An eCommerce tech stack is built around one website. The storefront sits in the middle, with apps around it for payments, email, reviews, shipping, and support. Most guides to the eCommerce tech stack stop there, and for a brand that only sells online, that is often enough.

A retail tech stack has to do more, because the same stock serves two kinds of customer: the person who walks into a store and the person who orders online. Stock now lives in many places, orders can be filled from many places, and every system involved holds its own version of the truth.

That is why the most useful question about a retail stack is not "which tools do we need?" but "which system owns each job?" Most stack problems start when two systems both think they own the same number, such as two stock counts for the same store, or when nobody owns a decision at all, such as which store should fill an online order. Overselling, orders keyed in by hand, and reports that never agree are usually ownership problems, not tool problems.

Most multi-store retailers end up with the same five core systems, plus a ring of supporting tools. Here is what each one is for.

eCommerce platform. The storefront: product pages, cart, checkout, and the customer's online account. Shopify is the common choice for brands with stores, because the storefront and the POS run on one platform. The eCommerce platform should own how products are shown and sold online, not how much stock exists or where an order is filled.

Point of sale (POS). The register and the tools store teams use with customers: in-store sales, returns, exchanges, and customer lookups. Retailers on Shopify often run Shopify POS. Larger chains may run Oracle Retail Xstore or another enterprise POS. For a closer look at the trade-offs, read is Shopify POS right for enterprise retailers.

Enterprise resource planning (ERP). The financial record of the business: purchasing, accounting, the value of inventory on the books, and reporting across legal entities. NetSuite is a common choice for mid-market retailers on Shopify. The ERP should stay the system of record for stock value and finance.

Order management system (OMS). The system that decides how much stock each channel can sell and where each order should be filled. An OMS pulls stock from stores, warehouses, and the ERP into one pool, works out what is safe to sell online, and routes each order to a store or warehouse. Once stores fill online orders through buy online, pick up in store (BOPIS) or ship from store, this is the job that needs a clear owner. Our guide to ERP vs OMS explains where the line between the two sits.

Warehouse management system (WMS). What happens inside the warehouse walls: receiving, putaway, picking, packing, and shipping at volume. Some retailers use the warehouse tools inside their ERP, and some use a third-party logistics (3PL) partner's system.

Supporting systems. These sit around the core and connect to it:

• Returns platforms, such as Loop or AfterShip, for return and exchange requests
• Shipping tools and carriers, such as FedEx, UPS, or a rate-shopping service, for labels and tracking
• Email and marketing, such as Klaviyo, for order updates and campaigns
• Payment providers for card payments and refunds
• Customer service tools for tickets and order lookups

Each system type has its own buying questions. Our list of 10 key retail systems for omnichannel retailers goes deeper on each one.

Which system should own each kind of data

This table is the short version of the whole page. For each kind of data, pick one owner. Other systems can read it or add to it, but only one system decides the number.

Two rows matter most for a retailer whose stores sell online: stock you can sell online, and where each order is filled. Neither the eCommerce platform nor the ERP was built to make those calls across many stores. Those two rows are usually the reason a growing retailer adds an OMS.

There is no single right stack. The right one depends less on revenue than on two things: how many locations hold stock, and whether your stores fill online orders. The store counts below are common patterns, not rules.

A few stores. With one website and a handful of stores, Shopify's own multi-location inventory and local pickup cover most needs, and accounting software handles the books. A store manager can fix a count by phone. Adding systems here mostly adds cost.

A growing chain. Somewhere past the first handful of stores, purchasing, finance, and stock value outgrow spreadsheets and basic accounting. This is where retailers add an ERP such as NetSuite, especially if they sell wholesale or run more than one company or currency. The ERP becomes the record for money and stock value.

Stores that sell online. The bigger change comes when stores start filling online orders. Now the stock in every store is part of what the website can promise, and every online order needs a decision about where to fill it. The storefront can show stock across locations, and the ERP can hold the stock record, but neither works out how much store stock is safe to sell online or which store should take each order. This is the stage where retailers add an OMS. This is also when store teams need simple tools for fulfillment, receiving, and counts, so they aren't working in ERP screens. Our store inventory management guide covers that store-side work step by step.

A retailer can reach this stage with 10 stores or with 100. A brand with eight stores that all fill online orders needs an OMS sooner than a 40-store chain whose stores only sell to walk-in shoppers.

Shopify and NetSuite show up together in many mid-market retail stacks, and for good reasons.

Retailers choose Shopify because the storefront and Shopify POS run on one platform, with one product catalog and one customer record. New storefronts and stores can go live quickly, and a large app ecosystem fills most gaps. Retailers choose NetSuite because it handles financials, purchasing, and reporting across several subsidiaries and currencies in one cloud ERP. Each does its own job well.

The two still need to talk to each other. Most retailers start with a connector, such as Celigo, that moves orders, items, customers, and stock numbers between Shopify and NetSuite. A connector does that job: it passes records from one system to the other.

The gap is in the decisions neither system makes. A connector can copy a stock number from NetSuite to Shopify, but it doesn't work out how much of a store's stock is safe to sell online after holding some back for walk-in shoppers. The connector can also pass an order to NetSuite without deciding which store or warehouse should fill that order. Cases like pickup orders, ship from store, split shipments, kit products, and gift cards tend to fall through to manual work. Our comparison of a standard Shopify NetSuite connector vs an OMS walks through those cases.

Retailers connect their stack in one of three common ways, and many use a mix:

• Direct connectors: One connector for each pair of systems, such as Shopify to NetSuite. Simple to start, though the number of connectors grows with every system you add.
• An integration platform: A middle layer that moves data between all the systems. Flexible, though your team or a partner owns the mapping and the business rules.
• An OMS as the hub for orders and inventory: The OMS connects to the storefront, POS, ERP, and warehouse, and owns the order and stock decisions. Other data, such as marketing or support records, can still flow through connectors.

Whichever pattern you choose, the same things have to move between systems: orders, stock changes, fulfillments, returns, refunds, and payments. Following one ship from store order shows how they fit together:

1. A shopper places an order on the Shopify storefront.
2. The OMS picks up the order, checks it, and routes it to the store best placed to fill it.
3. A store associate picks and packs the order, and a carrier label is printed.
4. The fulfillment goes back to Shopify, and the customer gets a shipping email.
5. The ERP receives the fulfillment and records the invoice and payment.

Timing matters as much as the path. Some flows run as events, the moment something happens. Others run in batches on a schedule. A delay of a few minutes on a finance record rarely hurts anyone, while the same delay on stock for a fast-selling item can sell a unit twice. When you review your stack, ask how often each flow runs and what happens when one fails, not only whether it is connected. Our engineering team explains why retail integrations need pipelines, not just API calls.