The Situation
Business Problem
Loyalty performance was steady but under-leveraged: offers weren’t tuned to neighborhood context, daypart patterns, or customer lifecycle, leaving acquisition and repeat frequency on the table. The aim was to combine first-party transactions and loyalty data with public, aggregated spend/rewards signals to identify where specific incentives resonate—and activate them through owned channels for both new and current customers.
The Challenge
Disparate data sources had to be unified in a privacy-safe way and distilled into actionable propensities that marketers and operators could actually use. In parallel, the team needed consistent eligibility logic across channels, rapid A/B testing with clean readouts, and guardrails that protect margin and minimize offer fatigue.
The Solution
We used AI to personalize rewards for new and current customers by combining first-party transactions and loyalty data with public, aggregated spend signals, then tested targeted offers to prove incremental lift.
- 01Automated ingestion and harmonizationAutomated ingestion and harmonization of transactions, loyalty events, and public spend/rewards panels with privacy-first controls.
- 02Orchestrated deliveryOrchestrated delivery across owned digital channels (app, email, SMS) with role-based dashboards for marketers and operators to iterate weekly.
- 03Creative rotationRotates copy and creative (e.g., flavor callouts vs. Points multipliers) to learn which nudge works best by lifecycle and daypart, improving redemption without deeper discounts.
