Tl,dr: This case study shows how Tannico reduced forecast error by 36%, using AI demand planning and inventory forecast software to enable more demand forecast accuracy and confident purchasing decisions across 10,000 SKUs.
Tannico, the world’s largest Italian online wine shop, was founded in 2012 by Marco Magnocavallo. Known for shipping over 1.5 million bottles annually across Italy and 19 countries, Tannico built its reputation on vast selection, unmatched delivery speed, and exceptional customer service.
By 2018, the company was managing roughly 10,000 SKUs. Seasonal fluctuations, promotional campaigns, and the complexity of managing vintages across multiple labels made accurate forecasting a constant challenge.
Co-Founder Cristiano Pellegrino realized that even small forecasting errors could disrupt stock allocation, delay shipments, and reduce customer satisfaction. Tannico needed a predictive system that could capture complex demand patterns and reduce uncertainty in planning.
The Forecasting Challenge
Tannico’s planning team faced:
- High forecast errors across a large, seasonal catalog
- Difficulty accounting for promotions, holidays, and vintage-specific demand
- Inconsistent insights leading to reactive purchasing decisions
- Risk of stockouts or overstock on high-demand SKUs
The company needed a solution that would deliver accurate, actionable forecasts across thousands of SKUs, allowing their internal buyers to make confident, proactive purchasing decisions.
The AI-Powered Solution
Tannico adopted Intuendi’s AI-powered demand planning platform, combining advanced demand forecasting, top-down planning, and data visualization tailored to winery products. The platform:
- Selected the best predictive model for each SKU automatically
- Incorporated causal factors such as promotions, seasonal peaks, and holidays
- Enabled buyers to determine optimal stock quantities for each SKU
- Supported daily decision-making to maintain service levels while controlling inventory
“Our buyers use Intuendi every day to estimate inventory needs, so we know what and how much to buy to keep our standards high and make customers happy.”
— Cristiano Pellegrino, Co-Founder
The Results: Measurable Impact on Forecast Accuracy
- 36% reduction in mean forecast error in the first months of usage (vs previous methods) from improved demand forecasting accuracy
- 62% reduction in absolute forecast error, with best-case MAE of 3 units over four weeks
- Enabled the company to plan confidently across thousands of SKUs, improving both service levels and operational efficiency
The Advantage Gained
Through the strategic vision of Tannico’s leadership, the company turned forecasting from a reactive task into a predictive, reliable engine for purchasing and stock management. Accurate forecasts allowed Tannico to reduce errors, better meet customer demand, and scale operations sustainably without compromising speed or service.
