System 04 · Commercial intelligence

Margin & Pricing Intelligence

True margin and recommended pricing across 1,153 SKUs, computed locally from real supplier invoices - turning pricing from a spreadsheet habit into decision infrastructure.

StatusOperational Scale1,153 SKUs DeploymentLocal internal application

Problem

Pricing decisions without true costs

Real margin lives in supplier invoices - PDFs with net purchase cost, VAT and shipping - not in the selling price alone. Across more than a thousand SKUs, keeping that picture current by hand is not realistic, so pricing decisions drift away from actual economics.

This system is part of the Commercial Intelligence flagship: proprietary decision infrastructure around revenue, margin and pricing instead of disconnected spreadsheets and third-party SaaS.

Architecture

Invoices in, pricing scenarios out

The system processes PDF supplier invoices, net purchase cost, VAT, shipping and selling price, with advertising data available through a separate sub-dashboard. Mathematical optimization is operational.

  • Input
  • Agent
  • Orchestrator
  • Governance gate
  • Output
  • Human judgment
Benchmark: approximately 1,000 real PDF invoices parsed in 74 seconds. The engine recommends; the human prices.

Outputs per SKU: current margin, recommended price, current vs recommended price, pricing scenarios, margin scenarios, profitability prognosis and advertising-spend context.

Outcome

What it changed

The system has informed real pricing decisions and contributed to improved margins. No percentage is attached to that claim - the honest statement is that pricing decisions are now made against computed economics rather than approximations.

It runs as a locally deployed internal application: no external SaaS dependency, no data leaving the building, and complexity limited to what the problem actually requires.

Related