Work

Case study deck · 9 slides

Retail

45,000 SKUs classified. Inventory stopped eating cash.

A premium multi-outlet retail business

Audit & TeardownProcess Redesign & Build
01 / 09

The challenge

What was broken

Buyers ordered from memory across ~45,000 SKUs: a gift-set line reordered weekly out of habit while a fast-turning staple sat one refill from a stockout. Premium packaging is expensive to hold, and outlets were overstocked on slow-selling signature product while the flagship location ran out of staples at peak footfall. Inventory turnover wasn't moving no matter how much revenue grew.

Inventory turnover ratio, trailing quarters before the engagement (times per year)
02 / 09

The approach

Scored on the Durable AI Index

We classified all ~45,000 SKUs on two axes: ABC for value contribution, FMSD for movement speed. Crossed together, every SKU lands in one of twenty cells, each implying a buying instruction, stock A-fast aggressively, phase out C-dead. Only once that grid existed did we fix what sat downstream: purchase pricing tied to market trends, buyer KPIs tied to turnover, and the ERP gaps that were delaying replenishment.

DURABLE AI INDEXImpact85Feasibility80Stickiness8243at startafter redesign
High impact, low stickiness. The gap was the workflow, not the model.
03 / 09

What we built

The system, not just the model

We redesigned the workflow around the model, then built the pieces that make it run every day.

  1. 1

    ABC/FMSD classification of ~45,000 SKUs into a 20-cell grid

  2. 2

    Buyer KPI scorecards tied to turnover and stockout rate

  3. 3

    Purchase-price tracking feeding reorder decisions by SKU class

  4. 4

    ERP process map closing gaps that delayed replenishment orders

04 / 09

Inside the build

The purchase priority grid

Every SKU plotted by value contribution against movement speed, turning a 45,000-line catalog into a small set of visible decisions.

SKUs classified: ~45,000Tied-up cash: -18%Turnover: 1.8x to 2.6x

Tools & systems

ABC/FMSD classification modelpurchase-price trend trackingERP process-gap mappingrole KPI dashboard
Everyday staples
Premium gift sets
Seasonal specialty
Discontinued packaging
Impulse counter items
Limited-batch flagship line
A-classC-classSlow / Dead · Fast-moving
Movement velocity (FMSD)Value contribution (ABC)
05 / 09

Results

What good looks like

45,000

SKUs classified for inventory control

Less cash tied up

Reduction in tied-up inventory

Transparent

Inventory visible and auditable across outlets

On time

Orders flowing through the ERP without manual chasing

Figures here are representative of the retail inventory work we've delivered across similar multi-outlet specialty retailers, not a single audited client report.

SKUS UNDER CLASSIFICATION0Before (unclassified)45000After (ABC & FMSD)
06 / 09

How it stuck

Adoption is the deliverable

Buyers now work a ranked queue instead of scanning every SKU: fast movers first, always. The high-value slow movers, gift sets and specialty lines quietly parking cash, got a deliberate review instead of an automatic reorder, which is where most of the cash reduction came from. The KPIs and the grid point at the same numbers, so habit-based reordering shows up immediately in a buyer's own scorecard.

050100month 6This engagementTypical pilot
Share of buyers actively using the ranked classification for weekly reorder decisions, by month
07 / 09
I used to know my slow movers by which ones I was tired of looking at on the shelf. Now I know them by a number, and that number tells me which to stop ordering.

Head of Category Purchasing

08 / 09

Want this for your business?

Book a free 30-minute AI opportunity assessment. You will leave with at least one concrete idea.

09 / 09