Fintech · Mobile product design

From Friction to Flow: Redesigning the Registration Process for a Fintech App

I led Dough’s onboarding redesign, using funnel data and user feedback to reduce friction and help more prospective investors complete registration.

Role

Lead Product Designer

Role

Lead Product Designer

Role

Lead Product Designer

Role

Lead Product Designer

Team

Product manager, four full-stack developers, marketing, and customer support

Team

Product manager, four full-stack developers, marketing, and customer support

Team

Product manager, four full-stack developers, marketing, and customer support

Team

Product manager, four full-stack developers, marketing, and customer support

Timeline

Approximately four months

Timeline

Approximately four months

Timeline

Approximately four months

Timeline

Approximately four months

Status

Shipped

Status

Shipped

Status

Shipped

Status

Shipped

The problem

A 21-screen brokerage application was losing 52% of prospective users, especially around verification, sensitive information, and a late-stage section.

The direction

I combined funnel data, user feedback, support themes, and comparative testing to identify the highest-friction moments, remove steps, and clarify unavoidable requests.

The outcome

The shipped flow removed five screens. Completed registrations rose 22% after an iterative rollout that also included technical fixes, clearer messaging, and a new-user incentive.

The outcome

The shipped flow removed five screens. Completed registrations rose 22% after an iterative rollout that also included technical fixes, clearer messaging, and a new-user incentive.

01 · Context

A product for beginners was losing them before they could begin

01 · Context

A product for beginners was losing them before they could begin

01 · Context

A product for beginners was losing them before they could begin

01 · Context

A product for beginners was losing them before they could begin

Dough was a mobile investing product for people with limited trading experience. Before users could invest, they had to complete identity checks, receive approval, connect a bank through Plaid, and fund the account. The journey stretched across at least 21 screens, and Braze showed 52% abandonment.

Many requirements were regulatory or tied to third-party services. The opportunity was to eliminate avoidable effort, explain what remained, and help users recover from delays or failures.

02 · Direction

Reduce both the effort and the uncertainty of opening an account

02 · Direction

Reduce both the effort and the uncertainty of opening an account

02 · Direction

Reduce both the effort and the uncertainty of opening an account

02 · Direction

Reduce both the effort and the uncertainty of opening an account

The redesign could not eliminate every requirement. I focused the team on the moments most likely to make a prospective investor stop: a journey that felt endless, sensitive questions without enough context, technical delays, and weak guidance about what would happen next.

01 · Follow the evidence

Prioritize the screens associated with the greatest abandonment instead of treating every part of the journey equally.

02 · Reduce perceived effort

Remove unnecessary steps and make the remaining process feel achievable.

03 · Earn trust before asking

Explain why sensitive information is required at the moment it is requested.

04 · Design beyond the ideal path

Help users continue when verification, bank connection, or funding does not happen immediately.

The broader customer journey continued beyond account opening through approval, bank linking, funding, interests, and a first trade. Delays and dependencies could compound across each phase.

Key product decisions

Three decisions that shaped the experience

Decision 01

The funnel showed where to investigate. Users explained why.

Situation

The 52% abandonment rate confirmed a problem but did not explain where or why people were leaving.

Evidence

Braze identified the largest losses around screens 9, 11, and 18. User and support feedback connected them to slow verification codes, sensitive-information requests, and a late section that made the journey feel as though it were restarting.

Decision

I centered the redesign on faster verification, contextual trust, and clearer late-stage orientation.

Result

The team moved from a broad onboarding problem to three specific issues that design, engineering, and product could address together.

The original application contained 21 required screens after the entry point. Braze identified screens 9, 11, and 18 as the largest abandonment points.

User conversations and support feedback connected those losses to delayed verification codes, sensitive info hesitation, and a late-stage reset.

Decision 02

Shorten the journey, then change how progress feels

Situation

The 21-screen application was divided into multiple sections, while a slow-moving progress bar repeatedly emphasized how much work remained.

Evidence

Users described the journey as too long. Funnel data also showed a major drop when a new section appeared late in the process, making the application feel as though it were starting over.

Decision

I removed five screens, along with the section interstitials and granular progress bar. Short contextual labels instead explained what users were completing and what would happen next.

