Simplifying a high-stakes mortgage application flow for first-time buyers, existing members, and refinancing journeys.
I led an end-to-end redesign of a digital mortgage application experience that was long, jargon-heavy, and split between MDX and the Loan Center. The redesigned experience clarified entry routing, reduced cognitive load, improved in-flow guidance, and helped members move through pre-qualification, pre-approval, and application steps with more confidence.
A short spoken overview of the case study for recruiters and hiring managers who want the quick story before scanning the full page.
Members needed to understand where they were, which mortgage path fit their situation, what information was required, and what would happen next. The previous experience made those decisions harder because the flow was fragmented, language was technical, and support moments lived outside the product.
All changes had to remain compliant with lending regulations and pass legal and risk review.
The front end still had to support existing back-end services and required fields.
The redesign needed to work within enterprise components, typography, and interaction patterns.
The team prioritized improvements that could ship within a fixed release timeline.
Some content could be clarified, but not removed or over-simplified beyond compliance requirements.
Members moved between app and web experiences, so continuity and re-entry mattered.
Surveyed about 100 members to understand where people got stuck, what they expected, and what pushed them to call.
Conducted six 60-minute interviews to capture recurring member confusion, rework, and support patterns.
Evaluated usability issues across language, hierarchy, progress, re-entry, and task flow.
Compared mortgage and lending experiences to identify clearer entry, guidance, and calculator patterns.
Mapped end-to-end friction across the online Mortgage Application and Loan Center.
Converted research themes into priorities for flow, language, tools, status, and re-entry.
Banking language made important steps feel more intimidating than they needed to be.
Questions were grouped around system needs instead of user expectations, which increased context switching.
Without calculators, examples, or contextual guidance, members relied on loan officers for questions the interface could answer.
Weak status cues made it hard to know what was complete, what was missing, and what to do next.
If members started in the wrong path, confusion compounded across later steps.
Support worked best when it appeared directly next to the decision or form field where uncertainty happened.
Members needed help choosing the right mortgage path before entering a long application.
Labels, helper text, and CTAs were rewritten to make expectations clearer.
Related questions were grouped together to reduce context switching and miscategorization.
Calculators, examples, and agent-connect options helped members answer questions without leaving the flow.
Progress and next-step guidance helped returning users regain confidence quickly.
The flow reduced visual weight by showing details when they were relevant, not all at once.
I used AI as an exploration and evaluation tool during early problem framing, UX copy variation, edge-case discovery, and synthesis. Final decisions stayed grounded in member research, compliance constraints, usability principles, and enterprise design system standards.
Members picked a goal, such as purchase or refinance, and received set-up guidance before entering the pre-qualification flow.
Users entered the wrong path too easily.
Added goal-based entry routing.
Reduced uncertainty before the long application started.
A scannable hub routed members into pre-qualification, application, or refinance paths with “Best for...” guidance and clearer CTAs.
Members did not know which mortgage path fit them.
Created a decision hub with plain-language descriptions.
Helped users commit to the right path with confidence.
The finances step broke income capture into clearer sections and used targeted guidance to prevent miscategorization.
Financial questions created confusion and context switching.
Grouped related inputs with helper text and progressive disclosure.
Improved accuracy and reduced the need for support.
The strongest product signal was improved completion and conversion from clearer language, better sequencing, stronger entry routing, and more useful in-flow support. The next layer of measurement focused on where members dropped, returned, called for help, or misunderstood path selection.
Mortgage applications are high-stakes because users are making financial decisions that affect their future. The design opportunity was to make a required, regulated process feel less intimidating without removing necessary detail. The result was a more guided experience that helped members understand the path, answer questions in context, and return with a clearer sense of what to do next.
Case study rebuild for senior product design interviews, portfolio reviews, and recruiter scanning.