Does the integration support both Sage X3 REST and GraphQL endpoints?

Yes. ml-connector can authenticate to both REST (HTTP Basic Authentication) and GraphQL (OAuth2 via Connected Application) endpoints. GraphQL is recommended if your Sage X3 instance is V12 or later. Either way, the same purchase orders, invoices, and supplier data flow into Procore using delta polling on the updatedDate field.

What happens when a Sage X3 purchase order is amended or cancelled?

ml-connector detects the change on the next poll and updates the corresponding Procore commitment to reflect the new line items, quantities, or amounts. If a PO is cancelled in Sage X3, ml-connector can mark the commitment as inactive in Procore. Every change is logged in the audit trail so you can trace when and why a commitment changed.

How does the integration handle Sage X3 GL accounts mapping to Procore cost codes?

GL accounts in Sage X3 are matched to Procore cost codes by a configurable mapping you define at setup. When a purchase order line in Sage X3 references a GL account, ml-connector looks up the matching cost code and applies it to the commitment line in Procore. If the GL account does not have a mapped cost code, the sync is flagged in the audit log so you can add the mapping before the next poll.

Sage X3Procore

Sage X3 and Procore integration

Sage X3 runs your financial operations, procurement, and inventory. Procore tracks project commitments, costs, and field operations. Connecting the two keeps your project budget aligned with actual purchase orders and supplier invoices, and ensures cost codes flow to both systems consistently. Purchase orders created in Sage X3 feed into Procore as commitments so project managers see the full financial picture in real time, and invoice data flows the same direction so project costs match vendor records.

How Sage X3 works

Sage X3 exposes suppliers, purchase orders, supplier invoices, sales invoices, products, GL accounts, and GL entries through REST and GraphQL endpoints. REST uses HTTP Basic Authentication against a customer-specific server URL and port; GraphQL uses OAuth2 client credentials via a Connected Application. Access tokens expire in 5 minutes with 30-day refresh token validity. Sage X3 has no native webhooks or event push, so records are read by polling using updatedDate and modifiedDateTime fields to detect changes. On-premise is the dominant deployment; cloud SaaS requires Sage-managed hosting.

How Procore works

Procore exposes vendors, purchase order contracts, requisitions, payment applications, budget line items, cost codes, and change orders through REST APIs secured with OAuth2 Client Credentials. All endpoints require a company_id; project-scoped endpoints also require a project_id. Tokens expire in 5400 seconds (1.5 hours). Procore supports webhooks for real-time push of create, update, and delete events on commitments, requisitions, direct costs, and budget. Webhooks must be publicly accessible HTTPS and return 2xx within 5 seconds. No HMAC signature is used; security relies on custom headers in the destination.

What moves between them

The main flow is Sage X3 into Procore. Purchase orders and supplier invoices from Sage X3 flow into Procore as commitments and requisitions, each tagged with the cost code matching the GL account in Sage X3. Supplier reference data syncs bidirectionally so vendor records are consistent. Cost codes and budget allocation dimensions sync from Procore into Sage X3 to ensure that when commitments are created, they map to valid GL accounts. GL entries themselves remain read-only; ml-connector does not create accounting entries in either system.

How ml-connector handles it

ml-connector polls Sage X3 on a schedule you define, using the updatedDate field to fetch only changed records since the last poll. It detects which Sage X3 purchase order lines map to Procore cost codes by matching the GL account from Sage X3 to the cost code in Procore. On the Procore side, it uses OAuth2 client credentials to authenticate and passes the required company_id and project_id on every request. When a new purchase order appears in Sage X3, ml-connector creates a commitment in Procore with line items for each cost code; supplier invoice totals flow as requisitions. If a Sage X3 PO is amended, the commitment is updated in Procore. Procore tokens expire every 1.5 hours, so ml-connector refreshes the bearer token on each call and backs off on 429 responses. Every record carries a full audit trail, and if a Procore API call fails, the full transaction can be replayed once the issue is resolved.

A real-world example

A mid-sized commercial contractor runs Sage X3 for corporate accounting and vendor management, and uses Procore as the system of record for project financials across 20 active construction projects. Before the integration, the accounting team exported purchase order registers from Sage X3 weekly and manually entered them into Procore as commitments, a process that introduced lag and data-entry errors. Project managers could not see the full committed cost picture without checking Sage X3 separately. With Sage X3 and Procore connected, every PO created by procurement appears in Procore within minutes, allocated to the correct project and cost code, and project managers see real-time commitment totals against budget. Month-end close is faster because committed costs are already in Procore and reconciliation with Sage X3 requires no manual verification.

What you can do

Sync purchase orders from Sage X3 into Procore as commitments, with line items allocated to the matching cost codes.
Flow supplier invoice totals from Sage X3 into Procore as requisitions, keeping actual costs in sync with commitments.
Keep supplier and cost code reference data consistent in both directions so allocations land on valid GL accounts and cost codes.
Authenticate Sage X3 via GraphQL OAuth2 or REST HTTP Basic Auth, and Procore via OAuth2 Client Credentials, with automatic token refresh.
Poll Sage X3 on a schedule you control, with full audit trails and replay capability if a sync step fails.

Questions

Does the integration support both Sage X3 REST and GraphQL endpoints?: Yes. ml-connector can authenticate to both REST (HTTP Basic Authentication) and GraphQL (OAuth2 via Connected Application) endpoints. GraphQL is recommended if your Sage X3 instance is V12 or later. Either way, the same purchase orders, invoices, and supplier data flow into Procore using delta polling on the updatedDate field.
What happens when a Sage X3 purchase order is amended or cancelled?: ml-connector detects the change on the next poll and updates the corresponding Procore commitment to reflect the new line items, quantities, or amounts. If a PO is cancelled in Sage X3, ml-connector can mark the commitment as inactive in Procore. Every change is logged in the audit trail so you can trace when and why a commitment changed.
How does the integration handle Sage X3 GL accounts mapping to Procore cost codes?: GL accounts in Sage X3 are matched to Procore cost codes by a configurable mapping you define at setup. When a purchase order line in Sage X3 references a GL account, ml-connector looks up the matching cost code and applies it to the commitment line in Procore. If the GL account does not have a mapped cost code, the sync is flagged in the audit log so you can add the mapping before the next poll.

Connect Sage X3 and Procore

Free to use. Add your credentials, ping your real systems, and see if we fit.

Get started