Shopify ERP Integration: The Complete 2026 Guide
A Shopify ERP integration connects your storefront to the system that runs your business, so orders, inventory, pricing and customers stay in sync without anyone re-keying data. A standard project takes six to fourteen weeks and costs between £15,000 and £60,000 in the UK. Complex builds with B2B, multi-warehouse or EDI go considerably higher.
Most integrations don’t fail because the API is difficult. They fail because nobody agreed what should happen when reality doesn’t match the happy path.
The three ways to connect Shopify to an ERP
Native connector. A purpose-built app for one specific Shopify-to-ERP pairing, maintained by the ERP vendor or a specialist. Shopify’s Global ERP Program certifies connectors that meet its standards for security and API maintenance — as of 2026 that programme includes Oracle NetSuite, Microsoft Dynamics 365, Acumatica, Infor and Brightpearl. Notably, Sage 100 and 300 are not in it, which doesn’t rule Sage out but does mean the maintenance responsibility sits with a middleware vendor rather than the ERP vendor.
Cost: roughly £6,000–£16,000 to implement, plus licensing. Best for: standard flows on a supported ERP, where your business rules are conventional.
iPaaS. A middleware platform — Celigo, Boomi, Workato, or one of the UK-focused managed platforms — with prebuilt content you configure into flows. Faster than building from scratch, vendor-maintained, and flexible enough to extend.
Cost: £15,000–£40,000 implementation, plus £500–£1,500 a month in platform licensing. Best for: mid-market brands wanting speed and a supported platform, with some non-standard logic.
Custom middleware. A service you own — typically Node.js — sitting between Shopify and the ERP, transforming and routing data exactly as your business requires.
Cost: £20,000–£60,000, more for complex scope. No ongoing licence. Best for: logic that connectors can’t express. Case-pack to unit conversion, unusual B2B pricing, multi-warehouse routing rules, legacy ERPs with no connector ecosystem.
The honest framing: a connector is cheapest until your business does something the connector’s authors didn’t anticipate, at which point it becomes the most expensive option because you’re now working around it forever. Custom is most expensive in year one and cheapest in year five — provided it’s documented well enough to survive the person who built it.
The hybrid most mid-market brands land on: iPaaS or a connector for the standard 70% of flows, plus a thin custom service for the 30% that contains your actual competitive logic.
What moves between the systems
ERP → Shopify: product data and attributes, inventory levels by location, price lists and customer-specific pricing, customer and company records for B2B.
Shopify → ERP: orders, customers, refunds and returns, payment and settlement data.
Both directions: fulfilment status, tracking numbers, stock adjustments.
Sounds simple written down. The complexity is entirely in the edge cases.
The edge cases that decide whether it works
This is the section most guides skip, and it’s the section that determines whether your integration survives contact with a real trading week.
Partial fulfilment. An order has three lines; two are in stock. The ERP ships two and the third goes on back order. Your integration needs to represent that in Shopify correctly, or your customer gets a “delivered” email for something still sitting in the warehouse.
Case packs and units. You sell wholesale by the case; the warehouse picks by the unit. Somebody has to translate — and if it’s a person doing it by hand, order by order, you’re paying for a problem that should have been solved in code. Mapping every wholesale SKU to its single-unit equivalent so orders arrive at the warehouse already in the right language is one of the highest-value things an integration does.
Inventory contention. Two channels sell the last unit within the same sync window. Whichever system you treat as the source of truth for stock, decide it explicitly and enforce it — don’t let it emerge by accident.
Order edits after the fact. Customer service edits an order in Shopify that’s already in the ERP. Does the change propagate? Does it create a duplicate? Most integrations get this wrong on the first pass.
Refunds and partial refunds. Financially the most consequential flow, and routinely the last one built.
Retry and failure handling. Your ERP will go down for maintenance. The API will rate-limit you. A payload will be malformed. Without a retry queue and a dead-letter path, those failures become silently missing orders — usually discovered days later during a finance reconciliation.
Reconciliation. A scheduled job that compares both systems and reports drift. Not optional. Every integration drifts eventually; the only question is whether you find out from a report or from a customer.
If your integration partner’s proposal doesn’t mention retries, idempotency, an audit trail and reconciliation, they’re quoting for the happy path only, and the real cost will arrive later.
