Replenishment models built on reconciled inventory data, evaluated against the promotional calendar, and owned by the merchandising team rather than a vendor.
The challenge
A national grocer’s forecasting relied on spreadsheets and experience at store level and on a black-box vendor model at head office, and neither informed the other. Promotions and weather turned both into guesswork at precisely the moments when accuracy mattered most.
Our approach
We spent the first weeks on the data rather than the model, reconciling inventory positions across systems until the numbers matched the shelves. The models followed, evaluated store by store against the previous year’s promotional calendar, with the merchandising team able to see and override every forecast.
What we built
- Inventory positions reconciled across ERP, point-of-sale, and warehouse systems
- Forecasting models with promotion and seasonality features
- Store-level forecasts with explanations and override capture
- Backtesting against historical promotional periods
- Hand-over to an in-house data team, with training
What changed
- Merchandisers trust the forecast because they can see the reasoning behind it
- Model changes are backtested before anyone acts on them
- The capability now belongs to the retailer rather than to a vendor
This engagement is representative of the work Loonwise undertakes; client names and specifics are shared in conversation.
Contact
From ambition to working software.
Whether you are accelerating AI adoption, modernizing a critical system, or building the capabilities to innovate faster, we bring the consultants who scope the work and the engineers who build it—the same people, from the first conversation. A senior engineer responds within one business day.
Get in touch