The Weather Company’s Early Bet on Agentic AI for Customer Engagement

Session Summary

In this session from MoEngage’s Agentic Marketing Tour, Hope Barrett, Sr. Director of Marketing Media and Technology at The Weather Company, shares what getting started with agentic AI actually looks like in practice. Rather than showcasing an already completed system, Hope walks through the realities of rebuilding a data foundation with a lean team, navigating customer pushback after early automated content tests, and overcoming operational bottlenecks. She explains why successful agentic adoption begins with clean data and foundational predictive models, while outlining a vision for customer engagement that combines first-party behavioral insights with dynamic, real-time context.

Top Takeaways

• Starting Before the Architecture is Complete: Launching an agentic AI strategy does not require a fully finished tech stack; early momentum comes from testing targeted use cases while actively rebuilding data foundations.

• Treating Customer Feedback as a Guardrail: Early pushback on AI-generated customer touchpoints delivers essential guidance for adjusting tone, managing expectations, and defining where human review must stay intact.

• Solving for Workflow Integration First: The primary obstacle to scaling agentic AI is rarely the technical capability of the model itself, but how smoothly it embeds into existing day-to-day marketing workflows.

• Prioritizing Foundational Predictive Models: Marketing teams should master clean data and high-impact predictive models, such as churn likelihood and RFM scoring, before deploying autonomous multi-agent systems.

• Pairing Behavioral Data with Environmental Context: Breakthrough personalization happens when brands connect first-party customer data with dynamic, real-time external conditions to deliver relevant, in-the-moment messaging.