An eCommerce ERP integration connects your online shop to the system that runs stock, purchasing, fulfilment and finance, so orders flow into the ERP and stock, prices and tracking flow back without anyone retyping them. There are three ways to build it: a native connector for one shop and one ERP, an integration platform (iPaaS) once three or more systems are involved, or a custom build for a process no connector supports. The choice of connector matters about as much as the choice of ERP, because the connector decides what you pay every month and what breaks at the next platform update.
This guide is for operations and finance leads at online retailers, wholesalers and manufacturers with a webshop who are planning the integration or the ERP behind it. It uses the vendors' own connector documentation and price lists, read on 2 November 2026, the pricing data behind the ERP Pilot comparison, and 230 real quiz runs. If you are still choosing the system rather than the integration, our ranking of the best ERP for eCommerce and retail compares the candidates by fit.
💡 Key Takeaway
Decide data ownership before you pick a connector. Every expensive integration problem, from overselling to a month-end close that takes a week, traces back to two systems that both believed they owned the same record.
When does an online shop need an ERP behind it?
When the shop is no longer the only place where stock, orders and money are recorded. A single store on one channel with ten orders a day runs fine on the shop platform's own tools. The setup starts to fail at predictable points:
- More than two sales channels. Your own shop plus Amazon, eBay or a B2B portal means the same stock is sold in several places. Without one inventory master, overselling is a matter of time, and the usual workaround of showing less stock than you have costs sales.
- Order volume that makes manual entry a job. Once someone spends hours a day copying orders into the accounting package, every copied order is a chance for a wrong amount, a missing tax code or a mismatched address.
- Stock counts you no longer trust. Discrepancies between system and shelf are an operations problem and a finance problem, because inventory valuation and margins are calculated from those counts.
- A month-end close of more than a few days. Reconciling shop payouts, marketplace settlements, shipping costs, returns and VAT across countries by hand is a symptom of fragmented data, not of a weak finance team.
- B2B customers or a second country. Customer-specific prices, payment terms, purchase order numbers, multi-currency and cross-border VAT are ERP territory.
| Signal | Shop tools are enough | ERP recommended | ERP critical |
|---|---|---|---|
| Daily orders | Under 20 | 20 to 100 | Over 100 |
| Sales channels | 1 | 2 to 3 | 4 or more |
| SKU count | Under 100 | 100 to 2,000 | Over 2,000 |
| Warehouses | 1 | 2 | 3 or more |
| B2B sales | None | Starting | Significant |
| Countries | Domestic only | Regional | Several or global |
| Month-end close | Under 2 days | 2 to 5 days | Over 5 days |
The thresholds are ERP Pilot planning guidance from typical selection projects, not hard rules. Three or more rows in the right-hand columns are a reason to start evaluating.
Most teams start a year late
The need for an ERP usually becomes obvious about a year after it arose. By then the workarounds have produced duplicate customers, inconsistent SKUs and spreadsheets nobody wants to give up, and all of that has to be cleaned before any integration can go live.
Who actually asks for integrated eCommerce?
Not only retailers. The ERP Pilot quiz asks every company in its second stage whether it needs integrated eCommerce, meaning a native or tightly integrated storefront with real-time stock and order sync. Between 14 February to 30 October 2026, 230 companies answered that question in the English version, and 83 of them (36%) said yes. Split by industry, manufacturers said yes slightly more often than retailers and wholesalers.
| Said yes | |
|---|---|
| Manufacturers (41 of 89) | 46% |
| Retail, eCommerce and wholesale (18 of 42) | 43% |
| Other industries (18 of 59) | 31% |
| Professional services (6 of 40) | 15% |
Source: ERP Pilot quiz submissions, English version, 14 February to 30 October 2026, every run that answered the eCommerce question (n = 230). Manufacturers = discrete and process manufacturing; trade = retail, eCommerce and wholesale distribution.
That matters for the integration, because a manufacturer's webshop is usually a B2B portal: dealers and distributors ordering at their own price lists, on account, against a purchase order number. Those are exactly the data flows that standard B2C connectors handle worst, and the ones the connector price lists charge most for, as the cost section shows. Professional services firms are the exception: only 15% need a shop at all.
