How does ml-connector handle Sage 100's lack of a cloud API and JAGGAER's cloud-only nature?

ml-connector connects to Sage 100 via SOAP Web Services if enabled, or through a local BOI COM agent running on the customer's Sage 100 server. The agent is a Windows service that ml-connector can call over your network, allowing Sage 100 data to be read and written without moving Sage 100 itself to the cloud. JAGGAER is accessed via REST API over the internet, so ml-connector handles the network difference transparently.

Does the integration work with Sage 100's limited SOAP surface that covers only sales orders and AR?

Yes. If the customer has only SOAP Web Services enabled in Sage 100, ml-connector can read and write sales orders and customer data. For full access to purchase orders, AP invoices, vendors, and GL accounts, the customer must deploy a local BOI COM agent on the Sage 100 server, which ml-connector can then call to access those modules. Your implementation partner will assess which approach fits your Sage 100 setup.

What happens if the local Sage 100 BOI agent is offline or unreachable?

If the agent is unavailable, ml-connector will retry with backoff for the configured interval and then surface the error so your team can restore connectivity. Invoices and purchase orders already in JAGGAER will be imported once the agent comes back online, since ml-connector tracks which records have been processed. A full audit trail is maintained so you can replay imports if needed.

Sage 100JAGGAER

Sage 100 and JAGGAER integration

Sage 100 runs finance and purchasing on-premises. JAGGAER runs procurement and sourcing in the cloud. Connecting the two keeps your purchase orders and invoices synchronized across the organization without manual re-entry. New purchase orders created in JAGGAER flow into Sage 100's AP module, invoices are matched and posted to GL accounts, and vendor master records stay current on both sides. ml-connector handles the distinct APIs and authentication on each platform and moves data on a schedule you define.

How Sage 100 works

Sage 100 is deployed on-premises on a Windows server and exposes purchasing, accounts payable, inventory, and general ledger data through SOAP Web Services (eBusiness Web Services) or through a local BOI COM agent that ml-connector's bridge can call. Sage 100 authentication is stateless - a username and password are passed with every SOAP call, and there are no tokens or OAuth flows. The system has no webhooks or event stream, so data must be polled. Key entities include purchase orders, vendors, AP invoices, GL accounts, and items. The SOAP surface is limited to sales orders and customers, so full access to AP, GL, and purchasing requires the BOI COM layer wrapped by a local agent on the customer's Sage 100 server.

How JAGGAER works

JAGGAER is a cloud Source-to-Pay platform that exposes purchase orders, invoices, suppliers, contracts, and catalog items through the J1P REST API, which requires OAuth 2.0 client credentials (client_id and client_secret provisioned during onboarding). JAGGAER also supports cXML for bidirectional supplier document exchange and SAP iDoc for ERP synchronization. The J1P API is pull-only - JAGGAER can push events to a registered endpoint, but event-driven configuration is not self-serve and requires JAGGAER professional services. JAGGAER does not publish a public Swagger or OpenAPI spec; detailed endpoint paths are gated behind customer login and provided during onboarding.

What moves between them

The integration flows from JAGGAER into Sage 100. After a purchase order is created or updated in JAGGAER, ml-connector reads it from the J1P API and writes it into Sage 100 as a PO, preserving the vendor reference and line-item detail. When JAGGAER invoices are received, they are fetched via J1P and posted into Sage 100's AP module, matched to the corresponding purchase order, and routed to the correct GL accounts based on the line-item category. Vendor master records are synchronized in both directions so Sage 100's vendor table reflects JAGGAER's current supplier roster. GL accounts are mapped so invoice posting lands on the correct dimension in Sage 100.

How ml-connector handles it

ml-connector polls JAGGAER's J1P REST API using OAuth 2.0 client credentials to fetch purchase orders, invoices, and supplier changes. On the Sage 100 side, ml-connector connects via SOAP Web Services if the customer has enabled it, or through a local BOI COM agent running on the Sage 100 server if full AP and GL access is needed. Sage 100 authentication is handled per-call with the credentials provided during setup - there are no tokens to refresh. ml-connector polls on a cadence you define, typically every 15 minutes for invoices and purchase orders and every hour for vendor changes. Because Sage 100 is on-premises, the local BOI agent or SOAP endpoint must be reachable from your network or a customer-hosted relay. ml-connector tracks update timestamps on both sides to avoid duplicate imports and maintains a full audit trail of every record synchronized so failed imports can be replayed or investigated.

A real-world example

A mid-market manufacturing company runs Sage 100 for accounting and inventory on-premises and uses JAGGAER as its Source-to-Pay platform across its three supply chain teams. Before the integration, procurement created purchase orders in JAGGAER, and the accounting department re-entered those orders manually into Sage 100 once they were approved. Invoices came back from suppliers through JAGGAER's portal, and the AP team matched them against the Sage 100 purchase orders by hand, a process that introduced errors and delayed posting. With Sage 100 and JAGGAER connected, new purchase orders flow automatically from JAGGAER into Sage 100, invoices are matched and posted without manual intervention, and vendor changes in JAGGAER are reflected immediately in Sage 100's master file. The AP team now spends its time on exception handling and compliance, not data entry.

What you can do

Read purchase orders from JAGGAER and write them into Sage 100 as POs, preserving vendor and line-item detail.
Fetch invoices from JAGGAER and post them into Sage 100's AP module, matched to the corresponding purchase order.
Synchronize vendor master records in both directions so Sage 100's supplier table reflects JAGGAER's current roster.
Authenticate JAGGAER with OAuth 2.0 client credentials and Sage 100 with SOAP Web Services or a local BOI COM agent.
Poll on a configurable schedule with a full audit trail on every record, supporting replays if a downstream sync fails.

Questions

How does ml-connector handle Sage 100's lack of a cloud API and JAGGAER's cloud-only nature?: ml-connector connects to Sage 100 via SOAP Web Services if enabled, or through a local BOI COM agent running on the customer's Sage 100 server. The agent is a Windows service that ml-connector can call over your network, allowing Sage 100 data to be read and written without moving Sage 100 itself to the cloud. JAGGAER is accessed via REST API over the internet, so ml-connector handles the network difference transparently.
Does the integration work with Sage 100's limited SOAP surface that covers only sales orders and AR?: Yes. If the customer has only SOAP Web Services enabled in Sage 100, ml-connector can read and write sales orders and customer data. For full access to purchase orders, AP invoices, vendors, and GL accounts, the customer must deploy a local BOI COM agent on the Sage 100 server, which ml-connector can then call to access those modules. Your implementation partner will assess which approach fits your Sage 100 setup.
What happens if the local Sage 100 BOI agent is offline or unreachable?: If the agent is unavailable, ml-connector will retry with backoff for the configured interval and then surface the error so your team can restore connectivity. Invoices and purchase orders already in JAGGAER will be imported once the agent comes back online, since ml-connector tracks which records have been processed. A full audit trail is maintained so you can replay imports if needed.

Connect Sage 100 and JAGGAER

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

Get started