What exactly should match?
Agree grain, period, currency, units, status rules, keys and the meaning of a successful match.
Synthetic composite case · FMCG operations
When three sources disagree, the exception process becomes the product.
An illustrative case showing how a supplier file, an internal ERP and retailer sell-out data can be reconciled into a controlled operating view.
Explore data reconciliationRead the case
Illustrative example · No client, logo or commercial result is implied
The situation
The business did not need another dashboard. It needed a trusted comparison.
A supplier sent shipments by product code, the ERP recorded invoices by internal SKU, and a retailer supplied sell-out by a different product and location structure. The numbers could be compared in principle, but the definitions, timing and identifiers were not aligned.
This is a synthetic composite case built from common FMCG operating patterns. It is designed to show the method, not to claim a client engagement or an outcome.
Where do the sources diverge?
Compare totals and records, classify timing, identifier, duplication, missing-data and genuine business exceptions.
Who resolves the next difference?
Assign owners, evidence, priority, status and a repeatable review cadence so the queue does not disappear into email.
Inputs and outputs
A reconciliation view that can survive the next reporting cycle.
Supplier file, ERP extract and retailer sell-out with owner, cadence, cutoff and known limitations.
Product, customer, location, date and unit correspondences with versioned rules.
Record counts, quantities, values, duplicates, missing keys and period-level checks.
Difference type, owner, priority, evidence, proposed action and resolution status.
Trusted totals and known caveats for replenishment, margin, service level or account review.
Definitions, assumptions, refresh steps and escalation rules that another team can follow.
Controls and boundaries
Automation can compare the records. Ownership still resolves the exception.
The example assumes structured files, database queries or existing workflow tools rather than a new platform. Automation can standardize fields, run checks and surface differences; it should not silently overwrite source data or decide which business definition is correct. The control owner remains accountable for the rule, evidence and final disposition.