Segmenting Physical Stores to Optimise Assortment and Space
The Operations Director of a chain with 350 stores wants to group the outlets according to the real purchasing behaviour of their neighbourhood, so as to stop sending the same standard catalogue to every store.
Where the money comes from
What to do now
- 1Segment the stores by real purchasing behaviour.
- 2Design 4-6 assortment modules, not 350.
- 3Pilot in 20 stores per segment versus a control.
For your committee
One page with the decision, the impact, the risks, the steps and the KPI to track.
The print dialog opens: choose "Save as PDF".
Watch out: more volume is not always more profit
Expanding the assortment for each neighbourhood multiplies SKUs, logistics complexity and stock. Tailor the assortment within a number of modules that logistics can serve profitably.
KPI for the committee
Margin per linear metre and shrinkage per store.
Detail for your data team
Proposed technique: K-Means Clustering
Why: It allows points of sale to be grouped by real demand patterns without relying on outdated geographic classifications (e.g. 'North vs South').
Cost of getting it wrong: Treating all stores the same leads to stock shortages in star products and excess slow-moving goods in others.
Data needed: POS by Store: Share of sales by category (fresh produce, ready meals, private label, organic products, quick purchases).; Commercial Environment: Foot traffic, peak hours and average ticket per store.