Lead, Product Analytics and Experimentation at Apree Health. Eight years of designing the experiment, building the measurement system, and then watching the decision change.
In one paragraph
Samuel Martin is a product analytics and experimentation lead who designs the measurement that decisions rest on. His outbound randomized controlled trial showed a 3.45x lift in appointment booking conversion at $99 cost per acquisition against $549 in first-year customer value, which turned an operating expense into a revenue product at roughly 80% margin and unlocked a $1.1M scaling roadmap. He has led analytics across 750+ enterprise clients and 5.5M users, and built the unified attribution warehouse that cut financial reporting latency from weeks to minutes.
These are not four different competencies. Samuel Martin designed the experiment, built the measurement, and the decision moved. Every number is stated here, so nothing on this page depends on a click.
Cost per acquisition landed at $99 against $549 of first-year customer value. Outbound engagement stopped being an operating expense and became a revenue product at roughly 80% margin. The $1.1M scaling roadmap that followed was sensitivity tested before it went to the executive team, and it won their buy-in.
Read the case studyFour source systems, one warehouse, one definition of a conversion. Finance had been waiting weeks on a reconciliation that was assembled by hand. Now the number arrives in minutes, and it is the same number marketing and finance both cite.
Read the case studySamuel Martin led the analytics function at that scale, which sets the real constraint on this work. A measurement has to hold across 750+ clients and 5.5M users, and it has to survive being read by people who did not build it.
See the full body of workA 30-day A/B test reads the leading indicators. A 9-month difference-in-differences study reads the outcome. Running both is how you avoid shipping the metric that looks excellent at 30 days and terrible at nine months.
Read the case studyThree more case studies sit alongside these, with two essays and two shipped tools. The shortest of the three is the callback events that were not there, which is the one about a KPI that was quietly, confidently wrong.
Each surface re-sequences the same case studies and swaps the framing sentence. Nothing is invented per surface, and none of the four is the real one.
San Francisco Bay Area. Lead, Product Analytics and Experimentation at Apree Health. Samuel Martin is looking for work where the measurement is the hard part and someone has to own it.
Most of this work sits behind protected health information. Figures on this site are redrawn on synthetic data, no client is named, and the method and reasoning are exact.