People abandon signup at the document step
Not because the form is ugly — because it asks for paperwork they don't have on them. The fix is sequencing, not styling.
Fintech product design
We design payments, banking and lending products for teams who can't just remove the awkward step. Identity checks, disclosures and limits stay. What changes is their order, their honesty and how a person recovers when something is declined.
Talk about your flow →
In most software, a confusing screen wastes a minute. In a financial product it makes someone wonder whether their money is safe — and that suspicion doesn't get repaired by the next screen. So the work is less about visual polish and more about certainty: what just happened, what happens next, when it will be done, and what to do if it isn't.
The second difference is that you can't design your way out of an obligation. Identity verification, source-of-funds questions and consent screens exist for reasons that outlive any redesign. Teams get stuck because they treat that as a dead end. It isn't — it's a sequencing problem, and sequencing is design work.
Problems we're usually called about
Not because the form is ugly — because it asks for paperwork they don't have on them. The fix is sequencing, not styling.
A flow gets designed, reviewed, then sent back because a disclosure or consent step was never in the wireframes. Involving compliance in week one costs a meeting and saves a rebuild.
Two teams shipped the same flow twice. Field validation, error copy and limits now disagree, and support absorbs the difference.
Balance and volume charts on top; the thing a merchant actually opens the app for — did today's payout land — three taps down.
Declined cards, failed verification, blocked transfers and hold periods carry the most emotional weight in a financial product and usually get the least design attention.
Without a named metric before design starts, a redesign gets judged on taste in a stakeholder review and everyone loses.
If several of those sound familiar, a UX research and audit engagement is usually the cheapest way to find out which one is costing you the most.
Capabilities
Progressive identity capture, resumable verification, document upload that survives a bad camera and a worse connection, and honest status communication while a check is pending.
Payout visibility, reconciliation views, dispute handling and limits explained in plain language rather than API terminology.
Transfers, payment requests, recurring collection and refunds — designed with confirmation, reversibility and failure states specified before the happy path is polished.
Eligibility, affordability, offer comparison and post-approval servicing, including what a declined applicant sees and what they can do next.
Disclosures, consents and fee explanations written to be read, then routed through your compliance review instead of pasted in at the end.
Components that treat pending, restricted, verified and blocked as real states, with tokens and accessibility baked in so front-end teams stop re-deciding.
Deliverables
Written into the proposal before we start, so there's no argument later about what "design" included.
How the engagement runs
Phase three is the one most teams skip, and it's the one that decides whether the redesign is allowed to ship.
01
One session with product, compliance and support to agree what the product is failing at and which number would move if we fixed it. We leave with a baseline, not a mood board.
02
Eight to twelve customers or merchants, on their own devices, doing the task for real. We record where they stop, what they go looking for, and what they say to themselves while waiting.
03
Every step gets a source: regulation, internal policy, or 'it's always been there'. Only the third category is free to delete, and it's usually bigger than anyone expects.
04
Fast, memory-answerable questions first. Verification on a resumable track with a real handoff. Waiting states that explain themselves. Then we prototype it.
05
A second round of testing on the prototype, then implementation in increments so you can read the metric between releases instead of after a big-bang launch.
Compliance, in practice
Identity verification, consent capture and record-keeping remain intact. We ask for the source of each requirement so we design against the real obligation rather than a team's memory of it.
Most of the abandonment we find comes from asking for the hardest thing first. Moving verification later — with a limited-capability account in the meantime — usually satisfies both compliance and the customer.
We draft plain-language disclosures and mark anything that makes a claim, then hand it to your reviewers. Nothing we write bypasses your sign-off process.
Where this applies
Merchant signup, settlement visibility, dispute and chargeback handling.
Account opening, card controls, limits, statements and dormancy re-engagement.
Application journeys, affordability, servicing and arrears communication.
Risk profiling, goal setting, portfolio views and honest performance display.
Quote flows, claims submission and status tracking people can follow without calling.
Approval chains, multi-user permissions, bulk operations and audit trails.
Fintech work

Fintech · 2025 · Research, Product design, Design system
Nova Pay was signing up small merchants at scale, but four in ten never finished opening an account. The team assumed it was a compliance problem. It wasn't.
We split the flow in two. Everything a merchant could answer from memory came first, in under two minutes, and unlocked a limited sandbox account immediately. Verification moved to a resumable track with SMS handoff, so a merchant could photograph a document that evening and pick up exactly where they left off.
Services behind this work
FAQs
Anything not covered here, put it in the form — a real person answers, usually within two working days.
Related reading
Send us the flow and the drop-off number. We'll tell you where we'd look first — no charge for that part.
Start a projectLet's talk
We'll schedule a call to discuss your idea. After a short discovery session we send a proposal, and once it's approved we get started.