Formation BioMuse
Muse is an AI-driven platform that simplifies the creation of tailored recruitment strategies and marketing materials. Users can interact with the AI assistant to generate content or manually adjust it based on their expertise. Muse also produces mock-ups of all recruitment materials for a comprehensive, streamlined process.
TeamDesign Lead (me)
Product Manager
Engineers (5)
Timeline12 Weeks*
(but also kind of a year)
ContextThe state of clinical trial recruitment today
Recruitment is often the single biggest bottleneck in clinical trials. Teams rely on a patchwork of tools, agencies, and manual processes to identify, engage, and qualify participants. This creates inefficiency, slows down timelines, and drives up costs.
Fragmented workflows
Different stakeholders make up most of this workflow; marketing teams, recruiters, nurses, and compliance officers, all operating in silos. Each group edits content independently with differing priorities, often duplicating effort.
One-size-fits-all strategies
Recruitment content and strategies are frequently generic, failing to account for differences across demographics, disease profiles, cultural contexts or unmet needs. This limits reach, especially when diversity and inclusion are critical for trial success.
What if recruitment teams could have a first draft of the recruitment strategy and content the same day they receive the protocol, skipping the bottlenecks and focusing their time on reaching patients instead?
Design Approach1
Starting with a Vision State
Instead of solving piecemeal pain points, I started with a high-level “vision” of what recruitment could look like if the bottlenecks disappeared. This vision was captured through loose wireframes and sketches that showed possibilities, not polished solutions. I chose to deliberately keep the fidelity low to encourage feedback. Putting early sketches in front of potential users, PMs, and stakeholders allowed them to react quickly surfacing what felt exciting, what felt unrealistic, and where the biggest opportunities were.
2
Learning Through Iteration
Feedback helped narrow the problem: the critical bottleneck was getting to a first draft. Each iteration got sharper, moving from “blue-sky vision” → to “plausible workflows” → to “testable prototypes.” We were able to build something within 6 weeks and start testing with real users and getting feedback. It wasn’t the best yet but got the ball rolling.
3
Balancing Immediate and Future Needs
We had to look beyond fixing today’s bottlenecks. For users, a “strategy” often meant a simple direction on paper. With AI, we reimagined it as comprehensive, adaptive, and actionable.
By designing for future challenges, iteration, compliance, personalization, Muse became more than a tool for speed. It surprised users with unexpected value, and that vision-first approach is now at the heart of why they love it.