Paytient is a B2B2C fintech platform that partners with employers & health plans to help members pay for healthcare over time. In 2024, the company entered the Medicare space to serve a massive & underserved market of 25M+ eligible Americans who face complex prescription costs, low digital literacy, & strict regulatory requirements.
I joined as the 1st & lead designer on this initiative. There was no existing product, no user research, and no design precedent for how a financial health platform should work for this population.
As the Lead product designer, I was responsible for defining the product architecture, architecting the design system, bridging design & engineering, & elevating org-wide design quality.
Product Managers
Engineers
Content Strategist
Legal & Compliance partners
Client partners
I built skills, scripts, & systems to compress my own discovery & design cycles. Once I proved the efficiency gains, I trained & present them to our entire product team. I developed & maintained our AI-Augmented workflow playbook.
I created design principles, voice & tone guidelines, research templates, design critique structures, etc. All artifacts I created that now live beyond my individual work.
I initiated complete competitive analysis of other Medicare products, conducted secondary research, built data dashboards entirely on my own initiative to shared it as a team resource. These digestible research decks continue to inform our product strategy & roadmap.
I built a decision-making system through designing out our UI component library. Each of the components carries embedded accessibility requirements, interaction logic, & usage guidelines. This ensures that when another designer or engineer picks up a component, they make the right choices by default.
My Deque training benefited the product I was designing & I used it to create documentation & templates that other product managers & designers would use when beginning their own accessibility work.
Designing a Medicare financial health platform presented challenges that went far beyond typical product design. Medicare beneficiaries had never used a product like this, but neither had the advocates serving them. There was no established pattern for how a trained operator completes a payment on someone else's behalf, under call-time pressure, with compliance liability attached to every field.
There was no precedent for how medicare beneficiaries navigate prescription financing, & none for how an advocate navigates it for them. I was designing both sides of a relationship that had never had a tool.
Every screen, every interaction, every piece of copy had to meet CMS compliance & client requirements (Humana, Express Scripts, & CarePlus). Design directions were led by a balance of UX decisions & legal decisions. A wrong edge case was not just a bad experience but a compliance exposure.
The members had varying digital literacy, many were non-native English speakers, and many interacted only through an advocate.
No user research existed. No design system. No component library. No voice & tone guidelines. Everything had to be built from nothing, which meant the responsible first move was to reuse what the org already trusted.
The stakes were high & the precedent was zero, so before I drew a single screen I made sure I understood the problem cold. I built a repeatable practice for doing that using AI skills I built. Using AI as a tool to push me in my understanding of the domain & problem at hand, allowed me spend my judgment where it mattered & compress the work that did not need it.
It got me to a complete first draft in an hour instead of a day. The thinking that mattered went into the editing.
I turned a scattered set of open questions into a plan we could actually run, and brought the team into it rather than handing it down. This is why my discovery held up later.
My first direction used the modal flow Paytient had already built & validated for its other internal admin tool. Reusing the house pattern was the cheap, safe, correct-by-default move. It cost engineering nothing new, it was already in the system, and no one would question it. Here is a glimpse at my first designs for the payment flow taking these deicisions into consideration.
When I led in-person & remote research sessions, working directly with Medicare beneficiaries & their advocates, one finding in particular undid the V0 designs.
Through our sessions, it became clear that an advocate has to read the payment history, find the failed charge, confirm the correct amount, & complete the payment without an error that becomes a regulatory problem. That investigative work needs room, trying to squeeze all the context into single modals would increase cognitive load & would not scale well long term.
The closer I got to the advocate's real workflow, the more the modal fought me. There was not enough room to show the payment history an advocate needed to diagnose a problem, and stacking modal on modal to compensate made the on-behalf-of flow harder to follow, not easier. So I redesigned the layout to address these pain points using a full page layout.
I had already committed to the modal. The evidence made me undo my own decision, and then argue my team into paying for the harder path I had just talked myself into. The full-page path cost more, it meant a separate design system for this user group, net-new design work, & net-new engineering. And it meant I had to win full team alignment to justify that spend rather than quietly shipping the 'cheap' thing.
I advocated for this redesign by focusing on 2 main arguments: the user's real needs uncovered in research and future scalability of these designs. With clients, I shared how more real estate meant fewer cramped, error-prone steps, & stronger compliance. And the argument that won engineering: full-page flows were more maintainable over time, the modal was cheaper to build once but more expensive to live with long term.
As I worked through revisions of the design solution for our CRM tool based on stakeholder input, I remained proactive about revising my solutions directly with the end user, shadowing their current workflows & running quick usuability sessions with prototypes. As I gathered more evidence my solutions improved to better improve business outcomes & address user needs as shown below through the payments tab.
in revenue generated. Making up ~33% of the company's revenue from a product I led as sole designer.
success rate achieved in over 1 million processed payments through UX layout decisions
ticket processing average achieved through the intentional layout of the CRM platform.
I pour everything into my work. Yet the end of every project reveals what might've been missed in the beginning, sharper questions I should have asked, frameworks I've since built, and tools that didn't exist when I started. That gap is where growth lives and I chase it deliberately.
Even before putting pen to paper, I sought out weekly sessions with direct end users. I'd either shadow their current workflows on other platform and/or share early design concepts for feedback giving me a deep & holistic understanding of my users through the entire process.
This is the first complex project I got to bring AI into my entire design process. I was intentional about using AI to expedite the time consuming task that could be automated, such as design decision documentation, while saving my focus on complex problem solving and strategy.
The training, documentation, and frameworks I created had real ROI. It ended up reducing onboarding time, fewer handoff errors, faster alignment. Had I known the potential impact, I wish I had been tracking those metrics from the start.
