About

This project investigates the behavioral drivers of upgrade conversion in a SaaS user base of 500 users. The objective was to identify actionable engagement signals that predict upgrade readiness and translate them into financially optimized growth interventions.

Using engineered engagement features, interpretable logistic regression modeling, and SHAP-based feature attribution, I identified a clear behavioral inflection point: users reaching six or more deployments convert to Pro with near certainty. Deployment activity emerged as the dominant upgrade signal, outperforming notebooks, runs, and tenure metrics.

Rather than stopping at predictive modeling, the analysis extended into economic optimization. I designed and evaluated three growth interventions; deployment paywall, targeted Pro trial, and email re-engagement, using cost-benefit sensitivity modeling and ROI ranking.

The results show that implementing a deployment threshold paywall yields an estimated 1,795 percent ROI with a 19-day payback period and a projected annual revenue uplift of $312,000 under mid-case assumptions.

This project demonstrates how behavioral analytics can move beyond prediction to drive capital-efficient product and growth strategy decisions.