Three cases.
One line of
judgment.

CAREER · PORTFOLIO REVIEW
Maya Chen — Senior Product Designer
01 / 12

Contents

01 / Context

Who I am, what I own

Role scope, the team, and the problems I was hired to solve.

02 / Proof

Three proof moments

The decisions that separated okay work from earned trust.

03 / Artifacts

The artifacts

Flows, prototypes, and the system decisions that shipped.

04 / Impact

Impact, measured

What moved after launch, in the panel's own metrics.

05 / Learning

What my team said

Feedback from the people who reviewed the work.

06 / Ask

The ask

The role I want, and why I'm ready for it now.

02 / 12
42%
Best checkout lift

Career
in numbers

Eight years designing checkout, risk, and onboarding flows for fintech products used by millions. Three case studies, one throughline: earn trust before you ask for money.

Snapshot · Portfolio 2026
3
Case studies presented
8
Years in product design
6
Cross-functional teams led
Users touched by shipped work
18M+
03 / 12

What I own

How I work
01

Own the ambiguity

I take briefs before they have a clean spec — fraud, compliance, and edge-case-heavy flows — and turn them into a shippable point of view.

02

Evidence over opinion

Every direction I ship is backed by a usability read, a funnel number, or a support-ticket pattern — not a hunch.

03

Ship, then defend it

I sit in the review room. Every claim on these slides can be traced to a number, a quote, or a screen.

04 / 12
Section 02 / Proof moments

Three cases.
One consistent
lift.

Completion and satisfaction lift across the three case studies presented here, measured against each product's own baseline before my redesign shipped.

Baseline vs shipped · indexed (baseline = 100)
Baseline
Shipped
CHECKOUTCHECKOUT NPSRISK ALERTSRISK NPSONBOARDONBOARD NPS
05 / 12
Section · Artifacts
03
The work, not the words

See the
flow, not
the pitch.

The next slides walk through the flows, prototypes, and system decisions behind the top-performing case study — the checkout redesign.

06 / 12
PORTRAIT / B&W
"
Maya was the first designer on our team to turn a fraud-ops complaint into a shipped fix within a sprint. The checkout numbers moved because she asked support the right question, not because she guessed better.
VP Product · Payments startup, 2023–2025
David Okafor
07 / 12
Three things I'm asking for

The
ask

01 · This round

A technical deep dive

Walk the panel through the Figma file and the decision log behind the checkout case study, screen by screen.

45 min
02 · If we move forward

A paired exercise

Pair with one of your designers on a live constraint from your roadmap — I want to see how your team actually works.

Half day
03 · What I'm looking for

Senior IC or lead

A team shipping fintech or trust-heavy products, where design sits in the room when risk and growth disagree.

08 / 12
Case study · 02

Checkout drop-off was a trust problem, not a UI problem — and the fix cut abandonment by 42% in six weeks.

What I found

Session recordings and support tickets showed users abandoning at the card-entry step for a reason no funnel metric had captured before.

  • Redundant re-entry — users retyped a card number three screens after they'd already confirmed it.
  • No visible security cue — the padlock and processor logos sat below the fold on mobile.
  • Error copy blamed the user — "invalid card" fired even when the real issue was a network timeout.
  • Save-for-later friction — returning users had no faster path than a first-time buyer.
N = 1,140 sessions · usability panel

Why it matters

$2.3M recovered checkout revenue — modelled, first quarter post-launch.

Checkout is the one screen where a user's trust and a company's revenue sit on the same pixel. Every extra second of doubt costs more here than anywhere else in the funnel.

The team had tried three copy tweaks before I joined. None worked, because the problem was never the words — it was the sequence of trust signals.

Replicated on two later launches with the same pattern. The lift held both times.

Modelled on Q3 2025 cohort

What I did

  • Redesigned the card-entry step to save state and never re-ask for a confirmed number. Shipped: Sprint 14.
  • Moved trust marks above the fold and added a real-time save indicator. Shipped: Sprint 15.
  • Rewrote every error state with the actual cause and a one-tap fix. Shipped: Sprint 15.
  • Built a returning-buyer fast path using saved payment tokens. Shipped: Sprint 17.
Case study 02 of 3 · full flow in appendix
09 / 12
Section · Impact

Completion
climbs, and
holds.

Checkout completion rate by week, indexed against the six weeks before my redesign shipped. The lift is not a launch spike — it holds through the full quarter.

Pre-redesign (control)
Redesign, week 1
Redesign, full quarter
% of sessions completing checkout, by week
1007550250
W-6W-3LaunchW+3W+6W+9W+12
10 / 12

From brief
to shipped,
in five moves.

The same process behind all three case studies — repeatable enough that my team could run it without me by the third time.
01 · Listen

Sit with support

Read raw tickets and watch three real sessions before opening Figma.

02 · Frame

Name the real problem

Turn the pattern into one sentence a PM and an engineer would both sign off on.

03 · Prototype

Build the smallest test

A clickable flow that isolates the one variable we're actually testing.

04 · Validate

Read it against a number

Usability panel plus the funnel metric it's meant to move — never one alone.

05 · Ship

Land it in production

Pair with engineering through launch — I don't hand off a spec and disappear.

Outcome

A case worth presenting

Three of these, back to back, is what's on the table today.

Week 1 · Listen Week 2 · Frame Week 3–4 · Prototype Week 5 · Validate Week 6 · Ship Total · 6 weeks
11 / 12

Why each
case earns
its seat.

Scored against the four things a senior portfolio review actually tests. All three case studies clear the bar on ambiguity and ownership.
Lever
Checkout redesign
Risk-alert redesign
Onboarding rebuild
Time-to-impact
6 weeks
10 weeks
6 weeks
Ambiguity owned
High
High
Medium
Cross-functional leadership
Design + eng + support
Design + fraud ops
Design only
Measured impact
+42% completion
-61% alert fatigue
+18% completion
12 / 12