Choosing an operations system for an online or multichannel retail business is one of the highest-leverage and most error-prone decisions a growing retailer makes. The market is crowded, the terminology is inconsistent, and almost every vendor describes its product as the obvious answer. This guide deliberately takes no side. It sets out what an ecommerce business actually needs from an ERP or operations platform, frames the real trade-offs, and flags the requirements that are specific to operating in Australia and New Zealand.
Use it the way you would use a buyer's checklist: as a neutral reference for scoping your own requirements before you sit through a single demo.
Start with the workflow, not the feature list#
The most common mistake is shopping for features before mapping the workflow. A feature list tells you what a system can do; your workflow tells you what you actually need it to do. Before evaluating anything, write down the full journey of an order in your business — from the moment a customer clicks buy, through payment, allocation, picking, packing, shipping, delivery and, inevitably, returns.
Do this for every channel you sell on, because the journeys differ. A marketplace order, a wholesale purchase order and a retail-store sale rarely behave identically. The points where those journeys diverge are exactly where generic systems struggle and where you will spend your integration budget.
Then identify your system of record. This is the single place where stock levels and financial truth live. Everything else — storefronts, marketplaces, label printers — should be treated as a satellite that reads from and writes to that record. Confusion about which system is authoritative is the root cause of most ongoing data problems, so settle it early.
Multichannel order management#
Order management is the heart of an ecommerce operations system. Its job is to ingest orders from every channel into one queue, normalise them into a consistent format, and move them through to fulfilment without manual re-keying.
At minimum, look for the ability to consolidate orders from your webstore, any marketplaces, wholesale and physical retail into a single view. Re-keying orders by hand is not just slow; it is the largest source of fulfilment errors in growing retailers.
Beyond consolidation, evaluate how the system handles the awkward cases: partial fulfilment when only some lines are in stock, split shipments from multiple locations, backorders, pre-orders, and order editing after the fact. These edge cases are rare individually but constant in aggregate, and they are where weak systems force staff back into spreadsheets.
Pay attention to order routing logic. A capable system decides automatically which location or 3PL should fulfil each order based on stock availability, customer location and shipping cost. Manual routing is tolerable at low volume and unworkable at scale.
Inventory sync across channels#
Real-time, accurate stock across every channel is the requirement retailers most often underestimate. When you sell the same SKU in several places, every sale must decrement a shared pool quickly enough that two channels never sell the same last unit.
The mechanics matter. Ask whether stock updates are event-driven (pushed the instant something changes) or batched on a timer. Batched updates measured in tens of minutes leave a window in which overselling is not just possible but likely during busy periods.
Allocation strategy is a genuine decision, not a default. You can expose your entire stock pool to every channel and accept occasional oversells, or ring-fence buffers per channel to protect marketplace performance metrics at the cost of hiding some sellable stock. Neither is wrong, but the system must let you choose and adjust the rule per channel.
Also confirm how multi-location stock is handled, how the system treats bundles and kits (where one sale depletes several component SKUs), and what happens when a channel's API is temporarily unavailable — does the system queue and retry, or silently drift out of sync.
Fulfilment, 3PL and WMS#
Fulfilment is where the operations system meets the physical world, and the right shape depends on who picks and packs. If you fulfil in-house, you need warehouse functionality: pick lists, pack verification, barcode or handheld scanning, bin and location management, and wave or batch picking as volume grows. A barcode-verified pack step alone removes a large share of mis-ship errors.
If you outsource to a third-party logistics provider, the requirement shifts to integration quality. The system must send orders to the 3PL cleanly and receive back accurate shipment confirmations, tracking numbers and stock-level updates. A 3PL relationship is only as good as the data flowing both ways.
Many growing retailers run a hybrid model — some SKUs in-house, some at a 3PL, some drop-shipped by suppliers — so check that the system can route a single order across these fulfilment paths without manual intervention. The distinction between a lightweight inventory tool and a true warehouse management system usually comes down to whether it can run a physical pick-and-pack operation, not just count stock.
Returns and reverse logistics#
Returns are a first-class workflow, not an afterthought, and ecommerce return rates can run far higher than physical retail. A weak returns process quietly erodes both margin and customer trust.
Look for a structured returns flow: the customer or staff initiate a return authorisation, the system tracks the inbound parcel, and on receipt staff inspect and decide the disposition — back to sellable stock, to a quarantine or damaged location, or to write-off. Each path has different inventory and accounting consequences, and the system should handle them distinctly rather than dumping everything back into available stock.
Tie returns to refunds and exchanges cleanly. The financial side — refunding the original payment method, issuing credit, or processing an exchange — needs to reconcile automatically with your accounting, or returns season becomes a manual bookkeeping burden.
