After the report, silence.

PharmEasy is a health tech platform built to remove friction from healthcare access: medicines, diagnostic tests and home sample collection in one place. It operates where trust, clarity and speed are critical, and every design decision reaches millions of people.

In 2021 the company moved into diagnostics through the Thyrocare acquisition, unlocking a high-frequency, repeat-use category. By early 2022 the goal shifted from acquisition to sustained engagement across the journey, from booking through report consumption. It wasn't working. Traffic grew, but engagement collapsed after checkout, and the experience had no continuity once a report arrived: low repeat usage, heavy support dependency, reduced trust in the whole diagnostics flow.

Method: quantitative funnel and time-to-re-engage analysis, plus 1:1 remote interviews structured around four themes: report delivery and understanding, post-report engagement, trust and retention, and a wrap-up on the most frustrating part of the experience. Small qualitative sample, so every interview claim was checked against the behavioural data before it shaped a decision.

The caregiver persona is the one that changed the architecture. Designing for someone managing three people's health, not their own, is what forced patient switching to be a first-class control rather than a setting.

Competitors had strong data systems and secure report vaults. Trends existed conceptually, but were not surface-level, visual or user-friendly. The category was competing on record access, not on continuous health understanding or action, which is exactly where the opportunity was.

I explored two structures for the same content. Both ended in the same place, guided next steps and continued engagement. They differed entirely in what a user meets first.

I chose Option 1. It starts with health trends rather than medical classifications, so users explore progressively instead of being sorted into clinical buckets on arrival. It reduces cognitive load, and it scales as more reports, parameters and insights are added over time.

Option 2 was tempting because it front-loads the answer: two parameters need attention, here they are. But for an anxious user, opening a health feature to a red "Out of Range" tab is an alarm, not an orientation. And structurally it ages badly: the tab a user lands on changes every time their results change.

The parameter card is the atom of the whole feature. It had to carry a value, its direction, its safe range, its recency and its urgency, without ever reading as an alarm to someone who is already worried.

Moderated usability testing with 10 users aged 18 to 44 of mixed tech proficiency, deliberately including chronic patients, caregivers and recent lab-test users. We tested live and prototype flows across discovery, understanding trends and re-testing, under three scenarios: a recent report available, a retest due, and no historical trends at all.

The point was to check it worked for first-time and repeat diagnostic customers alike. We measured discoverability, onboarding clarity, navigation patterns and graph readability.

The naming failure is the one I'd highlight. "Health Trends" was my own title, on the brief, in every review, in the file name. Eight out of ten people couldn't find it. Testing was the only reason it changed before launch instead of after.