Retail
Home goods, grocery, and multi-unit retail are defined by the asset they can't see: the lane, the promo, and the spread that tightens when a promotion goes sideways.
Retail problems are mostly assortment and load. Category teams plan on aggregate POS, and the strength of those plans only lands when the truck reaches the store and the allocation is already decided. A forecast that has to price every week must survive a promo, a seasonal swing, and a new store that gives no history.
The constraint is the same across the whole retail panel: the number the buyer cares about is the aisle and the store, not the national average. A model that hides a broken store inside a regional forecast has not done the work yet.
What you get
The promise, kept honest. Direction, not invented numbers.
- 01
Store-level forecasting that stays accurate when the promo is on and off
- 02
An inventory and allocation view that pulls shrink and in-store count into a single ledger
- 03
Assortment reads that show the buyer what the store sold — before the promo day
The constraints
The parts that make this industry different. If one of these sounds like your problem, that is the conversation to have.
- A multi-thousand-SKU forecast has to survive promos, holiday, and a store that does not resemble the model
- Inventory data is honestly stale — the count, the missing order, and shrink all live outside the system
- Warehouse plan and store plan both decide stock — the machine has to respect two masters, and it is wrong if it prefers either
- The promo effect looks like demand — the system has to separate the two or it will over-buy
How we deliver
The engineering gates that separate a demo from a deployment in this industry.
- Demand decomposition that splits the promo from the seasonal without double-counting either
- Store-level hierarchical forecasting with warehouse and store lanes in the same call
- An inventory ledger reconciled across POS, order book, and count — the forecast is only as good as the three agreeing
Call us when
- Promotional stock is the CFO's risk when the numbers are your lane
- The retail network is varied enough that a single national forecast hides a broken store
- Stock is planned to the day, but the promo was decided in minutes
Building in retail?
Describe the situation in a paragraph. We will tell you what we would look at first and what it would cost to find out.