Why duplicate SKUs and wrong units of measure break order automation (and how to find them)

Why does the item master matter to order automation?

Every order line ends as a reference to one product in your item master, the list of products your ERP knows. Catalogue matching can only be as right as the catalogue it matches against.

A person keying orders compensates for a messy item master without noticing. They know that 'CL500PET12' is the old code and the new one is 'COLA-500-PET-12'. They know that bottled water is listed as Each but everyone orders packs. That knowledge is real, and it lives in their head.

Automation cannot rely on it until it is written down. So when a business automates order capture, the item master's quiet defects become visible. That is not a setback. It is the first time the data has been checked against how the business actually trades.

What defects are most common in an item master?

Item masters grow over years, through supplier changes, ERP migrations, new staff and one-off products. These are the findings a Stock Data Audit looks for:

How do duplicate SKUs happen?

Rarely through carelessness. Usually through perfectly sensible events: a supplier changes a product code, an ERP migration brings old codes across, or someone creates a code in a hurry because the existing one could not be found.

Packaging changes are a common trigger. The GS1 GTIN Management Standard states that a change to the number of trade items in a case requires a new GTIN, the barcode number, at case level. When a supplier moves a case from 12 units to 24, the case barcode changes. If the item master creates a new code for the new case without retiring the old one, it now holds two codes for what customers see as one product.

How do unit of measure errors happen?

A unit of measure error is usually a mismatch between how a product is bought, stocked and sold. A distributor buys bottled water in shrink-wrapped packs of six and stocks it by the pack. It sells by the pack to most customers, but by the bottle to a few.

If the item master records the selling unit as Each, every customer who orders '10 water' is ambiguous. Ten bottles or ten packs? The person keying the order knows which customer means which. The item master does not.

How do item master defects show up as order errors?

Each defect has a signature in the orders. Knowing the signature is how you trace an order error back to its cause.

• A duplicate SKU splits stock and sales history in two. Half the orders land on one code, half on the other. One shows stock-outs while the other shows overstock.
• A wrong unit of measure ships twelve when one was ordered, or one when twelve were. Invoicing follows the wrong unit, and a credit note follows the invoice.
• An inconsistent pack size breaks pricing in one direction and stock in the other. A case price is applied to singles, or singles are received as cases.
• A missing price leaves an order that cannot be priced from the customer's list, so it stops for review.
• A near-duplicate description makes catalogue matching choose between two plausible products, which is a review every time.
• An inactive but sellable code accepts an order for something that will never ship.

In an automated flow, most of these appear as review reasons: 'SKU not in catalogue', 'quantity does not fit the pack', 'price differs from list'. Most order review reasons trace back to the item master. Fix the data and the review queue shrinks.

How do you find duplicate SKUs and unit errors?

A spreadsheet sort by description finds the obvious cases. The rest need evidence from outside the item master: how products are described, barcoded, packed, ordered and received.

1. 01Compare descriptions after normalising spacing, case, abbreviations and pack wording. 'Cola 500ml PET x12' and 'COLA 500 PET 12S' become comparable.
2. 02Compare barcodes. Two codes carrying the same barcode are almost always one product.
3. 03Compare pack and unit fields against each other and against the description. A description that says '6 x 1L' on a record with a pack of 1 is a candidate.
4. 04Compare against ordering patterns. If customers consistently order a SKU in cases while the catalogue says Each, the selling unit is probably wrong.
5. 05Compare against receiving. If recent deliveries arrive in cases of 24 while the catalogue says 12, the pack has probably changed.
6. 06Look for codes with no orders, no receipts and no stock over a long period. They are candidates to retire.

Weeva reads your stock master the way it reads orders. Duplicates are judged by description, barcode, pack and how customers order them. It shows its reasoning for each finding, so your team can disagree.

How do you fix item master defects safely?

Item master fixes touch stock, history, price lists and open orders. The safe approach is deliberate and recorded.

1. 01Assign each finding to a person who owns that product range. A finding without an owner is a spreadsheet nobody opens.
2. 02Merge duplicates rather than deleting them. Keep one code, move stock and open orders to it, and redirect the retired code so old references still resolve.
3. 03Set the selling unit and add the pack where a unit is wrong, rather than changing historical quantities.
4. 04Confirm a pack change with the supplier's current delivery notes before updating the catalogue.
5. 05Add missing prices, or retire the SKU if nobody should be selling it.
6. 06Verify each fix against the ERP on the next sync, and keep the history.

Stock Data Audit does not change data in your ERP. It finds and proposes. Your team makes the fix in Weeva, where the connector supports it, or directly in the ERP, and Weeva verifies it on the next sync.

How do you keep the item master clean?

A one-off clean-up decays. New products, new suppliers and new customers bring new defects every month. The durable answer is an audit that runs all the time.

Weeva runs the audit continuously, on every catalogue sync, with a summary each week. New findings become tasks with an owner. You can start without a connection: upload a CSV or Excel export of your item master and the audit runs on that.

The audit also feeds order automation, and order automation feeds the audit. The learning loop on orders and the audit on stock data share what they learn. A catalogue fix removes a review reason from future orders. Cleaner data, fewer reviews.

To see what an audit finds, book a demo and we will walk through a sample audit with you.