Result

The 16-screen journey had fewer interruptions and gave users clearer orientation without constantly reminding them how much work remained.

The revised flow reduced the core journey from 21 to 16 screens, removed section interstitials, and replaced the progress bar with contextual guidance.

Decision 03

Explain sensitive requests at the moment of hesitation

Situation

Dough requested sensitive personal and financial information before many prospective investors had developed trust in the product.

Evidence

Users hesitated when asked for information such as their Social Security number. Inconsistent hierarchy, inputs, and messaging made the unfamiliar financial process feel less reassuring.

Decision

I added concise explanations at the moment sensitive information was requested and established clearer, more consistent interface patterns. We also improved recovery messaging and added one-time-code autofill to reduce friction during verification.

Result

Users could better understand why information was required and complete verification with less effort, while the team gained a stronger visual foundation for the rest of the product.

The original screen provided limited visual reassurance during one of the application’s most sensitive requests.

The revised screen gives the regulatory explanation greater prominence, reinforces progress, and applies the new interface system.

03 · Experience

A clearer path from account creation to a funded account

03 · Experience

A clearer path from account creation to a funded account

03 · Experience

A clearer path from account creation to a funded account

03 · Experience

A clearer path from account creation to a funded account

The final experience guided a first-time investor through the required account-opening process with fewer interruptions, clearer context, and better support when a step did not happen immediately.

Step 01 · Create an account

Set expectations for the process while collecting the basic information needed to begin.

Step 02 · Complete identity and approval details

Explain why personal information is required and make verification easier to complete.

Step 03 · Connect a bank

Guide users through Plaid. Provide clear messaging when a bank is unavailable or a connection can't be completed.

Step 04 · Fund the account and begin

Set expectations about transfer timing and create a clear transition from account setup into the investing experience.

04 · Validation

Testing with engaged customers, unfamiliar users, and the team

04 · Validation

Testing with engaged customers, unfamiliar users, and the team

04 · Validation

Testing with engaged customers, unfamiliar users, and the team

04 · Validation

Testing with engaged customers, unfamiliar users, and the team

I tested the redesign with approximately ten users, supplemented by comparative sessions through UserTesting.com and internal dogfooding across teammates with different levels of investing experience.

Many direct participants came from Dough’s loyal livestream community. Their feedback was detailed and product-aware, but potentially more forgiving than feedback from first-time prospects. Remote testing helped us evaluate the experience without that existing familiarity.

Participants

Approximately 10 users

Participants

Approximately 10 users

Participants

Approximately 10 users

Participants

Approximately 10 users

Method

Moderated testing, comparative remote testing, and internal dogfooding

Method

Moderated testing, comparative remote testing, and internal dogfooding

Method

Moderated testing, comparative remote testing, and internal dogfooding

Method

Moderated testing, comparative remote testing, and internal dogfooding

Duration

Throughout the four-month redesign

Duration

Throughout the four-month redesign

Duration

Throughout the four-month redesign

Duration

Throughout the four-month redesign

What We Learned

Testing confirmed three sources of friction: delayed verification codes, hesitation around sensitive requests, and a structure that made the application feel longer than it was.

What Changed

We added code autofill and recovery messaging, explained why sensitive information was required, and removed five screens, section transitions, and the progress bar.

Why it mattered

The revised flow reduced effort and helped prospective investors feel more informed and oriented throughout a regulated account-opening process.

Testing Insight

Familiarity became a research variable: loyal customers exposed detailed interaction issues, while unfamiliar users revealed gaps in clarity and trust.

Testing Insight

Familiarity became a research variable: loyal customers exposed detailed interaction issues, while unfamiliar users revealed gaps in clarity and trust.

Testing Insight

Familiarity became a research variable: loyal customers exposed detailed interaction issues, while unfamiliar users revealed gaps in clarity and trust.

Testing Insight

Familiarity became a research variable: loyal customers exposed detailed interaction issues, while unfamiliar users revealed gaps in clarity and trust.

05 · Outcome

A shorter journey that helped more people complete registration

05 · Outcome

A shorter journey that helped more people complete registration

05 · Outcome

A shorter journey that helped more people complete registration

05 · Outcome

A shorter journey that helped more people complete registration

21 → 16 screens

Five screens and several interruptions were removed from the core account-opening journey.

