Back to work
Supertails2024Framework

Food Replenishment Framework

A customer level timing model that estimates when pet food is likely to run out and when to trigger the next reorder reminder.

ReplenishmentCustomer timingRetention
Business problem

Reorder reminders were generic and often mistimed. Customers could run out and buy elsewhere, or receive a reminder too early to be useful.

Key decision

Make timing the core decision. Estimate each customer’s likely run out window, then decide whether to remind now instead of using one fixed schedule for everyone.

What changed

Replenishment moved from generic reminders to customer level timing using purchase history, consumption logic and live intent signals.

Takeaway

Replenishment performance depends on timing as much as targeting.

Want the implementation details?

Open the full architecture, modeling, validation and lessons.