Assortment and space
Localised assortment and planogram decisions with measurable margin and availability impact.
Industry
Retail & consumer
Grocery, apparel, beauty, and consumer goods — margin lives in the last mile of availability.
Where we start
Retailers have more customer data than almost anyone and less ability to act on it than they expect. The gap between insight and shelf is where value is lost.
The winners pair demand-shaping models with supply capability, so a change in what is promoted is matched by what is physically available, in the right store, at the right time.
We also insist on honest measurement. Uplift reported against an unrandomised holdout does not survive a finance audit, and the credibility loss costs more than the slower experiment would have.
Typical results
- -12%
- inventory with no service loss, industrial distributor
- +2.8 pts
- on-shelf availability in grocery
- 1
- randomised holdout for every reported uplift
Illustrative client outcomes, representative of engagements in this sector.
Talk to this practiceHow we help
Pricing and promotion
Elasticity models and promotion optimisation with clean holdout design.
Supply and availability
Replenishment and allocation that respond to demand signals daily rather than weekly.
Customer experience
Personalisation with explicit preference controls and legible loyalty value.
Evidence
What this looks like in practice
- -12%
- inventory with no service loss, industrial distributor
- +2.8 pts
- on-shelf availability in grocery
- 1
- randomised holdout for every reported uplift
Related insights
Article
Retail's next chapter: hyper-personalisation without the creep factor
Personalisation works when customers can see the trade. The retailers winning with AI are the ones that made the exchange explicit.
29 May 2026 · 7 min read
Case study
Reinventing the supply chain with autonomous planning
A European industrial group replaced monthly planning cycles with continuous, agent-assisted replanning — and cut expedited freight by a third.
24 July 2026 · 8 min read
Next step
Bring us your retail & consumer problem
A ninety-minute working session is usually enough to agree the baseline, the owner, and the first evidence gate.