The allocation analyst is one of the most undervalued and overworked roles in retail planning. In a typical mid-size specialty retailer, a single analyst is responsible for allocating hundreds of SKUs across dozens or hundreds of doors — making micro-decisions about which store gets how many units, every week, often with incomplete data and insufficient time. The result is allocation that is inconsistent, reactive, and almost always suboptimal.
The Core Product Problem
The problem is most acute for core carryover products — the styles that are always in the line, always in the assortment, and represent a disproportionate share of total revenue. These are the items where allocation consistency matters most, because they drive the baseline sales that fund everything else. And yet they are almost always the items where allocation is most manual, most inconsistent, and most error-prone.
The reason is structural. New and fashion products get attention because they're new. Core products are assumed to take care of themselves. But 'taking care of themselves' in practice means an analyst making a judgment call based on last week's sales and whatever institutional knowledge they happen to have about each store. That is not a system. That is a person doing their best with inadequate tools.
"Core products don't need creativity. They need consistency. And consistency at scale requires a system, not a spreadsheet."
Peng Huang
What an Allocation Engine Actually Does
An allocation engine for core products is, at its simplest, a rules-based model that answers three questions for every SKU, every week: How much inventory does each door currently have? How fast is each door selling? And how much inventory does each door need to maintain optimal stock levels through the next replenishment cycle?
- 01Door-level inventory position: units on hand by store, updated weekly from the POS feed
- 02Door-level sell-through velocity: trailing 4–8 week average weekly sales, adjusted for seasonality
- 03Replenishment need: weeks of supply on hand versus target weeks of supply (typically 4–8 weeks for core products)
- 04Allocation recommendation: units needed to bring each door back to target weeks of supply
- 05Constraint layer: available inventory in the DC, vendor lead time, minimum ship quantities
The Door Tiering Model
The most effective allocation engines incorporate a door tiering model — a ranking of stores by their sales productivity for each product category. When inventory is constrained (which it almost always is), the engine prioritizes allocation to Tier 1 doors (highest productivity) before filling Tier 2 and Tier 3 doors. This ensures that the highest-velocity doors are never the ones running out of stock.
In one recent engagement, implementing a tiered allocation engine for a premium footwear brand's core carryover styles increased full-price sell-through by 27% and reduced allocation cycle time from 3 days to under 2 hours. The total inventory budget was unchanged. The difference was purely in how the inventory was distributed.