Modeling, Evaluation, and Opportunity Discovery

This unit explains how machine learning models learn, how product managers match AI capabilities to real problems, and how to evaluate model performance in context. Learners will work through a churn prediction case study, practice day-to-day AI PM activities, and identify AI opportunities along the user journey using a practical toolkit.

Learning Objectives

• Analyze how supervised learning algorithms (illustrated by k-Nearest Neighbors) form predictions and identify the implications for data needs and feature design
• Apply AI/ML capability-matching to map specific product problems (such as churn) to appropriate solution types, constraints, and required data sources
• Demonstrate an end-to-end AI product workflow by outlining data preparation, model selection, training, evaluation, and deployment steps for a subscription churn prediction example
• Evaluate and prioritize AI opportunities across a user journey using the AI-Experiences PM Toolkit and explicit feasibility, impact, and skill-gap criteria