DC-to-Store Allocation Fix
Fixing how an apparel retailer's stock moved from distribution centres into stores.
Problem
Stores across the client's network were running out of key SKUs while nearby distribution centres held excess stock — a classic allocation mismatch that was costing sales and inflating carrying cost at the same time.
Business background
The client is an apparel fast-fashion retailer allocating inventory from 3+ distribution centres into a large multi-store footprint, where fashion cycles leave little room for slow reordering.
My role
I analyzed inventory flow end to end — from DC receipt to store-level sell-through — and built the SQL and Power BI layer that made the imbalance visible and actionable, working with JDA and ODBMS as the allocation systems of record.
Tech stack
SQL ServerPower BIDAXPower QueryJDAODBMSExcel
Approach
Consolidated multi-store, multi-DC transactional data into a reporting layer, then modeled demand patterns in SQL to surface where allocation logic was systematically over- or under-shipping specific categories.
Solution
Automated Power BI dashboards replaced manual, spreadsheet-driven KPI tracking, giving planners a live view of allocation accuracy and inventory precision instead of a weekly static report.
Challenges
Reconciling data across DCs and store systems that weren't built to talk to each other, inside an allocation tool (JDA) with its own logic and constraints.
Impact
30% fewer stockouts, 10% better allocation accuracy, 85% inventory precision, 20% shorter reporting cycle, and a 15% revenue uplift in key categories — plus a 12% lower acquisition cost and 10% better marketing ROI from the demand signal this work surfaced.
Key learnings
The biggest wins came from making an existing manual process automatic and visible, not from adding new complexity — precision and trust in the numbers mattered as much as the model behind them.
