NOWINCLUDED: Product Direction, Measurement, and Delivery

NOWINCLUDED serves Black community members who are deciding whether to take a health action — a screening, a trial prescreener, a conversation with a provider — and the internal teams and partners who are trying to support that decision. The team needed to know whether the product was simply attracting users or actually helping members act.

There were no formal KPIs. Without a measurement model, every roadmap conversation risked becoming an opinion contest, and there was no way to tell a feature that drove behavior from a feature that only drove logins.

The original decision context was operational: there were no formal KPIs, no shared release history, and no dependable view of the path from onboarding to meaningful participation. That made the first priorities a decision log, a North Star, and clearer instrumentation — not another feature chosen on instinct.

Health Actions discovery also included approximately 50 survey responses collected by Laquita. The reviewed material does not preserve the findings or connect them to specific roadmap decisions, so I treat the survey as research context rather than evidence for a claim.

Later evidence helped assess the direction rather than originate it. An August 18, 2026 analysis found 54.5% trailing-90-day activity among members who engaged with Health Stories versus 6.6% among the comparison group. That association supports the rationale for story engagement; it does not prove stories caused retention or represent a prospective experiment.

The same retrospective analysis found a substantial gap between onboarding and meaningful participation, but its activation denominator and stated percentage-point gap were inconsistent. I use the qualitative finding, not the unreconciled calculation.

Defined the product's North Star — members taking a measurable health action (long-term ambition: 1M) — and the operating model beneath it: a Supply → Match → Mobilize loop and T1/T2/T3 health action tiers, so every feature has a "why this, why now" test.

Set the key activation metric — % of active users who engage with or share a health story — and wrote the instrumentation requirements with data and engineering; it runs live in production dashboards alongside story creation rate and a retention funnel.

The evidence report also showed profile-name completeness at approximately 53–56% for December 2025–January 2026 signup cohorts and approximately 99–100% from February 2026 onward. That supports the value of profile-flow improvements for CRM matching and reporting, without assigning the change to a precisely dated release until that date is confirmed.

Wrote acceptance criteria with engineering and ran a seven-phase QA process and a Go/No-Go gate across web, iOS, and Android, with our engineering team, an external development partner, and dedicated mobile testers. The delivery system is written up in The Release Engine (/work/release-engine).

Scoped the roadmap into a year of dated releases with design and engineering — the leaderboard and wellness launch (July 2025), the Story-Centered Feed (September 2025), then the Walking Challenge, Global Search, and the in-app calling stack (2026).

Created the Product Decision & Feature Log: the company's source of truth for every shipped release and the reasoning behind it.

Delivery also had measurable limits: client-side instrumentation could not support detailed feature funnels, mobile/web attribution was unreliable, authentication and verification needed reliability work, HubSpot configuration and production bugs blocked Inner Circle cohort testing, and Health Actions pilot planning still had unresolved SSO and integration ownership questions. These constraints belong to their specific workstreams; I do not use them to explain the story-upload decision without evidence.

Stories were the strategy, but the upload path carried a file-size limit, and phone-recorded video is exactly the kind of story we were asking members to send. The friction sat on the one behavior the roadmap depended on.

The alternatives: keep the constraint and coach members to compress their videos, keep pushing feature work while leaving that friction in place, or stop and rebuild the upload path.

My call was to pull direct upload forward in the roadmap ahead of new engagement features — it shipped in February 2026 and was intended to simplify story submission. The tradeoff was real: a release window spent on plumbing rather than something visible. I do not have a verified pre-launch upload drop-off rate, so I do not claim that the change reduced abandonment.

Engagement rate grew from 11% to 32% in Q3 2025. User posts drove 51% of all likes, supporting the community-content strategy. Mobilization is instrumented rather than asserted — a year earlier that instrumentation did not exist.

The honest limit: these are correlated outcomes across a year in which marketing, community, events, and business development were all driving demand. I can prove what product shipped and when, and that the curve moved with it; I cannot isolate product's share of it. The attribution work is in Year One as Product Lead (/work/year-one).

What I learned is that measurement is a prerequisite, not a reporting chore. Once the team could name the behavior a feature was meant to change, the roadmap arguments got shorter and better.

The Product Decision & Feature Log became the company's source of truth for what shipped and why, so decisions stopped living in people's heads.

The measurable-health-action model reframed the market conversation: it maps onto HEDIS measures, CMS Star Ratings, and SDOH goals, which supported a payor vertical thesis alongside the existing business.

Leading through an external development partner and a small internal team meant my influence ran through clarity rather than headcount — written requirements, a visible gate, and a shared definition of done.