Native connector, iPaaS or custom build?
Pick the simplest option that covers your real data flows. A native connector links one shop platform to one ERP with pre-built mappings for orders, stock, products, customers and fulfilment. An iPaaS (integration platform as a service) such as Celigo, Boomi or MuleSoft sits between several systems and runs the flows from templates plus your own logic. A custom build uses the shop's and the ERP's APIs directly.
| Approach | Setup | Running cost | Time to go-live | Fits |
|---|---|---|---|---|
| Native connector | $5,000 to $15,000 | $0 to a few hundred dollars a month | 2 to 6 weeks | One shop, one ERP, standard B2C flows |
| Middleware / iPaaS | $15,000 to $50,000 | $200 to $1,500 a month | 4 to 12 weeks | Three or more systems, for example shop, ERP, WMS and returns tool |
| Custom API build | $30,000 to $100,000 | Developer time, $10,000 to $30,000 a year | 8 to 20 weeks | A core process no connector supports |
Planning ranges from ERP Pilot for a mid-market project, including configuration and testing by a partner. They are not vendor list prices; connector list prices follow in the next section.
The decision rule is short. One shop and one ERP with standard flows: native connector. A warehouse system, a returns platform, a PIM or several storefronts in the picture: middleware, because point-to-point connectors between every pair become unmanageable. A custom build only when a process that earns you money, such as live configurator pricing from the ERP or your own warehouse allocation logic, is not supported by anything pre-built, and only if you have developers who will still be there in three years. Over-engineering is the more common mistake: middleware bought for a two-system link a native connector would have covered.
Which ERPs ship their own Shopify connector?
Several, and who builds the connector matters more than the feature list, because the builder is the one who fixes it after a Shopify API change. Shopify formalised this with its Global ERP Program, announced on 14 October 2021 with Microsoft Dynamics 365 Business Central, Oracle NetSuite, Infor, Acumatica and Brightpearl as the first partners building direct integrations into the Shopify App Store.
| ERP | Connector | Built by | Price (as published) |
|---|---|---|---|
| Business Central | Shopify Connector | Microsoft | Preinstalled for new sign-ups; Microsoft lists no separate price |
| NetSuite | NetSuite ERP Connector | Oracle NetSuite | $199.92 (B2C Standard), $249.92 (B2C Premium), $916.58 (B2B) per month |
| Acumatica | Native Shopify connector | Acumatica | Not published as a separate list price |
| Odoo | Own eCommerce app instead of a connector | Odoo | Included in the Standard and Custom plans |
Sources: Microsoft Learn, Shopify App Store listing of the NetSuite ERP Connector, Odoo pricing page, all read on 2 November 2026. Oracle notes that external charges may be billed by NetSuite separately from the Shopify invoice.
Three observations from the table. First, Microsoft's Shopify Connector ships inside Business Central and is installed for every new sign-up and trial, which makes it the lowest-friction starting point for a Microsoft shop. Second, the NetSuite connector is priced by scenario, and B2B costs more than four times the B2C Standard plan, because invoice posting, payment handling and B2B order sync are added. Third, Odoo takes the other route: its eCommerce app is a storefront in the ERP itself, so there is no connector to buy or maintain, at the price of running your shop on Odoo rather than on Shopify or Shopware.
Selling with Shopware in DACH
Shopware, built in Germany, is widely used by mid-sized online retailers in Germany, Austria and Switzerland. For Shopware the connector market is more fragmented than for Shopify: check whether yours is built by the ERP vendor, by Shopware or by a third party, who maintains it after a Shopware major release, and whether it covers B2B price lists if you need them.
What does an eCommerce ERP cost in the first year?
