Hershey revamps marketing mix modeling with agentic AI
Integrate Mutinex and Tracer to enable monthly MMM and cut analysis time to weeks.
Set up a data pipeline with Tracer and Mutinex to automate MMM and enable monthly media spend decisions.
Summary
Hershey is partnering with analytics platforms Mutinex and Tracer to automate its marketing mix modeling (MMM), moving from a slow, quarterly cycle to a monthly, real‑time process. Mutinex, powered by Claude and Gemini, deploys a multi‑agent system where each agent specializes in econometrics, pricing theory, or model diagnostics. Tracer acts as the plumbing layer, cleaning and standardizing fragmented data across marketing and retail systems so that Mutinex’s models can run faster and more reliably. The new setup reduces model run time from several months to as little as three weeks, enabling Hershey to make media and trade spend decisions on a monthly basis. The company expects a 4–5% lift in revenue attributable to media, and plans to scale the system to cover its entire brand portfolio with up to 12 analyses per year. This shift is part of a broader industry trend to use agentic AI for faster, data‑driven marketing decisions.
The collaboration demonstrates how AI can streamline complex, data‑heavy processes, turning historically lagging insights into actionable, near‑real‑time guidance. By integrating Tracer’s data preparation with Mutinex’s AI modeling, Hershey can iterate on spend allocation more frequently, potentially improving ROI and responsiveness to market changes.
Key changes
- Hershey partners with Mutinex and Tracer for automated MMM.
- Mutinex uses Claude and Gemini to power a multi‑agent system.
- Tracer cleans and standardizes fragmented marketing data.
- Models now run in as little as three weeks instead of months.
- Monthly decisions replace quarterly cycles.
- Expected 4–5% lift in revenue attributable to media.
- Up to 12 monthly analyses per year.
- Agentic AI assigns domain specialists for econometrics, pricing, and failure diagnosis.