5 to 10M€ in value leaks identified within weeks in a mid-size industrial company
How GenieFactory cross-referenced ERP data and business knowledge to map the hidden losses of a complex supply chain — and deliver a costed action plan.
In brief. A French industrial mid-size company with €115M in revenue, recently migrated to SAP S/4HANA, faces a supply chain under pressure: high working capital, structural delays, chronic emergencies. Within a few weeks, GenieFactory built a Knowledge Graph cross-referencing 8 SAP exports and 4 business interviews — and identified 5 value leaks, of which 2 to 4M€ are recoverable as quick wins within 6 months.
Context
An industrial sector under pressure. €115M in revenue, 2,000 employees. Product mix of standard (70–80%) and custom (20–30%). Multi-supplier network, long lead times, complex planning cycle.
The SAP S/4HANA ERP was recently migrated. All modules are in place — FI/CO, SD, MM, PP, PM. But planning parameters have not been recalibrated. Safety stocks are round numbers, unchanged for two years. The sector market contracted by 17.6% in 2024. In this context, every euro of immobilised working capital is a lever.
Management knows there are leaks. They don't know exactly where. And teams, caught in the operational flow, don't have the distance to objectify them.
The approach: Identify and Frame before acting
GenieFactory applied the first two steps of the ICPC framework — Identify and Frame — before any recommendation.
Structured business interviews. Four sessions with sales administration, procurement, production and logistics teams, guided by the platform. The goal is not to collect declared processes — but to capture the actual work. The gap between the two is systematically where losses hide.
SAP data extraction and modelling. Eight cleaned and cross-referenced SAP exports: items, stock, movements, orders, manufacturing orders, bills of materials, management accounting. A "planning-ready" pivot model is built from real data, not theoretical configurations.
Knowledge Graph construction. The four interviews are transformed into four process graphs, then merged into a unified Knowledge Graph. This graph links flows end to end — from quote to delivery — incorporating the real constraints and frictions reported by the teams.
Enriched scorecard. Sixteen KPIs calculated across the PLAN→BUY→MAKE→STOCK→DELIVER chain. Each indicator is enriched with root causes extracted from the Knowledge Graph. The result moves from "you have a problem" to "here's why and how to fix it".
The 5 value leaks identified
Structural raw material overstock — 3 to 5M€ in immobilised working capital. Supplier lead times in the ERP do not match actual lead times. The company orders and stocks too early. On €24.3M of raw material stock, the reduction potential is significant.
Excessive work in progress — 1 to 3M€. Manufacturing orders are launched without all components being available. Bottlenecks accumulate, WIP swells, lead times lengthen. "Full kit" is not systematically verified before launch.
Untreated MRP nervousness — high indirect impact. Since SAP parameters were not recalibrated after migration, the system generates an avalanche of contradictory "Reschedule In/Out" messages that no one processes. Planning reliability deteriorates, teams lose confidence in the ERP and work around it.
Compensatory premium transport — €200–500K/year. Express deliveries have become standard operating procedure to make up for structural delays. Urgency is managed in real time rather than prevented upstream.
Slow-moving and dead stock — €500K to 1M€. Capital immobilised without foreseeable use, for lack of a systematic review of the item portfolio.
Estimated total impact: 5 to 10M€, of which 2 to 4M€ recoverable as quick wins within 3 to 6 months — primarily through MRP parameter recalibration and the implementation of a tooled weekly stock review.
What AI changes in this type of engagement
A classic supply chain diagnostic analyses data or interviews — rarely both together, rarely within a few weeks.
The Knowledge Graph cross-references both sources continuously. A discrepancy between the lead time declared by procurement and the actual lead time in SAP data surfaces automatically. Root causes are linked to KPIs, not listed separately.
Speed also changes the nature of the engagement: teams receive an objectified diagnosis while they are still in the context of the problem — not six months later, when priorities have shifted.
Finally, everything built — business ontology, scoring rules, diagnostic patterns — is encoded in the Knowledge Graph and belongs to the client. The next diagnostic, in the same sector or an adjacent one, starts from this base.
A replicable model
The approach applies to any industrial mid-size company with an ERP (SAP, Sage, Oracle, etc.) and a supply chain with a mix of standard and configured products. The ERP connectors, extraction pipeline and Knowledge Graph engine are the same. What adapts is the sector-specific business ontology.
GenieFactory works with vertical partners who bring sector expertise. The client retains ownership of their data and Knowledge Graph — with no dependency on a tool or a service provider.
Does your supply chain have leaks you haven't spotted yet?
To understand the method behind this diagnostic: The ICPC framework