Reconciling Two Inventory Systems in a Major Insurance Investigation
When a significant discrepancy emerged between physical inventory and ERP records, our team reconciled two independent systems over time — identifying precisely when and where they diverged, and helping the investigation distinguish between potential stock loss and inconsistencies in the operational data.
A significant shortfall — and two systems that could not be compared
A major industrial processor reported a significant discrepancy between its physical inventory position and the figures recorded in its ERP system. The investigation needed to answer a fundamental question: did the shortfall represent actual stock loss, or inconsistencies in how inventory had been recorded across the organisation's systems?
The two inventory systems held relevant data, but they had not been designed to talk to each other. Different data structures, date conventions, and category definitions meant no direct comparison was possible from the raw records. Before the investigation could form a view, the systems had to be brought into alignment — and that alignment had to be built on a methodology the investigation could rely on.
Building a reconciliation methodology that would hold up
The core judgment was in the design of the reconciliation, not the extraction. We determined how to normalise data across two incompatible systems, how to handle gaps and inconsistencies in the operational record, and how to structure the comparison so that meaningful divergence could be distinguished from noise.
The pattern that emerged — periods of alignment followed by concentrated divergence in specific months — was not visible from either system in isolation. It became visible because the reconciliation was structured to surface it. That is the difference between processing records and analysing them.
From incompatible systems to a structured comparison
- Assessed both inventory systems and documented the structural differences requiring resolution before comparison was possible
- Ingested operational records, monthly exports, and ERP data
- Normalised and aligned data fields to enable direct cross-system comparison over time
- Built a month-by-month reconciliation identifying inventory values and variances across both systems
- Isolated and prioritised periods of material divergence for further investigation
What the reconciliation revealed
Inventory values across the two systems showed periods of alignment followed by divergence — with the pattern concentrated in specific time periods rather than evenly distributed across the record.
Discrepancies were not uniform. Specific months showed materially higher variance, making it possible to focus the investigation on the periods and records where the discrepancies were most significant.
The structured comparison created the basis for distinguishing between potential stock loss and inconsistencies in how inventory had been recorded operationally — a distinction the investigation needed to make before forming a view.
An investigation with a factual basis to act on
The investigation team received a structured reconciliation showing, month by month, where the two systems aligned and where they diverged. Every variance was traceable to underlying source records on both sides, and the analysis identified clearly which periods and records warranted further scrutiny.
The methodology was documented and repeatable — meaning the conclusions drawn from it could be explained and defended if the approach was later challenged.
Cross-system comparison revealed where inventory records diverged, helping the investigation distinguish between potential stock loss and inconsistencies in operational data — and focus its attention where it mattered.