Case Study — Consumer Health App

Ephira

Co‑Founder & Product / UX Lead, alongside project partner Alex Ulbrich · Sept 2025 – May 2026

Live app site GitHub (open source) App Store Google Play

The Problem

Most cycle-tracking apps monetize the most sensitive data a person has — and they still force a hard break the moment a user gets pregnant. Existing tools make people abandon their cycle history and switch to an entirely different app, losing continuity right when it matters most.

"Fragmented care leaves expecting parents in the dark."

That line, from our expo positioning, was the whole thesis: build one privacy-first product that adapts across a user's life stage instead of forcing a platform switch.

Research → Product Direction

Before touching UI, I synthesized peer-reviewed research on pregnancy-app usability (app-store sentiment analysis, a 31-app clinical quality review) alongside a competitive UX teardown, and kept surfacing the same four gaps: emotional isolation, confusing/scattered medical data, "pregnancy brain" memory lapses, and unclear nutrition guidance. That research became the argument for Ephira's pregnancy mode — the same local-first, no-account architecture, extended rather than replaced.

My Role

I owned product design and code quality on a three-person engineering team, working directly with our founder/stakeholder to keep every UI decision privacy-first — reviewing PRs specifically to confirm no accidental analytics or network calls slipped into the app.

Gradient cycle-flow visualization — replaced the old dot-based flow indicator with a color-coordinated gradient (spotting → light → medium → heavy flow) that reads instantly at a glance, unified across iOS and Android.
Home screen "Current Phase" card — redesigned the phase indicator and next-period countdown for faster comprehension.
Calendar tab — rebuilt the flow-logging dropdown around the same unified color system.
Pregnancy Mode v1 — designed and shipped the first tab states (home, info, calendar) and the structural layout for 40 weeks of educational content.

A bug worth naming My first version of the gradient only transitioned correctly in one direction — tapping flow states in reverse (heavy → light) or skipping a level broke the visual entirely. I traced it to a directional assumption baked into the color-interpolation logic, rebuilt it to handle arbitrary state transitions, then formalized 16 edge-case and 8 fallback test scenarios around cycle prediction and flow logging so the same class of bug couldn't ship silently again.

UI Implementation

The three screens below are shipped, in-app — not mockups. Each one reflects a UI decision I owned end to end.

Ephira home screen with the gradient cycle tracking ring

Home — gradient tracking ring replacing the old dot indicator

Ephira cycle tab showing current phase, day count, and next period estimate

Cycle tab — "Current Phase" card + next-period estimate

Ephira calendar with color-coded flow and birth control logging

Calendar — unified color system for flow, birth control & symptom logging

Go-to-Market & Positioning

I led the product demo and UI/UX walkthrough in our stakeholder presentation, then presented Ephira's go-to-market and positioning strategy to a panel of judges at the OSU Capstone Expo — framing the product around its core differentiator (local-first privacy with zero telemetry) against cloud-based competitors, backed by the app's MIT-licensed, publicly auditable codebase as proof.

Process

Across three academic terms the team shipped through 156+ pull requests under a two-reviewer approval policy with CI-enforced lint and test gates — no PR merged without passing automated checks and a peer sign-off.

Outcome

500+downloads across iOS & Android

16 / 8edge-case / fallback test scenarios

156+PRs shipped across 3 terms

React NativeExpoSQLite + SQLCipher (AES-256)Drizzle ORMMIT open source

What I'd carry forward

The clearest lesson from the retrospective was the simplest one: an idea that sounds great in a design review means nothing if it doesn't hold up for a real, everyday user — which is exactly why the edge-case testing mattered as much as the visual redesign.