80%
of women interviewed had never brought about a full lifestyle change despite knowing it was medically recommended
Drawn from primary interviews conducted alongside secondary research citing similar adherence gaps in PCOS management literature.
HEALTHCARE · SERVICE DESIGN · 2021
Ariete
Individual service design research project reimagining PCOS management as a shared responsibility, not a solitary one.
Objective
What this project set out to do
To understand how young women (20–25) manage PCOS on a day to day basis, and to design a system that makes that management a shared responsibility between the woman, her partner, and her close support network rather than it being something she carries alone.
Challenges
What made this hard
PCOS sits at an uncomfortable intersection: it's under-diagnosed, poorly understood even by those who have it, and rarely discussed openly, which made both the research and the design genuinely difficult.
Approach
How the work was structured
The project ran in three phases over four months: Phase 1 (Sept) - primary and secondary research, persona mapping, and opportunity framing. Phase 2 (Oct–Nov) - ecosystem mapping, co-creation workshops, and three separate ideation sprints as the concept evolved. Phase 3 (Dec) - prototyping, user testing, and documentation. Two concepts were built and discarded before the final direction emerged with each discarded idea directly shaping the next and leading to the final outcome.
Methodology
Methods, and why each one
Secondary research established the clinical baseline; primary research supplied the lived experience that no paper or article could. Given the sensitivity of the topic, methods were adapted per participant: voice notes, text responses, and one supported video call rather than a single standardised format. A facilitated focus group (run by a peer) was used specifically to reduce the researcher's presence as a barrier to honesty.
Solutions & Findings
What I found, and what I built from it
Research kept circling back to one pattern: women were not short on willpower or information. They were short on people who understood what they were managing, and a system to make that support concrete.
Recommendations
What was recommended, and what was built
The first concept, a subscription-based diet and meal-delivery model was proposed, tested against clinical literature, and was dropped because different women have differing symptoms when it comes to PCOS. No dietitian consulted could support a single "optimal" PCOS diet, so building the product around one would have misled users. The direction that replaced it made partner involvement the core mechanic rather than a feature: syncing activities, shared journaling, and visibility into the woman's day made the whole ordeal a shared journey.
Outcomes
Where it landed
Ariete exists as a validated concept and tested hi-fidelity prototype, not a shipped product. The direction was strong enough to catch the eyes of entrepreneurs at the MIT Hacking Racism hackathon, and I intended to follow up with a hackathon mentor to explore bringing it further but nothing came out of that; the project's real output was a rigorously tested service concept and a clear articulation of what a shared-responsibility approach to PCOS care could look like.
Reflections
What I would do differently
I'm aware the project centres a fairly narrow demographic (urban, English-speaking, 20–25) and I'd want to test the shared-responsibility model against more conservative family structures, where the idea of a partner or father tracking a woman's cycle data could land very differently. The hardest part wasn't the interface, it was getting people to talk honestly about something they'd never said out loud before.
Learnings
What transfers to the next project
Co-creation with the actual affected group beats generalist brainstorming, even with skilled participants. Two ideation exercises with designers who didn't have PCOS produced far weaker ideas than one session with a participant who did.
A concept can be well-researched and still be wrong. The subscription/diet model was backed by real interviews, yet it collapsed the moment it met clinical literature. Validating an idea against research you did isn't the same as validating it against research you didn't do.
Designing for a taboo topic meant reassessing how I went about my research. I had a real sense of curiosity and openness to learn and support. The project has to always fit the participant's comfort and needs and not a researcher's convenience.
Next