Free email sprint · 3 days
Draft the study that gets your AI device paid — in 3 prompts, over 3 days.
FDA clears your device. A clinical-utility study is what makes a payer cover it — and it's the gate most cleared AI devices never planned for. This sprint walks your team through drafting your own, one prompt a day.
- Three short emails, one prompt each — run in your own AI platform, against your own device.
- It surfaces your evidence gap in your own words. It doesn't hand you an answer.
- Written by Ross Prior (Saolyn), from The State of AI in Medical Devices 2026 with Kelly Coverdale (Cover Biomedical).
Start the sprint
We'll email you a link to confirm. Part one arrives the moment you click it.
Confirming starts the sprint and adds you to Saolyn MedTech Signals — occasional email about new reports and analysis. Unsubscribe any time, one click. See our Privacy Policy.
What the three days cover
- 1
Name the decision, and the claim
Accuracy is a property of the model. Clinical utility is a property of the decision the output changes — and payers reimburse the second one. Day one is one sentence, and it's harder than it sounds.
- 2
Design the comparison
A one-page study skeleton: the comparator, a management-change endpoint, and an outcome endpoint that isn't a diagnostic-performance metric.
- 3
Stress-test it against a payer
Hand the draft to a hostile reviewer before a real one gets it. What comes back is your clinical-utility gap, in your own words.