Tiered Health Actions: Defining What Meaningful Action Looks Like

Before the framework, every action in the product carried equal weight toward the North Star Metric. A walking challenge could count the same as a clinical trial enrollment — even though one is general engagement and the other is a partner-reportable action tied directly to contract value.

That created a measurement problem, not a reporting inconvenience. The product was generating plenty of activity, but the business could not prove whether that engagement was converting into the actions partners pay to drive. The people served by fixing it were members — who deserve to be guided toward something that matters, not everything at equal volume — and the teams accountable for proving the product works.

Health Actions discovery included approximately 50 survey responses collected by Laquita. Because the reviewed material does not preserve the findings or connect them to specific decisions, I treat that work as research context rather than claim that it shaped the tier definitions or numerical weights.

The supported rationale was strategic and operational: partner-reportable outcomes, the steps toward those outcomes, and foundational engagement are different kinds of value and should not be treated as one flat activity count.

Tier 1 — Partner-reportable health actions. The highest-value actions clients pay to drive: clinical trial enrollment, prescreener completion, blood work completion, health screening completion, vaccination received, specialist appointment attended.

Tier 2 — Mobilization steps toward Tier 1. The bridge between engagement and contracted action: setting a health goal, scheduling an appointment, registering for a screening event, completing a health risk assessment, receiving an ambassador warm handoff.

Tier 3 — Engagement baseline. Foundational behaviors that keep members active and feed the matching engine: quiz completion, article reading, walking challenges, community posts, login streaks, push notification engagement.

I defined tier classification as a measurement requirement and used the framework in roadmap prioritization. By July 28, 2026, it was used by senior leadership and reflected in the P1/P2 schema for planning and reporting; that is adoption of the framework, not proof of complete production instrumentation.

Weighting the metric meant the headline number would look worse than the flat count everyone was used to, and Tier 3 work — the engagement features people enjoyed building — would now have to justify itself as a feeder into Tier 2 and Tier 1 rather than as an end in itself.

I recommended taking that hit. A metric that rewards volume tells you to build more volume. The consequence is a roadmap where engagement work still ships, but has to answer what it moves members toward.

The organization got one taxonomy for planning and prioritization, separating partner-reportable outcomes from the steps and engagement that may lead toward them. Senior leadership used it, and it appeared in the P1/P2 schema.

The implementation limit matters: an August 18 database review found no instrumented Health Actions prompt, intent, step, or completion events in the database reviewed. I therefore do not claim a complete measured tier funnel, verified T3 → T2 → T1 movement, or an attributable partner outcome.

The learning transfers directly to AI product work. Deciding what counts as a meaningful outcome — before you build the thing that is supposed to produce it — is the same discipline that tells you whether an AI feature is actually working or just producing output.

The framework gave leadership a clearer way to plan around partner-reportable actions instead of treating every activity as equal value.

Its business contribution is a strategy and measurement design that connects mission to the business model. Production measurement and attributable partner value remain work to prove, not outcomes I claim here.