Costs and timelines by ERP
| ERP | Typical UK project cost | Timeline |
|---|---|---|
| Xero / QuickBooks | £4,000–£25,000 | 2–6 weeks |
| Brightpearl | £12,000–£45,000 | 4–10 weeks |
| Acumatica | £16,000–£90,000 | 6–14 weeks |
| Dynamics 365 Business Central | £20,000–£110,000 | 6–16 weeks |
| Oracle NetSuite | £24,000–£150,000 | 8–20 weeks |
| SAP Business One / S/4HANA | £40,000–£220,000 | 12–28 weeks |
Add four to six weeks for multi-entity, multi-currency, or running B2B alongside DTC. Scope drives cost far more than the ERP does — but the ERP sets the floor.
How to run the project
Discovery first, and take it seriously. The single largest cause of timeline overrun is inadequate requirements work at the start. Map every field, both directions, with the person who processes orders in the room — not just the systems owner. The undocumented workaround someone has been doing manually for three years is exactly the thing that breaks the launch.
Define the source of truth per data type. Inventory, pricing, customer records, product data. Each one has exactly one owner system. Write it down.
Build the observability before the features. Logging, an audit trail of every message, alerting on failure. You cannot debug an integration you can’t see into, and you will need to debug it.
Sync deltas, not full datasets. Full syncs are simple to build and will eventually take longer than the interval between them.
Run parallel before you cut over. Two weeks of both systems running with reconciliation reports catches the discrepancies that testing doesn’t.
Frequently asked questions
How long does a Shopify ERP integration take? Six to fourteen weeks for a standard project on NetSuite, Business Central or Acumatica. Faster for Xero or QuickBooks. Add four to six weeks for multi-entity, multi-currency or B2B alongside DTC.
What does it cost? £6,000–£16,000 for a native connector implementation, £15,000–£40,000 plus licensing for iPaaS, £20,000–£60,000 for custom middleware. Complex builds involving EDI or unusual pricing exceed those ranges.
Do I need Shopify Plus for ERP integration? No. Standard plans expose the same Admin API. Plus matters when you also need B2B, custom Functions or checkout customisation alongside the integration.
Connector or custom — which should I choose? Start by listing the things your business does that a generic connector wouldn’t anticipate. If that list is short, buy a connector. If it includes case-pack conversion, customer-specific pricing rules, or multi-warehouse routing logic, you’ll be building custom eventually — better to know that at the start.
Can I integrate Shopify with Sage? Yes, via middleware or a third-party connector. Sage 100 and 300 aren’t in Shopify’s Global ERP Program, so maintenance responsibility sits with the middleware vendor rather than Sage. Sage 200, Intacct and X3 are commonly connected through iPaaS.
What is iPaaS? Integration Platform as a Service — middleware such as Celigo, Boomi or Workato that provides prebuilt connectors and a environment for building flows. More flexible and more expensive than a single-purpose connector.
How do I keep inventory accurate across channels? Designate one system as the source of truth, sync deltas frequently, and run a daily reconciliation job that reports drift. Real-time sync in both directions without a defined owner is how stores oversell.
What happens if the ERP goes down? A properly built integration queues messages and retries with backoff, then alerts a human if retries exhaust. Without that, orders placed during the outage are simply lost.
Can I integrate a warehouse system as well? Yes, and it’s common — ERP for finance and stock, WMS for picking and dispatch. It adds a third system to the mapping exercise and typically four to eight weeks.
Should the integration live in an app or a separate service? For anything beyond simple flows, a separate service. It’s easier to monitor, scale, version and test than logic embedded in a Shopify app, and it isn’t coupled to Shopify’s app lifecycle.
Is a custom integration risky if the developer leaves? Only if it’s undocumented. Insist on repository ownership, written architecture documentation and a runbook as contractual deliverables. That’s the difference between an asset and a liability.
Last reviewed July 2026. Shopify Global ERP Program membership should be confirmed against Shopify’s current documentation.
Connecting Shopify to your ERP or warehouse system? We build custom middleware for UK brands — including case-SKU to unit mapping, B2B pricing sync and fulfilment automation — with retries, audit trails and reconciliation built in from the start. Get in touch.