Coca-Cola uses AI to improve retailer ordering in Malaysia

Details

Coca-Cola has added an AI recommendation layer called Perfect Basket to its Coke Buddy self-ordering platform in Malaysia. The feature uses Coca-Cola's Central Recommendation Engine to suggest which products retailers should order and in what quantities, drawing on multiple data signals before a retailer submits a purchase. Retailers access Coke Buddy via app, website, or WhatsApp, and retain full override control over what they buy. Perfect Basket layers in signals beyond simple reorder history: seasonality, weather, ordering frequency, and purchasing patterns from comparable businesses. This moves the system beyond basic replenishment nudges toward a more dynamic demand-sensing tool — though the underlying approach is applied recommendation tech, not novel research. Coca-Cola reported results from a January–April 2026 campaign: 83% of participating outlets adopted the recommendations, and those retailers recorded higher sales revenue growth than comparable non-participant outlets. However, the company did not disclose the magnitude of the revenue difference, nor figures for forecast accuracy, stock availability, or logistics costs. The 83% figure also applies only to campaign entrants — coverage across the full 39,000-outlet network was not reported. The deployment fits a broader pattern of FMCG companies using AI to tighten distributor and retailer relationships through digital ordering tools. Coca-Cola noted similar capabilities exist in other markets, though this release focused exclusively on Malaysia.