21 → 16 screens

Five screens and several interruptions were removed from the core account-opening journey.

21 → 16 screens

Five screens and several interruptions were removed from the core account-opening journey.

21 → 16 screens

Five screens and several interruptions were removed from the core account-opening journey.

22% more completed registrations

The total number of completed registrations increased following the iterative rollout.

22% more completed registrations

The total number of completed registrations increased following the iterative rollout.

22% more completed registrations

The total number of completed registrations increased following the iterative rollout.

22% more completed registrations

The total number of completed registrations increased following the iterative rollout.

Three critical drop-off points addressed

The redesign focused on verification delays, sensitive-information requests, and late-stage progress.

Three critical drop-off points addressed

The redesign focused on verification delays, sensitive-information requests, and late-stage progress.

Three critical drop-off points addressed

The redesign focused on verification delays, sensitive-information requests, and late-stage progress.

Three critical drop-off points addressed

The redesign focused on verification delays, sensitive-information requests, and late-stage progress.

A reusable UI foundation

New patterns for fields, typography, spacing, hierarchy, and explanatory content improved consistency beyond onboarding.

A reusable UI foundation

New patterns for fields, typography, spacing, hierarchy, and explanatory content improved consistency beyond onboarding.

A reusable UI foundation

New patterns for fields, typography, spacing, hierarchy, and explanatory content improved consistency beyond onboarding.

A reusable UI foundation

New patterns for fields, typography, spacing, hierarchy, and explanatory content improved consistency beyond onboarding.

The redesigned experience shipped during my tenure. Completed registrations rose 22% after an iterative rollout that also included technical fixes, clearer messaging, and a new-user incentive, so the increase cannot be attributed to the UX redesign alone.

06 · Delivery

Designing the journey while creating the system behind it

06 · Delivery

Designing the journey while creating the system behind it

06 · Delivery

Designing the journey while creating the system behind it

06 · Delivery

Designing the journey while creating the system behind it

As Dough’s lead and only product designer, I owned funnel analysis, research, interaction and visual design, prototyping, testing, and delivery. I partnered with the product manager and four engineers, while marketing and support contributed incentive strategy, community access, and recurring customer issues.

The redesign also established reusable patterns for typography, inputs, spacing, controls, and explanatory content, giving the team a more consistent foundation beyond onboarding.

Tools Used

Figma · Braze · UserTesting.com · Plaid · Cross-functional product delivery

Tools Used

Figma · Braze · UserTesting.com · Plaid · Cross-functional product delivery

Tools Used

Figma · Braze · UserTesting.com · Plaid · Cross-functional product delivery

Tools Used

Figma · Braze · UserTesting.com · Plaid · Cross-functional product delivery

Early sketches established the screen structure. Those decisions evolved into reusable patterns applied across onboarding, verification, transfers, and account funding.

07 · Reflection

What I would carry into the next regulated product

07 · Reflection

What I would carry into the next regulated product

07 · Reflection

What I would carry into the next regulated product

07 · Reflection

What I would carry into the next regulated product

What worked

Combining screen-level analytics with user conversations and support evidence turned a vague funnel problem into specific product hypotheses. It helped the team distinguish required account-opening complexity from friction we could actually remove or explain.

What I would reconsider

I would define the measurement plan before rollout and stagger the redesign, technical fixes, messaging, and incentive experiment. I would also recruit more prospective users with no prior relationship to Dough.

What I would explore next

I would go deeper on recovery paths for delayed verification, unsupported banks, connection errors, and funds that were not immediately available. In financial products, these moments can determine whether someone trusts the experience even when the ideal path works well.

Next project

HappyPath

Designing and building a field-ready CRM that connects lead discovery, follow-ups, and route planning for vending operators.

View case study →

Next project

HappyPath

Designing and building a field-ready CRM that connects lead discovery, follow-ups, and route planning for vending operators.

View case study →

Next project

HappyPath

Designing and building a field-ready CRM that connects lead discovery, follow-ups, and route planning for vending operators.

View case study →

Next project

HappyPath

Designing and building a field-ready CRM that connects lead discovery, follow-ups, and route planning for vending operators.

View case study →