A payment infrastructure overhaul that reduced churn, unlocked a new B2B market segment, and replaced manual account manager firefighting with a system that works on its own.
Before I opened Figma, I needed to understand why this mattered to the business, not just to users. The data told a clear story: a meaningful portion of B2B churn wasn't caused by product dissatisfaction. It was caused by payment failures that went completely unnoticed.
When a card expired or a charge failed, nothing happened. No alert. No email. No indicator in the UI. Users discovered problems only when their service was interrupted. By then it was too late to prevent it.
B2B clients often manage multiple budgets across departments. They needed to store and switch between payment methods, but the platform forced manual card re-entry on every single transaction. Enterprise workflows, consumer-grade tooling.
No notifications. No alerts. The only signal something was wrong: their trees stopped being planted.
Account managers were spending hours each week chasing failed payments, manually contacting clients, updating details on their behalf, resolving issues a well-designed system should handle automatically.
SEPA Direct Debit wasn't supported at all. An entire European B2B segment couldn't pay the way they expected to.
“B2B users sometimes have multiple budgets and multiple payment methods, but we didn’t support that. They had to manually re-enter details every single time.”
Key insight from stakeholder researchEvery transaction required full card entry from scratch. No saved methods, no defaults, no flexibility. For a B2B client managing multiple subscriptions across departments, this was a genuine operational burden, not a minor inconvenience.
I combined data analysis, stakeholder interviews, competitive benchmarking, and journey mapping to build a complete picture: not just of what users wanted, but of what the business needed to recover.
Failed payment rate analysis confirmed the hypothesis: churn was being driven by silent payment failures, not product dissatisfaction. This reframed the project from “improve UX” to “recover revenue”, giving it the business priority it deserved.
Working with the CEO and PM, I uncovered the full operational burden. Account managers were manually chasing payment issues every week. The team had integrated Stripe in a previous iteration but hadn't built the layer that let users actually manage their methods.
Reviewing how Stripe, Shopify, and comparable SaaS platforms handle payment management informed the UI patterns and confirmed that SEPA support was table stakes for European B2B, not a nice-to-have.
Collaborative sessions with the CEO and PM made the problem undeniable. Mapping the full payment lifecycle (from adding a method, through a silent failure, to a disrupted service) aligned the team and made solution priorities clear.
Payment flows are deceptively complex. The visible UI is the tip of the iceberg. The real work lives in the edge cases, and getting those wrong in a payment context isn't a UX issue, it's a revenue issue.
I mapped every failure state before writing a single UI spec: deleting the only card linked to an active subscription, expired cards mid-billing-cycle, failed retries, reassigning defaults. Each scenario had a deliberate, tested resolution path.
"Most people design the happy path. Designing the 'you can't do that, here's why, here's what to do instead. That's the work."
Design principle guiding this projectThe final system let users store multiple payment methods with custom aliases, set a default, add SEPA Direct Debit, and receive proactive alerts, all within a clear, low-friction interface.
The 30% recovery target was ambitious. Reaching ~20% in the initial rollout is meaningful, and the foundation is in place to continue improving as retry logic and notification flows mature.
That's worth naming directly. The discovery and workshop phase ran longer than expected. From the outside, it can look like a designer doing research when you could just start building. From the inside, I was making sure we didn't build the wrong thing, or the right thing with the wrong edge cases, which in a payment flow can be very expensive to fix post-launch.
The outcome validated the investment. But the tension was a useful signal: in future projects, I'd front-load a tighter alignment session at kick-off to set explicit timeline expectations, and identify earlier where the process could be compressed without sacrificing quality.
This is the calibration I think about most as I move toward Product Lead roles. There's no universal answer to "how much process is the right amount." Getting better at reading that, and negotiating it with stakeholders in real time, is the craft above the craft.
Running collaborative sessions with the CEO and PM at every major decision point meant faster alignment, fewer revision cycles, and genuine stakeholder ownership. The best decisions came from the room, not from a Figma handoff.
With more time, I'd have tested with real B2B clients, especially around the SEPA flow and failed payment notifications. Internal testing caught issues, but live enterprise feedback would have added confidence and surfaced edge cases faster.
Payment infrastructure is where product design gets real. The stakes are financial, the edge cases are endless, and “we’ll fix it later” is not an option.
I started with a business problem (revenue leakage, operational debt, an underserved B2B market) and worked backwards to a system that resolved all three simultaneously. That’s the kind of structural thinking I bring to product work. And the kind of Product Lead I’m growing into.