E-commerceOmnichannelUX StrategyRedesign

UCOOK

Sign-up funnel redesign.

Role

UX/UI Designer

Company

Silvertree (UCOOK)

Year

2025

Methods

Mixed methods

3.7%

old completion rate

9.3%

new completion rate

~R1.3M

annual revenue at risk

Customer service data showed users completing what they believed was sign-up, then contacting support expecting deliveries.

60+

Incomplete sign-ups per week

1.43%

Retention through the sign-up path

This was not on the roadmap. I initiated the investigation.

The Problem

User pain

  • Users confused sign-up with account creation.
  • Could not find meals or navigate back.
  • Subscription vs on-demand was unclear.

Business pain

  • 60+ incomplete sign-ups per week.
  • ~R1.3M/year revenue leaking through funnel.
  • 56% of first billings 1–5 days post sign-up.

Goals

01

Validate whether sign-up was a genuine usability and conversion issue worth prioritising

02

Identify where and why users were dropping out of the journey

03

Define a clear, measurable happy path sign-up funnel

04

Reduce incomplete sign-ups and improve confidence around subscription activation and billing

Constraints

No baseline funnel existed for testing.

Billing logic was complex and misaligned with user mental models.

Earlier payment introduced operational refund risk.

Research

Seven stages. Mixed methods.

01

Customer service interviews

Tracked complaint categories. Flagged incomplete sign-ups.

02

Internal survey (n=16)

55% flagged unclear communication. Sign-up scored 6/10.

03

Competitive benchmarking

Marley Spoon, Taste Box, Hello Fresh. Clear step patterns.

04

Heuristic evaluation

Nielsen's 10. Three theme clusters identified.

05

Unmoderated testing (Maze)

Blocked. No defined happy path to test against.

06

Moderated usability testing

Three sessions. Navigation and pricing clarity issues surfaced.

07

Measurement planning

Set up GTM triggers per funnel step. GA, PostHog, heatmaps.

Critical finding

All participants expected to pay immediately. Payment was the task completion signal.

This single insight reshaped the solution.

Insights

01

Sign-up felt unclear -- users confused subscription activation with simple account creation

02

Navigation and orientation issues: users couldn't find meal options or return to previous steps

03

Checkout created confusion around subscription vs on-demand ordering and when payment would occur

04

All test participants expected to pay immediately, with payment acting as task completion

05

56% of first billings occurred 1-5 days after sign-up, reinforcing expectation misalignment

06

No clearly defined baseline funnel existed, preventing effective unmoderated testing

Option considered

Pay on Sign Up

Move payment to the start. Use the cart as the entry point. Auto-deduce customer profile from order.

01

Meal Kit Page

02

Add to Cart

03

Checkout

04

Address + Banking

05

Pay

06

Auto-profile

Risks identified

01

Subscription model not communicated before payment. Users commit without understanding recurring billing.

02

Pause and cancel functionality unclear at point of commitment.

03

Solves acquisition only. The retention problem remains unresolved.

04

Order changes post-payment introduce refund risk. Reduced orders take 2+ days to reflect, creating billing confusion at the point of highest user trust.

Solution

Two workstreams. Quick wins shipped independently. The structural redesign tackled the funnel itself.

Quick wins

Communication, navigation, UI clarity.

Structural redesign

Measurable funnel with payment decision.

Revised user journeys

Homepage→Sign Up→
Plans→Activated
Homepage→Sign Up→Get Started→
Plans→Activated
Homepage→Sign Up→Meal Kits→Get Started→
Plans→Activated
Homepage→Let's Get Started→
Plans→Activated

Four paths. One outcome. No dead ends.

Wireframes

Full sign-up flow wireframe showing 6-step funnel
Step 5: Delivery address form with Google Places
Step 6: Payment details form with R1 verification note
Confirmation page showing payment, address, and delivery review

Handoff specifications

scrollIntoView()

Step transitions, error states.

Sticky header

Keeps users oriented during sign-up.

Form validation

Inline errors. Prevent incomplete submits.

Progress tracking

GTM triggers per funnel step.

Results

Old funnel

Entered405
Step 236
Completed15

3.7%

New funnel

Entered1994
Step 2499
Step 3236
Completed185

9.3%

+5.6%

Absolute increase. Abandonment at the top of the funnel dropped from 91.1% to 75%. More users reached later stages.

~R255k

6-month redesign impact ROI · Aug 2025 – Jan 2026

R405k × 63% (3.7% → 9.3%) = ~R255k

Pending

  • Payment timing decision outcomes.
  • A/B testing results.
  • Post-release refund and billing impact.

Reflection

01

Self-initiated research can drive business priorities -- proactively identifying unmet user needs shifted roadmap focus

02

Form friction on mobile is a strategic revenue issue with outsized effect on conversion and trust

03

Leading discovery in low-UX maturity environments requires demonstrating value through data, not theory

04

Mixed methods research builds unshakeable evidence -- combining qualitative with quantitative created a defensible case for change