2024 · Product Designer & Researcher
AIRA
A circadian rhythm app for PCOS management
Role
Product Designer & Researcher
Team
4-person team (shared build)
Research Methods
User Interviews, Survey, Behavioral Science Review, Expert Consultation
Platform
Mobile App
Duration
12 weeks
Hook
PCOS doesn't have one fix. It has forty, sleep, food, stress, cycle, mood, all feeding into each other, and almost every tool out there asks you to manage one of them like it's the whole problem. I have PCOS. I know what it's like to open four different apps just to understand why today feels harder than yesterday, and still not have anything track the one thing, mood, that actually explains it.
My Role
Team size
4
What I led
Concept ideation, problem identification, secondary research on mental models and what existing PCOS/tracking tools miss, and design of the period tracking and food tracking screens.
What was shared
Sleep tracking (teammate), mood tracking (teammate), sign-up/onboarding (teammate).
Research
Methods
5 interviews, 20+ survey respondents, one endocrinologist consultation.
Key findings
PCOS is psychologically loaded, and because symptoms present so differently person to person, there's no one-size-fits-all fix. Women with irregular periods were put off by phase-based period trackers (luteal, follicular, etc.) because irregularity didn't fit the categories, and that mismatch read as shame about a symptom they couldn't control. And tool abandonment wasn't a motivation problem: people stopped logging when the effort outweighed the visible benefit, or when the tone felt judgmental instead of supportive.
One thing that stuck with me
A participant described managing PCOS across separate sleep, food, and period apps as feeling like a full-time job, and pointed out that nobody was tracking mood, even though mood is one of the clearest ways to spot what's actually triggering a bad symptom day.
The Insight That Changed Everything
An app built to help women with PCOS was making some of them feel ashamed of a symptom they had no control over. That's what the phase-tracker interviews surfaced, women with irregular cycles avoiding the very tool meant to support them, because it kept sorting them into phases their body didn't follow.
That reframed the whole project. It stopped being about tracking habits and became about tracking capacity, and giving people permission to rest instead of a reminder that they'd failed to log something.
[The guilt cycle diagram — log a bad day, feel judged, stop using the app]
How Might We
How might we help women track their reality without the app quietly telling them they're doing it wrong?
Design Decisions
Decision 1
Phase/season toggle on period tracking
A switch between clinical phase names (luteal, menstrual) and a season metaphor (spring, winter) for the same cycle data. Built directly for the women who found phase-based tracking shame-inducing when their cycle was irregular, same information, a kinder frame.
Decision 2
Readiness Score over streak counter
Streaks punish inconsistency, exactly what someone managing fluctuating PCOS symptoms can't promise. A Readiness Score reflects current capacity instead, which fits the “permission to rest” reframe.
Decision 3
Non-judgmental logging (photos, no calorie counts)
Directly answers the tool-abandonment finding: judgmental tone was a bigger dropout cause than effort. Photo logging removes the numbers that read as diet culture.
Decision 4
Dark mode as the default
Dark by default keeps AIRA calm and personal instead of clinical. Most health and medical apps default to stark white, which reads sterile and diagnostic, the opposite of the permission-to-rest tone this project is built on. A softer, darker interface feels more like a companion than a chart. Honest note: this was locked in late in the process, and a teammate had pushed for light mode. A light-mode version and a full accessibility pass are on the v2 list.
Decision 5
The Energy Wave visualization
Nobody in the interviews was tracking mood, despite it being one of the clearest ways to spot symptom triggers. The Energy Wave gives a non-numeric, pattern-over-time view instead of reducing energy or mood to a single score.
[Readiness Score diagram — sleep, stress, symptoms, energy inputs]
[Annotated home dashboard — Energy Wave + Readiness Score]
The Solution
AIRA: a PCOS management app built around capacity, not compliance.
Period tracker — toggle between phase names and seasons
Food tracking — photo-based, no calorie counts
Readiness Score — reflects today's capacity, not a streak
Energy Wave — mood and energy, tracked as a pattern, not a number
Prototype scope (v1): a few happy paths of an established “day 15” user, not a day-0 signup. The point is to show the payoff, tracking that has run long enough to surface patterns, not the onboarding plumbing. This is why the sign-up/onboarding screens (built by a teammate in light mode, at a different screen size) are out of scope for v1. Dark mode is the locked default throughout. Priority flows: quick daily log → readiness/energy pattern → what your body needs this week.
Energy Wave
Menstrual Phase
.png&w=828&q=75)
Winter (Menstrual)
Follicular Phase
.png&w=828&q=75)
Spring (Follicular)
Ovulation
.png&w=828&q=75)
Summer (Ovulation)
Luteal
Luteal (alt)
Workout — Cycle Insight
Talk About Feelings
Learn — Global Nav
Outcomes
Validated
The core premise, PCOS needs multi-symptom, non-judgmental management, held up across every interview and the survey. Nobody questioned why mood tracking was missing from existing tools once it was pointed out.
Not yet done
No confirmed prototype or usability testing yet, this validated the need, not the interface. Honest next step, not a finished outcome.
What I'd Do Differently
Get the interface in front of real users, not just the concept. The research is strong; the readiness score formula and the phase/season toggle haven't been tested with anyone yet.
I'd also revisit the dark-mode call properly instead of defaulting to it under deadline pressure. A teammate pushed for light mode, and the honest answer is we never validated either way. A light-mode version and a full accessibility pass (contrast, tap targets, not relying on color alone) are the clear next step, alongside reworking the period and seasonal-tracker screens with stronger data visualization, since those screens are the direct solution to the irregular-cycle shame finding and deserve to be the strongest visual in the project.
What This Taught Me
A condition that shows up differently in every person needs a tool that doesn't moralize consistency. Health isn't one number, and neither is a good day.