AI Forecasts Beat the Incumbent Forecast in 15 of 15 Customer Deployments
The Demand Forecasting Accuracy Report 2026 measures AI against the forecasts companies already run, in each company's own metric. No surveys.
New York, October 1st, 2026 - DemandForecast.ai, a service powered by Pecan AI's predictive engine, has released The Demand Forecasting Accuracy Report 2026, comparing its AI-generated forecasts to the incumbent forecasts used by companies. The report evaluated all 15 deployments conducted by DemandForecast.ai from 2022 to 2026, each against the existing forecasting methods like statistical, consensus, or manual.
The findings reveal that the AI forecast consistently outperformed the incumbent in accuracy, reducing forecast error by 14% to 56%. On average, the AI forecast showed a 32% improvement over the incumbent. Additionally, the report noted a 70% reduction in manual planning effort on average, with 75% of forecast volume in a deployment at Mars categorized as 'touchless.'
The AI-driven approach has significant implications for inventory management and sales. Better forecasts led to a reduction in overstock by up to 50% and increased sales by 10% to 25%. One example highlighted in the report is Nucor, which recovered $4 million to $5 million in annual sales at a single site through improved forecasting.
Deployments varied in scale, from a few hundred SKUs to over 100,000, with forecasting horizons ranging from 8 weeks to 36 months. Demand drivers were simplified in some cases, such as at a consumer hardware company where more than 20 assumed sales drivers were reduced to four, achieving 80% forecast accuracy across SKUs and channels. Notable customers included Kenvue, Dorman, Mars, Nucor, Nanit, and CAA Club Group.
Zohar Bronfman, Co-Founder and CEO of Pecan AI, emphasized the importance of these findings, stating, "For supply chain leaders, the useful question now is how much accuracy their current forecast is leaving on the table." The full report, including methodology, limitations, and downloadable charts, is accessible at demandforecast.ai/resources/forecast-accuracy-report.
Written by urgent.news from HackerNoon's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
