Three hats, one job
Product manager, designer and researcher were not three roles handed to three people here. They were one job.
The research told me what the proposition had to be. The proposition told me what the interface had to prove. The interface told me what I still did not know and had to go back and test. Splitting that into separate lanes would have added handoffs to a problem that did not need them.
Mapping the whole journey, not the happy path
I mapped the service across six stages: discovery entry, onboarding, payment details and eligibility, reminders, consent to subscribe, and confirmation and post payment.
Discovery alone had three entry points, because people would arrive from a dashboard icon, an SMS with a link, or an in-app notification timed to salary day. Eligibility had to handle the case where someone simply does not have the balance to pay the premium, which for this audience is not an edge case. That branch had to be designed as carefully as the success path, and it opened the question of whether a goal-based saving nudge belonged there.
Every state a policy can be in
Most of the value in an insurance product is not in the moment you buy it. It is in every state afterwards.
I designed for all of them: in progress while payment settles, active, expiring with five days left, expired with a renewal prompt and a warning about the fine, and the case where someone already holds a policy taken outside myZoi and needs reminders rather than a sale. Auto-reminders fire a month before expiry, two days before, and after.
The expired state is the one that matters most. It is where the fine happens, and it is the state most products treat as a dead end.
Walkthrough
The subscription journey as it runs in the app.
Comprehension, not just usability
The subscription flow carries a short explainer video, switchable between English and Hindi, sitting inside the decision rather than buried in a help centre.
That came directly out of testing. People were not failing to tap the right button. They were being asked to commit money to something they could not yet explain back to me in their own words, and could not have claimed against if they had needed to.
The partnership
Launching this meant a working relationship with Dubai Insurance, not just an integration. I built and ran that partnership, understood what each side needed to be ready, and removed the blockers that were holding up launch.
This is the part of product work that rarely appears in a portfolio and usually decides whether anything ships.
Where the research went next
The insights from this work did not stop at the product.
They became a report shared with the Central Bank of the UAE, setting out the gaps in financial inclusion and human-centred design that other fintechs offering job loss insurance can act on. They also produced a lending sandbox business case, now in the works.
Full cycle: research to product, product to policy, policy to the next business line.