Case study · Department store

How a multinational apparel retailer compressed planning from three weeks to one day

A team of 12 spent three weeks every month producing forecasts. With RabbitHawk, all forecasts now run in a single day, shifting a reactive, spreadsheet-driven process to an adaptive, goal-led decision engine.

Client: Multinational apparel & department-store retailer
Industry: Department store
Country: USA
Revenue: Just under US$100M

Measured outcomes

The results of the engagement

Reduction in new inventory: 52%Reduction in new inventoryobserved sales revenue maintained during the evaluated period
Reduction in MASE: 96%Reduction in MASEversus saved legacy forecasts at style × colour × store × brand
Reduction in manual workload: 90%Reduction in manual workloadfrom weeks to hours
Decision cycle: 3 wks → 1 dayDecision cyclefrom debate to coordinated action

Client identity withheld under NDA.

Basis of comparison: The client's saved legacy forecasts compared with RabbitHawk forecasts over the evaluated period
Planning grain: Style × colour × store × brand
Evidence status: Measured client outcome approved for publication

This public summary does not disclose the exact evaluation dates, series count, exclusion or missing-data rules, or independent-check status. RabbitHawk does not infer those details.

Enterprise evaluation

Before: ERP-rich, planning-poor

Numbers everywhere. One trusted decision, nowhere.

The business was ERP-rich but planning-poor. A leading ERP proved rigid and slow to adapt, so the planning team had reverted to Power BI and spreadsheets. Numbers were everywhere; a single, trusted decision was nowhere.

The planning bottleneck

Three weeks of debate, every month

Each monthly cycle consumed weeks of debate and reconciliation. Twelve people spent three weeks arguing over whose forecast was correct: time spent reconciling numbers instead of acting on them, while long lead times and tight cash flow windows punished every delay.

The customer

A distributed supply chain under cash flow pressure

A multinational department-store operator with revenue just under US$100M. The company designs locally and outsources production across the USA, South America and Asia: a distributed supply chain with long lead times and tight cash flow windows. Its leading ERP proved rigid and slow to adapt, so the team had reverted to Power BI for planning. Each monthly cycle consumed weeks of debate and reconciliation, with teams arguing over whose forecast was correct.

The solution: an adaptive layer, not a replacement

RabbitHawk on top of the existing ERP

RabbitHawk deployed its Forecasting & Optimization product as an adaptive layer on top of the existing ERP, with no rebuild. It ingests client data exports, forecasts by style × color × store × brand, surfaces ageing inventory, projected stock-outs, MOQ impacts and margin risk, then exports optimized purchase orders back to the ERP: a governed workflow from data to action.

Multi-level forecasting

SKU, category, store and region, each with cross-functional visibility.

Probabilistic forecasting

Risk-aware predictions that show not just what is likely, but what is possible.

Goal-driven optimization

Purchasing that balances cash flow, margin, revenue and constraints (MOQ, lead time, risk, budget).

The workflow

A closed loop, with humans in control

Forecasting and optimization run as one governed sequence, and every recommendation is reviewed by a planner before it returns to the ERP.

01
Connect
Ingests client data exports from the existing ERP, with no rebuild and no migration.
02
Forecast
Forecasts probabilistically by style × color × store × brand, with honest ranges.
03
Optimize
Surfaces ageing inventory, projected stock-outs, MOQ impacts and margin risk, then optimizes purchase orders against cash flow, margin and constraints.
04
Approve
Planners review the assumptions, evidence and expected impact behind every recommendation, so humans stay in control.
05
Export
Optimized purchase orders are exported back into the ERP, completing the governed workflow from data to action.

“It’s the first time our meetings started with what’s possible rather than what went wrong.”

Planning Director·Multinational apparel & department-store retailer

Why it worked

Impact, not incrementalism

Extension, not replacement
Human-centered automation
One coherent planning truth
Probabilistic planning
Constraint-aware optimization

Extension, not replacement

RabbitHawk extends the ERP rather than displacing it: “add intelligence without rebuild” delivers immediate ROI.

Goal-aligned optimization

Optimizes for cash flow, GMROI and revenue within real constraints, not pure accuracy. The system explicitly represents what success means for the retailer.

Human-centered automation

Eliminating manual drudgery redirects expertise toward scenario planning and proactive risk management.

Demonstrable ROI

Outcomes quantified in both time and capital saved: practical, compounding returns over time.

What other retailers can learn

Transferable lessons

You do not need to replace the ERP to fix planning; add an adaptive intelligence layer on top of it.
Optimize for the outcome (cash flow, margin, service) rather than forecast accuracy alone.
Keep planners in the loop: auditable recommendations beat opaque automation for trust and adoption.

What’s next

Phase two expands RabbitHawk in two directions: a top-level UI that focuses daily decisions on broad company goals, and store-specific micro-analysis that finds opportunities in local demographics and buying behavior.

Explore retail applications

Could this be your planning story?

Tell us about your planning environment and the decisions you need to improve, and we’ll show you how RabbitHawk would approach it.

Discuss Forecasting & Optimization