Payments and accounting integration#
For ANZ retailers, accounting integration is frequently the make-or-break requirement. Xero and MYOB are the dominant platforms in the region, and the difference between a maintained, native integration and a generic CSV export is the difference between automatic reconciliation and hours of manual journal entry.
Confirm exactly what syncs and in which direction. Strong integrations push invoices, payments, refunds, cost of goods sold and inventory valuation into the accounting ledger automatically, and reconcile payment-gateway settlements against orders. Weaker ones export a flat sales summary that loses the detail your accountant needs.
GST handling deserves explicit scrutiny. The system must apply 15% GST in New Zealand or 10% in Australia correctly, support GST-inclusive and GST-exclusive pricing, and handle the wrinkles of cross-Tasman and export sales. Get this wrong and you inherit a compliance problem, not just a data problem.
On payments, check that the system reconciles across every method you accept — card gateways, buy-now-pay-later providers, PayPal, marketplace-collected payments — because each settles differently and on different timelines.
Demand forecasting and replenishment#
Forecasting separates reactive retailers from planned ones, but be honest about the sophistication you need. At its simplest, the system should show sales velocity per SKU, flag low stock against reorder points, and suggest purchase orders. That alone prevents most stockouts and a good deal of dead stock.
More advanced systems layer in seasonality, lead-time variability and trend detection. These are valuable, but only if your data is clean and your sales history is long enough to be meaningful. A sophisticated forecast built on six weeks of patchy data is false confidence.
Judge forecasting tools by how they handle uncertainty, not by the elegance of the model. Useful systems expose their assumptions and let you override them with human judgement, because no algorithm anticipates a supplier delay or a viral product moment.
Peak-season scaling#
Australian and New Zealand retail is intensely seasonal, and peak events compress months of volume into days. Black Friday, Cyber Monday, Boxing Day and the pre-Christmas rush will stress every part of your operation at once.
Ask vendors for concrete throughput numbers and how the system behaves at several multiples of your normal daily order count. Probe the architecture: does it scale automatically, or does someone have to provision capacity in advance? What happens to integration queues when a channel floods you with orders? Does stock sync degrade gracefully or fall behind dangerously under load?
Peak readiness is operational as much as technical. The ability to onboard temporary pick-and-pack staff quickly, support handheld scanning so casual workers are productive within an hour, and stage or freeze non-urgent jobs so the system prioritises getting orders out the door — these matter as much as raw server capacity. A system that buckles for three days in late November can wipe out a year of efficiency gains.
Marketplaces, EDI and real-time stock#
Selling on marketplaces or supplying large retailers brings its own integration demands. Marketplaces such as Amazon, eBay, Catch, The Iconic and TradeMe each have their own listing, order and stock APIs, and their performance metrics penalise overselling and late dispatch harshly. Real-time stock sync stops being a nice-to-have and becomes a condition of staying on the platform.
If you wholesale into major retail chains, you will likely encounter EDI — electronic data interchange — the structured document standard those retailers mandate for purchase orders, advance shipping notices and invoices. EDI support is specialised; if it is in your future, confirm the system handles it natively or via a maintained connector, because retrofitting EDI later is painful and costly.
Shipping and carrier integration#
Carrier integration directly affects both cost and customer experience. The system should generate compliant shipping labels, select services by cost and speed, produce tracking that flows back to the customer, and ideally rate-shop across carriers per order.
In the ANZ market, that means working cleanly with New Zealand Post, NZ Couriers, Aramex and the major Australian carriers including Australia Post, StarTrack and CouriersPlease, often through a shipping aggregator. Address validation and rural-delivery surcharges are quiet margin killers in New Zealand in particular; a system that cannot model rural surcharges will under-quote shipping and erode your margin one order at a time.
Reporting and visibility#
Finally, the system must tell you what is happening across the whole operation. At a minimum you need sales by channel and SKU, margin by product, inventory valuation, fulfilment performance and returns rates, available without exporting to a spreadsheet and rebuilding the analysis by hand.
Consolidated cross-channel reporting is one of the strongest arguments for unifying your operations in the first place. If you have to stitch numbers together manually from each channel's own dashboard, you do not really have a single source of truth — you have several, and they will disagree.
Weighing the trade-offs#
There is no single right architecture. An all-in-one suite minimises integration seams but may compromise on any individual capability. Best-of-breed gives you the strongest tool for each job at the cost of building and maintaining the connections between them. A core ERP extended by connectors sits in between and suits many growing retailers.
The discipline that serves every retailer, whichever path they choose, is the same one this guide opened with: keep your system of record consolidated, map your real workflows before you evaluate anything, and weigh each requirement against the cost of the integration effort it demands rather than the polish of the demo.