Throne
Science
Introduction
Throne Science is a health tech startup building a smart toilet sensor that tracks gut health, hydration, and urinary function. The scope covered homepage, blog, purchase page, paid media ads, founder's story, product photography, and 3D motion imagery. I led homepage design, blog design, and the majority of Webflow development — owning the Spline 3D integration, Lottie animations, and WebM optimization,, while partnering with a developer on e-commerce and a video timeframe implementation.
Challenge
Selling a smart toilet sensor requires earning trust fast. The strategy was motion-led storytelling: a 3D product hero, scroll-triggered setup video, and lightweight animations stitched together inside Webflow without losing performance. The technical bar was as high as the narrative one.
Role
Lead Product Designer
Narrative-first homepage
The homepage was designed as a guided story rather than a feature list — leading with the product, building credibility through physician endorsements, and walking users through how Throne fits into their daily life before asking them to buy. Every blade had a job: introduce, validate, demonstrate, and direct. Users who wanted to go deeper were routed to a health-focused blog covering urinary health, stool health, and how Throne's data could be used — keeping the homepage clean without sacrificing depth.
Product-forward storytelling through motion
The hero blade centered a Spline-rendered 3D rotation of the device — putting the product itself front and center from the first scroll. A scroll-triggered video guided users through three key moments: how to install Throne, how it pairs via Bluetooth, and how health data appears in the app. The goal was to make an unfamiliar product feel intuitive and approachable at a glance.
Performance-first technical execution
Bringing this narrative to life inside Webflow required solving real technical constraints. A looping WebM in the footer needed to be small in file size, smooth on playback, and within Webflow's upload limits — so I used Terminal to iteratively remove frames and reduce quality until it hit the threshold without losing visual fidelity. That same rigor applied across Spline, Lottie, and WebM integrations throughout the site.
