Reporting Placed Quantities: The Data Duty Behind the Fees

EPR reporting states, per member state and period, the quantity of packaging the producer placed there - broken down by material and by the register's categories - from which fees are calculated and the system's statistics built. The data comes from two sources joined: packaging master data (the weight of every component of every SKU, in the component map's terms) and sales data (units shipped into each state), adjusted for the flows that are not placings in that state (exports onward, returns, B2B flows under their own arrangements where the rules distinguish them).

The pipeline is a data engineering exercise with legal consequences for error in both directions: under-reporting is free-riding; over-reporting is a donation. This guide builds the pipeline properly.

The join's quality is decided in the packaging master data: per SKU, per component, material and weight - maintained under the same change control as the technical files, because a specification change is a reporting change. The classic defects: weights inherited from a predecessor product; component lists missing the label, the adhesive, the cap's seal; multipack and transport layers omitted or double-counted across the join; and promotional and seasonal variants absent from the data entirely. A periodic physical audit - weigh a sample of SKUs against their master data - prices the defect rate, and the correction feeds every downstream calculation (fees, content percentages, fee modulation modelling) that drinks from the same data.

Three adjustments carry most of the judgement. Exports: packaging sold to a distributor in state A who ships to state B is the producer's placing in B under the distance logic in defined cases and the distributor's in others - the roles analysis decides, and the reporting must follow the decision state by state, with the evidence (sales terms, delivery data) filed. Returns and unsold stock: netted where the rules allow, on evidence of actual return or destruction, not on commercial assumptions. Channel splits: household versus commercial packaging follow different systems in many states, and the split must follow the actual channel, SKU by SKU where SKUs straddle both. Each adjustment is a documented position in the reporting file - the file an EPR audit will open first.

Reports are filed per cycle and corrected under the scheme's process when errors surface - and errors surface in every honest system, which is why the correction record is a sign of health rather than shame. Keep: the source extracts, the join logic's version, the adjustment positions, the submitted reports and the scheme's confirmations, per state per cycle, in the documentation architecture. An EPR audit (by the scheme or the state) reconstructs a cycle from these; a business that can replay its own reporting in an afternoon passes the audit's hardest test before the auditor asks a question. The replay drill, run annually on one state, is this station's equivalent of the documentation station's retrieval drill - schedule it together.

Close on reporting's second function: it is the mirror in which the producer sees its own packaging estate clearly - tonnage by material and state, variant discipline, the tail of zombie SKUs each carrying registrations, fees and files. Businesses routinely discover, in their first rigorous reporting build, products they did not know they sold and packaging they did not know they used. Treat the discovery as the system's dividend: the reporting pipeline, built for the fee, becomes the estate's census, and half the other stations in this guide series (grades, content, modulation modelling) draw their population from it.

The pipeline matures in stages most businesses recognise: spreadsheet joins maintained by one person's heroics; a scripted extract from the ERP and packaging master data with the join logic documented; integrated reporting where specification changes propagate to reporting automatically and the cycle's file assembles itself for review. Each stage is legitimate at its scale; the risk sits in staying at stage one past the portfolio size it can carry - visible when the heroic person is on holiday at filing time. The register discipline from the documentation station is the stage test: when the registers are data, the pipeline is a query; when they are memory, it is a person.