Senior Product Manager — Predictive Growth (pLTV)

As Senior Product Manager for Predictive Growth at Flo, you'll own Flo's predicted Lifetime Value (pLTV) platform across iOS, Android, and Web. pLTV is the shared signal behind UA bidding, budget allocation, creative-testing prioritisation, finance forecasting, and roadmap decisions across the organisation — making it one of the highest-leverage PM roles in growth.

You'll work with a team of data scientists, machine learning engineers, and data engineers running Flo's production pLTV models across short-horizon (intra-day) and long-horizon (M15+) predictions. You'll drive accuracy improvements, expand signal coverage across platforms, and extend pLTV's reach into new decision domains. You'll also help scale Flo's in-house agentic experimentation system — already delivering 30%+ quality gains — and contribute to the next generation of Flo's marketing measurement stack, including Marketing Mix Modeling and incrementality testing.

This is a senior IC role with company-wide visibility, partnering with leadership across User Acquisition, Finance, Analytics, and Engineering. It's equally hands-on in delivery: co-leading the team day-to-day with your engineering manager pair, writing the specs, defining the epics, staying close to the work.

Your Experience

Must have:

• Proven experience owning machine-learning or data-driven products in production — predictive scoring, propensity, ranking, forecasting, or similar data-rich products. Strong fit for current Product Managers; also open to Engineering Managers, ML Engineers, or tech leads who have shaped product direction end-to-end and are ready to make the switch into a PM seat.
• Hands-on delivery instincts — equally comfortable in Jira (writing specs, defining epics, refining the sprint backlog, unblocking the team) as in strategic planning and stakeholder management. Co-leads the team day-to-day with your engineering manager pair.
• Hands-on partnership with data science and machine-learning engineering teams — partnering on model evaluation, retraining cadence, feature decisions, and release discipline.
• Strong analytical fluency — SQL proficiency, comfort interpreting model evaluation metrics, and the habit of interrogating performance at segment, cohort, and campaign level rather than blended aggregates.
• Demonstrated ability to influence senior cross-functional stakeholders across marketing, finance, analytics, and engineering — translating technical model behaviour into commercial decisions.
• Comfortable in fast-paced, ambiguous environments — balancing platform investments against fast-cycle experimentation, with shifting priorities and competing demands.

Nice to have:

• Working knowledge of mobile and web user acquisition — how bidding, attribution, and campaign optimisation actually work, including privacy frameworks such as SKAN/ATT or Privacy Sandbox.
• Hands-on experience with mobile attribution platforms (AppsFlyer, Adjust, Branch, Firebase) and modern measurement frameworks (SKAdNetwork, GA4).
• Exposure to Marketing Mix Modeling, incrementality testing, holdout design, or causal inference techniques.
• Familiarity with feature stores (Databricks, Tecton, Feast) and modern MLOps tooling.
• Subscription-app, health & wellness, or femtech domain background.
• Exposure to agentic AI or LLM-orchestrated workflows for ML experimentation, research, or model evaluation.

What you'll be doing

You'll be responsible for:

• Owning Flo's pLTV product roadmap end-to-end across iOS, Android, and Web — driving accuracy improvements, expanding feature coverage, extending prediction horizons, and pushing pLTV into new decision domains (creative testing, product roadmap, pricing, hypothesis validation) alongside its existing UA and Finance use cases.
• Partnering with data scientists and engineers on feature strategy — deciding which inputs the model consumes and ensuring those features are available reliably in production.
• Operating rigorous release discipline — gated rollouts, shadow-mode evaluation, drift detection, automated retraining, and post-launch calibration — so production models stay accurate as data distributions and platform policies evolve.
• Working closely with data scientists and analysts to investigate model quality issues — country-level miscalibration, cohort drift, signal anomalies — and feeding findings back into the next iteration.
• Helping scale Flo's agentic ML experimentation system from hundreds toward thousands of model runs — compounding quality gains back into production.
• Acting as the primary product partner to User Acquisition leadership — translating model behaviour into bidding decisions, framing trade-offs, and aligning on attribution-window strategy.
• Reporting model performance and strategic recommendations into monthly leadership reviews and quarterly investment decisions.
• Contributing to Flo's next-generation measurement stack — including Marketing Mix Modeling, incrementality testing, and causal lift methodology — so marketing investment decisions are grounded in causal, not correlational, signal.

#LI-Hybrid #LI-AJ1