Decision 1

Strip setup to 2 inputs (amount and frequency), add smart defaults and a confirmation screen

Why: the multi-input form was the highest-friction setup flow anywhere in the product.
Old multi-field setup form
New 2-input flow, side by side
Decision 2

Add Payday as a 3rd onboarding option, after Automatic and Weekly

Why: most automation adoption happens at onboarding, and Payday was invisible there entirely.
Onboarding step before/after

Final UI

This is the strongest visual asset in the case, lead with it.

Old multi-field setup
New simplified flow
Onboarding placement screen

Results

Weekly adoption rose from 55% to 64% (+16% relative), beating the 45 to 50% target.
Monthly Payday users grew from 250K to 265K in 2 months.
AUM increased from £19M to £22M, 98% attributable to this change.
Conversion improved from 35% to 39% (+9% relative).
Halo effect: all-rules automation adoption up 22%.
Honest miss: 39% conversion is still bottom of pack. Friction was reduced, not solved.
Honest limitation: no events tracking or GTM at launch. Onboarding placement and the simplified flow shipped together, so their individual effects can't be cleanly separated.

Reflection

The onboarding amount nudge didn't land as designed: 46% picked £0-50 vs. 31% in-app. The preset and framing logic needs rework.

Next: isolate the onboarding effect from the simplification effect with better instrumentation.