The Problem
The company had invested heavily in digital infrastructure across its Global Logistics and Network Services organization. p44 provided visibility. A global TMS held shipment records. Trax handled the freight audit. Yet 60% of effort across procurement and audit teams remained manual — because each system worked individually while the end-to-end process did not work as one. Trax processed invoices against rate cards but offered no root-cause analytics on why exceptions occurred, no first-pass match rate tracking, and no audit savings measurement. The exception backlog accumulated with no structured tracking. Finance teams managed 200-plus GL allocation rules manually across an 8-segment structure. Rate cards for spot quote validation were managed by email with no automation or approval workflows.
The procurement side had its own inefficiency. The 8-week global tender cycle ran through Excel and email with limited benchmark linkage. By the time rates were negotiated and awards finalized, market conditions had often shifted. Benchmarking against external rate sources — DAT, Xeneta, FreightWaves — required manual research that happened after bid events closed rather than during them. The process produced contracts. It did not produce intelligence.
The compliance stakes were high. SOX 404 required auditable controls across all freight transactions — every payment traceable from invoice to contract to shipment to GL posting, with segregation of duties and immutable audit logs. EY audited the company’s systems. The manual processes that had worked at smaller scale could not meet the audit readiness standard the company’s size demanded across 35 legal entities, 1,480 cost centers, and 18 currencies.
What Freehand Did
Freehand replaced Trax in a phased deployment — Phase 1 activated the AI Procurement Analyst and Rate Manager Agent, replacing the 8-week Excel tender cycle with AI-driven sourcing events benchmarked against live market data. The first procurement cycle ran 70% faster than the prior manual process. Phase 2 extended to Freight Audit and Pay: the Audit Agent runs across all 370,000-plus annual bills — invoice against contracted rate, rate against p44 and TMS shipment actuals, accessorial charges against entitlement schedules. Every invoice carries a confidence score. High-confidence invoices auto-approve with a full SOX-compliant audit trail. Low-confidence items route to the Exception and Dispute Agent with evidence assembled.
The 200-plus GL allocation rules that had required manual intervention are now automated through the GL Coding Agent — every freight cost allocated to the correct legal entity, cost center, and financial period based on shipment characteristics, with self-correcting logic for edge cases. The 8-segment GL structure that had produced recurring exceptions now operates without manual overrides. Segregation of duties is enforced in the workflow architecture: the Audit Agent validates and recommends, the Spend Intelligence Agent reports, and human approvers retain execution authority within a governed approval matrix. Every decision is logged, timestamped, and traceable — EY-ready without additional reporting overhead.
Phase 3 connects the procurement and audit layers through the Spend Intelligence Agent — accruals running against actual audited data rather than estimates, ESG tracking through SmartWay carrier filtering and GLEC-aligned CO2 metrics integrated into procurement scoring, and predictive budget accuracy giving the GLNS finance team visibility before close rather than during it. The AI Teams that replaced manual coordination across procurement, audit, and finance now operate as a single learning ecosystem. The 2–4% overpayment recovery through duplicate elimination and the under-12-month payback are the financial output of replacing a fragmented point-solution stack with one platform.
