Creating a Social Phenomenon in HealthTech

BabyPeek is a consumer add-on to UNITY, BillionToOne's prenatal genetic screening test. UNITY screens for serious chromosomal conditions. BabyPeek lets expectant parents discover lighter traits from the same blood draw — eye color, hair color, cilantro aversion — before birth.

The product sits at an unusual intersection: a consumer experience built on top of a clinical test. Results are genetic probability predictions, not certainties. That distinction isn't just a legal disclaimer — it's a core design constraint. Every part of the experience had to communicate “this is our best guess” without making non-scientific users feel like they're doing statistics homework.

I led design and product management of BabyPeek end-to-end, from 0 to 1.

Before designing anything, I needed to understand who was buying BabyPeek and why. I combined quantitative and qualitative methods.

Data Analysis

I analyzed key demographic and behavioral factors influencing BabyPeek purchasing decisions — age, ethnicity, insurance type, marital status, and location — using spreadsheets and Tableau. This surfaced which user segments were more or less likely to purchase, giving us a targeting model to inform product and campaign design.

User Interviews

I interviewed 20 participants who had interacted with BabyPeek in the past week — understanding motivations, emotions, and behaviors, not just funnel drop-off.

From the research, I developed six user personas representing the diversity of expectant parents who might interact with BabyPeek. I shared these with stakeholders before any feature work began — aligning VP of Product, VP of Marketing, and Engineering Lead on who we were designing for. The personas became the foundation for every brainstorm that followed.

With research in hand, I led a collaborative brainstorming session with the cross-functional team to identify opportunities. I analyzed the existing user flow alongside conversion numbers at each step — identifying where users dropped off and what drove the most engagement. Tracking this alignment was crucial for designing a product that met user needs and achieved business goals.

We assessed each potential feature against level of effort, impact, and risk — creating an actionable roadmap that prioritized the highest-value features first without overextending the team.

Hypothesis 3 led to the most interesting design problem. We wanted to reduce the barrier to purchase, but we couldn't offer BabyPeek for free — doing so would create legal issues around inducement in a clinical context.

The solution: users could unlock BabyPeek for free by sharing it on social media. This created a viral loop — every share was both a distribution channel and organic social proof. But research showed many users would need help with the sharing step. I designed two assets:

• Sample sharing images — pre-made visuals users could post directly
• Step-by-step guide — a clear walkthrough of how to share and apply the promo code

After the initial campaign design, I ran user testing to validate the flow before launch.

Based on user and stakeholder feedback, I refined and iterated the design — simplifying the sharing flow, surfacing verification above the fold, reducing steps, and adding pre-written copy options so users had a starting point without requiring them to announce their pregnancy publicly.

Personas aren't just deliverables

Building six personas before any design work forced every idea in the brainstorm to answer for a real person. When someone proposed a feature, the first question became “which persona does this serve?” — not “does this sound good?”

Legal constraints produce better design problems

The inducement restriction that prevented us from offering BabyPeek for free turned out to be a design gift. It forced a mechanism that was simultaneously legal, viral, and user-respecting — and the social sharing unlock was a better solution than “just make it free” would have been.

Creating a social phenomenon requires owning every touchpoint

The campaign visuals, the sharing copy, the post-share confirmation — all of it had to be coherent for the phenomenon to happen. I held the thread across design, product, and lifecycle marketing to make sure it did.

Conversion data is a research tool

Tracking where users dropped off in the flow — and treating those drop-off points as design questions — shaped how we iterated. Every percentage point was a signal about something the experience wasn't communicating clearly enough.