For 10 users, between roughly $18,000 and $232,400 in our pricing data, depending on the system and how much implementation help you need. The first year is the expensive one, because implementation lands in it in full. The ranges below are list subscription for 12 months plus the implementation range for 10 users, and for NetSuite also 12 months of the B2C Standard connector ($2,399 a year).
| First year | |
|---|---|
| Odoo Enterprise, shop included: $18,000 to $85,400 | from $18,000 |
| Business Central + Shopify Connector: $34,600 to $163,200 | from $34,600 |
| NetSuite + ERP Connector B2C: $64,300 to $220,400 | from $64,300 |
| Acumatica, connector not included: $71,600 to $232,400 | from $71,600 |
Source: ERP Pilot pricing data and TCO model (list price, 10 users, 12 months of subscription plus implementation). NetSuite includes the ERP Connector B2C Standard plan from the Shopify App Store, 2 November 2026. Acumatica is priced by resources, not users, and its connector is not included. Bars show the low end of each range.
The middle of the market sits close together. What separates the systems is less the licence than the integration that comes with it: Odoo needs no connector if you accept its shop, Business Central brings Microsoft's connector, and NetSuite adds a connector subscription that rises to $10,999 a year on the B2B plan. Per-user prices and what they include are on the pricing pages for Odoo, Business Central, NetSuite and Acumatica, and the ERP price comparison puts all of them side by side. What drives the implementation line is set out in our guide to ERP implementation costs.
Budget the integration as a running cost
Shop platforms, marketplaces and ERPs all ship updates several times a year, and each one can change an API the connector relies on. Whoever owns the connector has to test and adjust it every time. With a vendor-built connector that is mostly the vendor; with middleware or a custom build it is you or your partner, every year.
Which system owns which data?
The ERP owns everything that is counted or booked; the shop owns the moment of purchase. Shopify, Shopware and WooCommerce are storefronts and checkouts. They are not an inventory system of record and not a finance system, and an integration that treats them as one ends in two versions of the truth.
| Data | System of record | Direction of sync |
|---|---|---|
| Product master and prices | ERP (or a PIM) | ERP to shop |
| Stock levels | ERP | ERP to shop, near real time |
| Orders | Shop at capture, ERP afterwards | Shop to ERP |
| Fulfilment and tracking | ERP or WMS | ERP to shop |
| Customer records | Contested: decide explicitly | Usually both ways |
| Payouts and settlements | ERP | Shop to ERP, reconciled |
Two flows deserve the most attention. Stock is the one that prevents overselling: push it from the ERP to every channel by event (webhook) rather than in a nightly batch, and decide which warehouse's stock each channel may sell. Customer records are the one teams forget to decide: B2C customers are usually created in the shop, B2B customers with credit limits and price lists in the ERP. Write the rule down, or the integration will create duplicates in both directions.
Where do eCommerce ERP integrations break?
Rarely at the order sync, which every connector demonstrates well. They break at the edges:
Payout reconciliation sinks more projects than anything else
Shopify Payments, PayPal, Klarna and the marketplaces settle net of fees, on their own schedules and in batches that do not match individual orders. Teams scope the order sync carefully and treat settlements as an afterthought, then close the books by hand for months after go-live. Put settlement reconciliation in the requirements and make the partner demonstrate it with your own payout files.
- Returns and refunds. Partial refunds, exchanges, restocking fees and marketplace-specific return rules are where simple connectors stop. Test the returns flow as thoroughly as the order flow.
- Bundles and kits. A shop product made of several stock items does not map one to one onto an ERP bill of material. If you sell bundles, have the vendor show that exact case with your SKUs.
- Plan limits on the shop side. Shopify's API rate limits vary by plan, and Shopify Plus adds features such as B2B company accounts that a connector may or may not support. Confirm which Shopify plan the connector quote assumes.
- Dirty master data. Inconsistent SKUs, missing weights and duplicate customers are amplified, not fixed, by an integration. Plan a clean-up phase before the first sync.
- Too much scope at once. Start with the core loop of orders, stock and financials, then add product sync, fulfilment details and CRM in later phases. Our ERP implementation timeline shows how the phases are usually sequenced.
What changes when your shop sells B2B?
Almost everything that makes a connector cheap. A B2B portal needs customer-specific price lists, net payment terms, credit limits, purchase order numbers, approval steps and invoices instead of card payments. That is why the NetSuite connector's B2B plan costs $916.58 a month against $199.92 for B2C, and why it is the scenario to test hardest in any demo. Wholesalers face the same list, usually at higher order volumes; the ERP guide for wholesale distribution covers the warehouse and pricing side of that case.
