Enter Password
Making it Painless to Connect to Your Bank
Over 80% of YNAB's customers rely on our 3rd party providers such as Plaid to import transactions and account balances from their banks. This source of data is the lifeblood of the entire YNAB experience.
Connecting to your bank should be seamless, but our experience was obtuse, confusing, opaque, and the source of 45% of our customer support tickets.
Company
YNAB
Industry
Fintech
Year
2024 – 2025
Timeline
6 Months
Role
Lead Product Designer
The Challenge
The experience of connecting to an outside bank was actively breaking our customer's trust.
We had multiple words that were synonyms but meant radically different things
The seemingly obvious path for adding new accounts would break existing connections
There was no way to see which accounts belonged to which banks
Our default settings practically invited users to destroy their budgets
When users can't trust their transaction data, they don't trust the app and they leave.
This project wasn't on our original roadmap. I had to sell leadership by connecting broken bank connections directly to our biggest business challenge: retention. Users who lose trust in their financial data don't stick around.
My Role
I was the lead product designer on this project. In addition to design, I pitched and advocated for this work, created the product roadmap, helped get executive buy-in, and conducted most of the project management.
My Team
1 Part-time PM
2 Full-stack Engineers
1 QA
2 Customer Support Reps
1 Marketing Rep
The Solution
We identified and addressed the most common pain points that led to friction and user error, aiming to create an experience that users could barely remember because it just worked.
The Impact
32%
Drop in related customer support tickets
20%
Increase in support resolution times for related tickets
28%
Reduction in related feature requests
How We Got There
To understand the problem, I shadowed customer support, led hands-on workshops, and documented every nook and cranny of the existing experience.
In a world of potential problems, it quickly became clear that our core user flows were leaving users scratching their heads and wondering, "What is going on??"
I tested changes to the flow in weekly usability tests, helping me hone in on a mental model that worked for our users.
Using quick mid-fidelity wireframes, I was able to test multiple approaches and quickly validate our assumptions, as well as what felt intuitive to our customers.
Solutions & Design Decisions
We shipped multiple solutions in iterative cycles, allowing us to learn as quickly as possible.
Prevent users from accidentally re-adding existing accounts
In the past, users could easily re-add existing accounts, causing headaches. Now, we prompt users with existing accounts to prevent them from accidentally forgetting what already exists.
Proactive account matching and smart defaults to prevent disasters
Added guardrails that prevent users from shooting themselves in the foot. The old system let users create duplicate connections that would break everything.
Bringing in impact & celebration
When users finished adding their accounts, we created a new screen that not only connected their actions to their overall budget but also added a touch of YNAB's personality.
More helpful connection management
We clarified key vocabulary terms and provided users a modern way to visualize and manage their connected bank accounts.
Key Learnings
Break complex redesigns into bite-sized chunks
Instead of a six-month death march, we shipped value monthly. Each piece stood alone but built toward the bigger vision. This approach is now my default for any large project.
Weekly prototype testing is worth its weight in gold
We tested work-in-progress prototypes with users every week – often even before the entire flow was figured out. These sessions helped us catch problems and incorrect assumptions very early, which made them very cheap to fix.
Smart defaults > user education every time
Whereas we'd previously relied on our users reading a mountain of help documents to understand how to complete fairly straightforward flows, we leveraged smart defaults to eliminate as much cognitive load as possible and reduce the number of decisions a user had to make to the absolute minimum.
"Invisible" features require the highest level of usability
Features that should fade into the background need enormous attention, testing, and iteration. Any friction in these kinds of supporting features immediately undermines the core experience and the product's fundamental value.
