Transfer Chase Points to Travel Partners

Chase's most strategically valuable redemption feature was about to change its rules. I redesigned the experience before users felt the confusion.

Client

JPMorgan Chase

Year

2025-2026

Timeframe

Ongoing

Role

Lead UX Designer

Outcome

Complaint -13.5% MoM · -11.3% YoY 1 month post P1 release · Completion rate +61 - 67%

WEB

NATIVE

TASK FLOW

Desktop

Mobile

chase.com

Before

Now

Future

10B

UR points redeemed every month

45%

mobile completion rate

37.1M

Rewards users

424K

Monthly visits

35%

Hyatt redemption

24%

United redemption

15%

Southwest redemption

The volume was there. The experience wasn't keeping up.

Transfer Points is the second most-used redemption across Chase's premium card portfolio — Sapphire, Ink, and JPMorgan cards. The feature was working. But working isn't the same as working well.

THE CHALLENGE

Three failure points. All happening before users could complete a transfer.

01 — Wrong input increment

The field accepted any number. Only 1,000 increments would work.

No constraint, no hint, no prevention — just a silent failure after submission. Flagged consistently in customer complaints and VOC aggregation.

02 — Insufficient balance

Users didn't find out until the confirmation step.

By then, they'd already gone through the entire flow. Identified through backend error log analysis.

03 — Identity mismatch

A backend limitation that looked like user error.

Cardholder name and loyalty program credentials didn't always match — causing the transfer to fail at the final step. Top complaint theme across social listening and support ticket data.

45%Mobile

52%Web

Completion rate across platforms, monitored in production via internal analytics dashboard.

Completion was underperforming, but usage was high. That made the opportunity clear: use these three failure points to set priorities across UX, content, and tech — then look deeper to understand what was really getting in the way.

HOW I WORKED

I didn't start with solutions. I started with signal.

AI-assisted research

Equipped before exploration

Used AI to surface current redemption flow trends across mobile and web, and to aggregate raw VOC into themes — so I entered exploration with signal, not assumptions.

Cross-functional alignment

Alignment as a continuous practice

Worked with PM, content strategy, UX research, and tech throughout. As lead designer for the full book of work, alignment wasn't a phase — it ran the length of the project.

Production analytics

Every problem has a number

Pulled completion rates, drop-off points, and error frequencies directly from backend data. Every failure in the previous section has a metric behind it.

For the detailed design and user testing process, please request the full case study.

PHASE 1 — WHAT I SHIPPED

Three decisions. Each one targeting a specific failure point.

01 — Points Calculator

Transparency before commitment

Added a pre-transfer calculator to the entry point. Users could see estimated redemption value before entering the flow — removing the discovery-too-late pattern that caused drop-off after commitment.

Design rationale

Informed decisions require visible consequences. Surface the outcome before asking for the action.

Instant Value

Check the conversion rate immediately so you can decide if you want to proceed without completing the full form first.

02 — Input constraint

Error prevention at the field level

Introduced a visual ",000" suffix that made the 1,000-increment constraint visible within the input itself. No error message needed — the format communicated the rule before the mistake could happen.

Design rationale

The best error message is one that never has to appear. Encode constraints in the UI, not in the feedback loop.

Smart Increment

Appends ",000" to your entry to match the required thousand-unit increments.

03 — Credential clarity

Reducing identity mismatch at the final step

Redesigned the credential entry step to surface inline guidance: which name to use, where it comes from, and why it matters. Turned a silent backend constraint into an informed user action.

Design rationale

Backend limitations aren't user problems — unless we make them one. Own the explanation.

Name Matching

Ensure this account number belongs to the cardholder named above.

These fixes made the transfer flow clearer, but they also pointed to a larger principle: redemption systems should prevent errors before users collide with them, using UX, technology, and content strategy to reduce both input mistakes and system-driven confusion.

BEYOND THE SCREEN

Three fixes. One pattern that scales.

The immediate problem was three error states. The underlying opportunity was a design principle that applies across every transfer surface in Chase's redemption ecosystem — and eventually, across the full points portfolio.

Error prevention

Encode constraints in the UI — not the error message

The ',000' suffix pattern isn't limited to Transfer Points. Any numeric input tied to a business rule — minimum amounts, batch sizes, denomination steps — benefits from the same approach. One pattern, scalable across the full product surface.

Transparency

Surface outcomes before asking for commitment

The calculator pattern removes a class of drop-off that occurs when users discover consequences too late. Applied upstream — at product entry, at feature discovery, at onboarding — this principle reduces abandonment across the redemption journey.

Identity clarity

Own the explanation when backend has limitations

Credential mismatch is a systemic risk wherever loyalty accounts interface with payment infrastructure. Inline guidance at the point of mismatch — not after failure — is a reusable pattern across partner integrations.

Design system

These principles were translated into new UI and pattern logic, giving the design system stronger material to evolve from. Scalability is at the core of UX work.

RESULTS

The errors went down. The completions went up.

Measured in production. Data collected over 60 days post-launch across mobile and web platforms.

Mobile completion rate

45%→61%

↑ 16pp lift on mobile — largest single-release improvement in the transfer flow's history.

Web completion rate

52%→67%

↑ 15pp lift on web. Parity with mobile closing for the first time.

Error 01 — Wrong input

↓ 91%

Input-related errors effectively eliminated after the ',000' suffix shipped.

Error 03 — Identity mismatch

↓ 64%

Credential clarity guidance reduced mismatch failures at final submission.

Error 02 (insufficient balance) reduction tracked separately via backend error logs. Full dataset pending Q3 reporting cycle.

OUTCOME - REDEMPTION FLOW

Six steps. One seamless journey.

TRANSFER POINTS TASK FLOW

Landing Page

Partner Detail Page

Point Input

Partner Account Information

Review

Confirmation

WHAT'S NEXT

Phase 1 fixed the failures. Phase 2 makes the product intelligent.

The polished UI and flow fixes matter — they make Transfer Points clearer, calmer, and much easier to finish. But that was never the whole story. The real question became: how do we make this experience feel smart enough that users actually want to come back to it?

AI recommendation engine

In discovery

Personalized transfer suggestions at the point of entry

Using redemption history, partner affinity, and current point balance to surface the most relevant transfer partner — before the user has to choose.

Proactive balance alerts

Backlog

Know you're eligible before you open the flow

Push notification or in-app alert when a user's balance crosses a meaningful threshold for their most-used partner. Remove the need to check manually.

Smart input defaults

Backlog

Pre-fill the amount based on redemption history

If a user transfers 10,000 points to Hyatt every time, the field should say 10,000. The constraint is already known — the default should reflect it.

These are early-stage directions, not confirmed roadmap items. The design work exists as concept specs shared with stakeholders and discussed in the Phase 1 retrospective.

CONFIDENTIALITY NOTICE

This case study contains information regarding project details and design exploration. To comply with non-disclosure agreements (NDAs), certain sensitive data—including specific merchant partner identities and proprietary metrics—has been redacted or modified. Additionally, while the core design strategy, collaborative processes, and analytical methodologies remain accurate to the original project, specific UI elements and final design outcomes shown here may differ from the launched product.

Should you require further clarification or wish to discuss the project in greater detail, please feel free to reach out to me directly.

CONFIDENTIALITY NOTICE

Should you require further clarification or wish to discuss the project in greater detail, please feel free to reach out to me directly.

More Projects

Tracker Design System Framework

Chase Ultimate Rewards Core Experience