It also changes which ERP fits. We replayed the ERP Pilot scoring engine on 2 November 2026 for two companies with 21 to 100 employees in 2 to 5 countries and changed only one answer: whether they need integrated eCommerce. For an online retailer the answer barely moves the result: NetSuite and Odoo stay tied at the top, and Business Central (130 to 145) and Acumatica (125 to 145) both gain and finish level behind them. For a manufacturer in standard discrete manufacturing the same answer reshuffles the shortlist.
| Score | |
|---|---|
| Acumatica, no eCommerce (rank 1) | 115 |
| Acumatica, integrated eCommerce (rank 1) | 135 |
| Business Central, no eCommerce (rank 4) | 95 |
| Business Central, integrated eCommerce (rank 2) | 110 |
| NetSuite, no eCommerce (rank 7) | 70 |
| NetSuite, integrated eCommerce (rank 4) | 100 |
| Odoo, no eCommerce (rank 9) | 65 |
| Odoo, integrated eCommerce (rank 5) | 95 |
| SYSPRO, no eCommerce (rank 2) | 110 |
| SYSPRO, integrated eCommerce (rank 6) | 95 |
Source: replay of the ERP Pilot scoring engine, 2 November 2026, English set of 20 systems. Baseline: discrete manufacturer, 21 to 100 employees, 2 to 5 countries, one legal entity, cloud, budget $100 per user, priorities total cost, manufacturing and scalability. Only the eCommerce answer differs between the two bars of each system. Scores have no unit.
With integrated eCommerce, NetSuite climbs from rank 7 to 4 and Odoo from 9 to 5, both because they bring their own storefront. Business Central rises from rank 4 to 2 on the strength of Microsoft's own Shopify Connector, with a smaller gain than the systems that contain a shop: it connects to a storefront rather than being one. For a manufacturer that already runs Shopify, that distinction is exactly the integration decision this guide is about. SYSPRO, a strong fit for the manufacturing itself, falls from rank 2 to 6. Acumatica leads in both runs. The manufacturing side of the decision is covered in our ERP guide for manufacturers.
What changes for sellers in the EU and DACH?
Tax determination moves to the customer's country, and German finance teams expect their tax advisor's formats. The integration has to carry the data for both.
- Distance sales and OSS. Once your cross-border B2C sales to EU consumers exceed the EU-wide threshold of EUR 10,000 a year, VAT is due at the customer's rate, declared through one One-Stop-Shop registration. The ERP must determine tax by delivery country, not by your company's location, and produce invoices to match.
- Different rates per country. Germany (19% and 7%), Austria (20%, 13% and 10%) and Switzerland (8.1%, 3.8% and 2.6%) each need their own tax codes on the same product.
- Switzerland is not in the EU. Swiss customers mean customs declarations for goods, CHF, Swiss VAT registration rules and the Swiss data protection act, not just another country code.
- DATEV. Most German companies close their books with a tax advisor who works in DATEV. Microsoft documents the GoBD audit export and ELSTER VAT advance notifications for the German version of Business Central, but no DATEV export among them. For Business Central and every other system on the list, ask who supplies the DATEV export and whether it is part of the standard product or a partner add-on.
- B2B e-invoicing in Germany. Invoices from a German B2B shop have to be issued as structured e-invoices under the German e-invoicing mandate. The ERP produces them, not the shop, so the integration must hand over every field the invoice needs.
- Packaging law. Online sellers in Germany must register with the Central Agency Packaging Register (LUCID) and report packaging volumes. The figures come from your shipments, so the ERP has to record packaging per order.
💡 Key Takeaway
Choose the integration by who maintains it, not by the demo. A vendor-built connector for one shop and one ERP, middleware from three systems up, custom code only for a process that earns money. And if you sell to businesses, test the B2B flows first: 46% of the manufacturers in our quiz data asked for integrated eCommerce, and B2B is where connectors charge the most.

