10 discovery interviews with women aged 28–44 navigating a new PCOS or hormonal imbalance diagnosis
Halo Health's care platform connects women with hormonal health specialists for diagnosis and treatment. The team observed high drop-off between initial diagnosis and treatment start, and a pattern of cancellations in weeks 3–5. This research explored the user experience of receiving a new diagnosis and what information and emotional needs shape whether someone moves forward with treatment.
Participants were recruited via panel — all had received a PCOS, thyroid, or hormonal imbalance diagnosis within the previous 6 months, and had used at least one digital health platform during that process.
Four themes emerged with high consistency across participants. Each one points to a place where the product's communication assumptions diverge from users' actual emotional and informational state.
Most participants described receiving a diagnosis as a moment of high anxiety followed by a flood of unanswered questions. The platform delivers the clinical fact but not the emotional scaffolding or personalized context that makes a diagnosis livable.
A diagnosis isn't an endpoint — it's the beginning of a user's need to understand. The moment of diagnosis delivery requires more than a label; it requires a framework for what this means for this person's specific situation.
The product currently treats diagnosis as a clinical event. Users experience it as a life event. That gap is where trust is lost or built.
Participants ranged from first-time patients to self-educated health advocates. The platform offers a single onboarding path that satisfies neither. Over-explanation frustrates knowledgeable users; under-explanation alienates new ones.
The intake questionnaire collects symptom history but not knowledge level. A single question — "How familiar are you with [condition]?" — could unlock a branching content experience that meets users where they are.
This isn't about dumbing things down for some and elevating for others. It's about establishing a starting point so the explanation feels like it was written for this user.
Many participants had been dismissed or delayed by prior healthcare experiences. The need to feel believed was a prerequisite for trusting the diagnosis itself. This is a relational need, not an informational one — and the current platform experience is purely informational.
The intake and diagnosis flow could explicitly acknowledge this dynamic. Language that validates the user's journey — "We know many women wait years for answers. We want to change that." — does more for trust than clinical accuracy alone.
This is a copy and tone problem more than a feature problem. The platform has the clinical credibility. It needs the relational credibility to match.
Participants consistently distinguished between receiving a treatment instruction and understanding a treatment journey. They wanted to know what would happen at week 4, what a good outcome looked like, and what to do if something felt wrong. Without this, starting treatment felt like a leap of faith rather than an informed decision.
Post-diagnosis, the product should deliver a personalized "treatment roadmap" — a simple visual timeline that shows: what to expect in weeks 1–4, what indicates progress, what indicates a problem, and when to check back in.
This is the single most consistently cited missing piece. It also directly addresses the week 3–5 cancellation pattern — users who have a roadmap know that feeling uncertain at week 3 is normal, not a signal to quit.
A mental model gap is where users' assumptions about how something works diverges from how it actually works. These gaps are where trust collapses and comprehension fails.
What users expect vs. what they currently receive
These findings point to three types of work — two of which can start immediately without engineering resources:
Copy & tone sprint (this month): Address validation language (Opportunity 4) and the "why this treatment" explanation (Finding from Theme 1). These are writing tasks. A focused sprint with a PM and writer could ship in two weeks.
Concept test (next 4–6 weeks): Prototype the treatment roadmap (Opportunity 3) and test it with 5 users from this same profile. We already know the need — validation testing will tell us whether the execution lands.
Product design sprint (Q2): Build the knowledge calibration question and adaptive content path (Opportunity 2). This requires design and light engineering, but it's the highest-leverage structural change in this set.
I'm available to run the concept test on the treatment roadmap and to support the copy sprint with a review pass. Both can be scoped as a follow-on engagement.
Legible Research is a specialist UX research practice for consumer health and beauty products. Questions about this report or next steps: hello@legibleresearch.com