RedCloud Validates AI Engine with 80%+ NDCG Accuracy Across 3.7M FMCG Transactions
RedCloud's AI engine R.A.I.D. achieved over 80% NDCG accuracy on 3.7 million live FMCG transactions, shifting decision-making in volatile global trade. CEO Justin Floyd declared traditional ERP systems obsolete as R.A.I.D.'s agentic infrastructure embeds real-time forecasting, replenishment, pricing and risk management, with commercialization planned later this year.
1. R.A.I.D. Performance Metrics
RedCloud’s R.A.I.D. engine processed 3.7 million completed FMCG trade transactions, delivering NDCG scores exceeding 80%, well above the threshold for commercial impact on ranking systems. Trained on real orders and volumes, R.A.I.D. offers product recommendations that reflect actual purchase intent rather than inferred behavior.
2. Shift from Systems of Record
CEO Justin Floyd argued that legacy ERP systems, built to document past transactions, fail to serve the real-time demands of fragmented global trade. He introduced the concept of agentic infrastructure, where intelligence is embedded in the transaction flow to optimise decisions continuously as market conditions evolve.
3. Commercialization and Applications
R.A.I.D. will be integrated into the RedAI platform later this year, powering demand forecasting, replenishment optimisation, stockout prevention, pricing effectiveness, credit signals and distribution route planning. The engine’s scalable architecture and confidence scoring enables deployment across new markets and datasets within RedCloud’s aggregated transaction network.