A guide to unit of measure conversions when migrating inventory data

When Australian retailers move from one ecommerce platform to another, the technical work behind the scenes often determines whether shelves stay accurate or stock counts fall out of sync. Unit of measure conversions sit at the heart of that work. Each case, each pallet, and each metre of fabric in a Brisbane warehouse may be recorded differently across legacy systems and modern ecommerce engines, and getting the math wrong at this stage ripples into pricing, fulfilment, and customer trust long after the cutover.

Australia's retail landscape adds its own texture to the challenge. Most operators record quantities in metric units such as kilograms, litres, and millimetres, yet many global software templates default to imperial measurements. GST at 10% sits on top of almost every transaction, which means rounding errors in unit conversions can quietly inflate or shrink taxable amounts. Retailers from Sydney's CBD boutiques to regional stores in Hobart are expected to keep pricing consistent across online and in-store channels, even when back-office systems disagree on how a single box is defined.

The team behind ncrretailonline.com spent years helping Australian retailers tie online storefronts to point-of-sale and inventory layers. With the original platform now discontinued, many businesses are rerunning their stock files through partner ecosystems built on Magento, WooCommerce, and Counterpoint. That rerun is precisely where unit conversions either hold steady or fall apart.

This guide walks through the practical decisions a retailer faces when moving stock data between systems, from mapping cartons and dozens through to auditing mass and volume records. It is written for store owners, operations managers, and the consultants who support them during a platform transition.

Common units of measure used by Australian retailers

Metric units dominate Australian retail software, but the way they are grouped varies widely. Grocers in Perth may stock rice by the kilogram while a Melbourne gift shop records candles individually. Apparel retailers in Adelaide frequently lean on inner packs and cartons, whereas hardware sellers near Geelong deal with lengths of timber, square metres of flooring, and litres of paint.

The challenge emerges when one system stores "each" as the only unit of measure and another expects a hierarchy of inner pack, outer pack, and pallet. A single SKU can therefore be entered as 12 units, 1 inner, or 0.25 of a master carton, depending on where the file originated. Migration scripts need explicit rules for how those layers relate, otherwise totals will not reconcile against the original system.

Some Australian businesses also import goods labelled in imperial units from overseas suppliers and then resell under metric descriptors. That dual labelling quietly introduces conversion drift if it is not caught early, and the drift tends to show up first on imported electronics and tools.

Mapping stock keeping units between old and new systems

A reliable mapping exercise starts with a clean SKU audit before any data moves. Each SKU in the source file should be paired with its destination counterpart, and any unit attributes attached to it should be catalogued. For businesses transitioning from NCR Retail Online, this is often the first opportunity to clean up decades of free-text entries that no longer match current product ranges.

When the mapping is complete, conversion logic can be applied. If the old system records a 24-bottle case as 24 each and the new system expects 1 case, the migration script must divide by 24 and add a new unit tier. Doing this manually for thousands of SKUs across a warehouse in Parramatta or a distribution centre in Townsville quickly becomes unsustainable, which is why repeatable rules matter more than ad hoc edits.

Weight, volume, and dimension handling

Weight-based products bring the most arithmetic risk. A Perth-based bulk food retailer might buy flour in 25 kg bags, sell it loose by the gram, and quote shipping in tonnes. If the migration tool records only the bag weight and ignores the loose-sell variant, online listings will understate available stock by a factor of a thousand. Volume creates similar friction, particularly for liquids where litres and millilitres can be misread across decimal place conventions.

Dimensions add a third layer. Furniture sellers in Brisbane frequently work in centimetres for product pages but in metres for warehouse planning. A migration that ignores the unit shift can produce items that appear too large or too small for delivery vehicles. Local courier networks, particularly those serving regional Queensland and the Pilbara in Western Australia, charge by dimensional weight, so a miscalculation here feeds straight into freight costs.

The fix is to enforce a single canonical unit per dimension class, then convert every legacy value into that unit before the data lands in the new platform. A canonical centimetre for length and a canonical kilogram for mass keep downstream calculations predictable.

Pricing, GST, and rounding tied to units

Australia's Goods and Services Tax sits at 10% and is calculated on the taxable sale price, which is itself derived from per-unit values. If a system records a price per kilogram and the migration script misreads it as a price per gram, the displayed price will look a thousand times higher than intended. Customers notice quickly, and the Australian Competition and Consumer Commission takes a dim view of misleading pricing under the Australian Consumer Law.

Rounding rules deserve their own line in the migration plan. The Australian Taxation Office expects GST to be calculated to the nearest cent, so rounding direction can shift the tax payable by a fraction of a cent on each line. Over thousands of transactions, that drift adds up and produces headaches at BAS lodgement. Building a rounding tolerance into the conversion script protects against mismatched reports and keeps the bookkeeper onside.

Catching conversion errors before they reach customers

Pre-launch audits save more time than post-launch apologies. A useful starting point is a side-by-side report that compares source and destination quantities for a sample of high-volume SKUs. Choose items that move quickly, such as a popular breakfast cereal in an independent supermarket, because those expose rounding bugs fastest and reveal them to the team before customers spot them online.

Sign-off on conversion logic should sit with someone who understands both the old and new systems, not just the migration tool itself. For Australian businesses working through automation after transition, that person often holds an operations or category manager title rather than a developer role. Their judgement matters when the data looks technically correct but operationally suspicious.

A short list of common failure modes worth checking:

• Singular versus plural unit labels that the destination system reads as separate fields.
• Decimal commas in supplier feeds arriving from European suppliers rather than Australian dots.
• Missing "each" defaults when a product is sold loose without an explicit unit attached.
• Pack hierarchies broken when a parent SKU is retired but child SKUs remain live in the catalogue.

Building conversion rules into automation

Once manual mapping proves out, the next step is codifying it. Most modern ecommerce platforms accept unit of measure setup through CSV imports or API endpoints, and the same rules can be reused for future product additions. That reuse is where the real return on migration effort shows up, particularly for retailers managing thousands of SKUs across multiple stores from Cairns to Canberra.

Automation should still include human checkpoints. A weekly reconciliation report comparing on-hand quantities between the new ecommerce engine and the point-of-sale system catches drift caused by either side introducing new units without updating the conversion rules. For multi-store Australian retailers, this is especially important where one location might sell timber by the metre while another quotes the same product by the length.

The disciplined approach keeps stock trustworthy for customers searching online and for staff pulling orders on the floor. It also feeds clean numbers into the periodic stocktake, which any retailer running a full wall-to-wall count will appreciate.

Testing, validation, and go-live confidence

A staged rollout reduces risk. Migrating a single store or category first lets the team observe how conversions behave under real sales volume before the full catalogue moves. Choose a category with mixed units, such as a hardware range that includes screws by weight, paint by volume, and brackets by piece, to exercise every conversion rule in one pass.

Validation scripts should compare live stock counts against expected counts after the migration, with tolerances that account for in-flight orders. Once those tolerances hold for a full trading week, the team can extend the cutover to other stores with confidence and avoid the midnight panic of a stock count that refuses to balance.

A short checklist for the staged rollout:

• Pick a pilot store with the widest unit variety to stress-test conversions.
• Reconcile on-hand stock between POS and ecommerce every trading day of the pilot.
• Sign off rounding rules with the bookkeeper before processing live orders.
• Document every unit rule in a shared handover note for the wider team.

The practical takeaway for any Australian retailer approaching an inventory migration is to treat unit of measure conversions as a first-class workstream rather than a footnote in the technical plan. Audit the source, lock the canonical units, automate the rules, and reconcile often. Done well, the migration lands cleanly and stock counts match across every channel on day one.