JapanRetail and e-commerce2024 – ongoingBelc
AI sales forecasting for new store sites
Supermarket chain Belc forecast new-store sales with a median error in the 7% range using SENSY's site-selection AI, SENSY says
Problem
Belc, a supermarket chain, needed to judge candidate store sites and estimate their sales before opening.
Approach
Belc began trialing SENSY GeoScope in autumn 2024. The model combines Belc's own POS data with trade-area, population and competitor data. A user marks a site on a map and the AI predicts sales. After the trial, Belc used it in full operation for initial site evaluation, trade-area and competitor analysis, and sales forecasting.
Results
Takeaway
Belc trained the forecast on its own sales data, and the article reports only an error figure, with no before-and-after comparison.
Categories
- Industry
- Retail and e-commerce
- Business function
- Strategy and organizationProduction and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Vendor solution
- Evidence
- Vendor case study
Sources
Source summaries
The article describes how Belc uses SENSY GeoScope, which forecasts sales for candidate sites from Belc's POS data plus trade-area, population and competitor data. It reports trials from autumn 2024, full operation since, and a median forecast error in the 7% range on the verification targets. It gives no baseline, and the headline's claim of zero opening mistakes has no support in the body text.
Full details are in the original sources.
