EAFlow
Book demo

ANONYMIZED CASE · CPG

Multi-actor service desk on the graph, with self-registered external suppliers

A leading mass consumer goods organization validated Supplier Service Desk by migrating its service desk with external suppliers from a legacy low-code platform to EAFlow: external-supplier self-registration, multi-role queue with a DRAFT → ASSIGNED → RESOLUTION → CLOSE state machine, graph-assisted routing by legal entity and category, audited bulk cleanup of phantom tickets and aging and SLA reporting.

The validation showed that the solution works on the client's operation and the value it delivers: a multi-actor desk governed on the graph, with self-registered suppliers, assisted routing and a cleaned-up queue whose aging and SLA reporting makes sense again.

See the supplier-service-desk solution Operational validation

The challenge

In a consumer-goods manufacturer that runs a service desk with external suppliers, that desk usually lives on a legacy low-code platform. The problem is not the lack of a desk: it is that the desk lives trapped in proprietary platform structures, isolated from the rest of the operation.

  • The desk is locked inside the legacy LCDP: ticket types, service-center roles, per-area focal points and the state machine sit in proprietary structures; cost grows and upgrades are risky.
  • The supplier relationship starts by mail: to onboard an external supplier, someone sends an email and records it in a local spreadsheet, with no link between the supplier and its legal entity in a governed model.
  • Routing is by hard-coded rules or by mail: deciding who handles each ticket by legal entity and category is done with rigid rules or with emails that get lost between areas.
  • Phantom tickets pile up: opened in the system but resolved by email, they are closed one by one or never; the queue ages and reporting loses meaning.
  • The operation lives in a saturated inbox and supplier data is requested twice.

The service desk is not sustained when locked in. It is sustained by tickets that are governed and connected to the legal entity, the category and the focal point, with assisted routing and traceable evidence.

The EAFlow solution

Supplier Service Desk is a CPG vertical accelerator built on the cross-cutting solution Operational Graph for Service Operations plus Workflow + federated identity and the common Operational Graph layer of EAFlow Platform, operating with multi-actor federated identity for internal users and self-registration for external suppliers. The validation covered, on the client's service desk:

  • Migration from the legacy low-code platform, preserving the multi-actor model. Ticket types, service-center roles, per-area focal points and the state machine that lived on the legacy low-code platform were migrated to EAFlow, preserving the intake model, without rewriting the process from scratch.
  • Each ticket connected to the legal entity, the category and the focal point. Originator, ticket, category, legal entity, CSC role and area focal point become first-class entities of the Operational Graph.
  • External-supplier self-registration: the supplier registers from the portal, receives credentials by email and is linked to its legal entity in the graph, with no mail to the service center and no local spreadsheet.
  • Multi-role queue with a state machine DRAFT → ASSIGNED → RESOLUTION → CLOSE: Supplier, Internal User, L1 Agent, L2/L3 Focal Point and Power User see the same queue from their role; each transition signed by its author.
  • Graph-assisted routing: the graph knows the legal entity, the category and the applicable focal point, and suggests type and recipient to the originator, with no hard-coded per-country rules, with governed agents and human-in-the-loop.
  • Controlled bulk-cleanup procedure: phantom tickets are resolved with prior diagnosis, audited bulk update and subsequent verification, instead of being closed one by one.
  • Operational aging and SLA reporting: queue dashboard, aging by category, within-SLA closure rate and supplier-data freshness.
  • Governed querying over the ticket corpus: tickets, states, categories and focal points are answered on the graph, citing the ticket and the node, always with governed agents and human-in-the-loop.

Category coverage, per-area focal points and supplier onboarding are established by a scope agreed during discovery. The accelerator runs the multi-actor desk; it coexists with the client's corporate ITSM tool where one exists, without replacing it.

What was validated

The exercise ran on the client's service desk with external suppliers, migrated from the legacy low-code platform. The service center walked the full cycle: migration of the multi-actor model, supplier self-registration, multi-role queue with a state machine, graph-assisted routing, audited bulk-cleanup procedure, aging and SLA reporting and governed querying — with the roles of Supplier, Internal User, L1 Agent, L2/L3 Focal Point and Power User on the same governed desk.

Demonstrated capabilities

  • Operational Graph as the common context layer.
  • Migration of the multi-actor model from the legacy low-code platform, preserving structure.
  • Traceability originator ↔ ticket ↔ category ↔ legal entity ↔ CSC role ↔ focal point.
  • External-supplier self-registration linked to its legal entity (multi-actor federated identity).
  • Multi-role queue with a state machine signed by author.
  • Graph-assisted routing, with governed agents and human-in-the-loop.
  • Controlled audited bulk-cleanup procedure.
  • Aging and SLA reporting with supplier-data freshness.
  • Natural-language querying over the ticket corpus, citing the source.

Observed outcome

The validation made it possible to verify the use of EAFlow to govern the multi-actor service desk: moving it from being trapped on a legacy low-code platform to operating governed on the graph, with self-registered suppliers and assisted routing. The supplier relationship stopped starting by mail and moved to a self-registration linked to its legal entity; phantom tickets stopped being closed one by one and moved to an audited bulk-cleanup procedure; and the operation stopped living in a saturated inbox to rely on aging and SLA reporting.

The validation confirmed that the accelerator migrates and governs the multi-actor service desk preserving the client's intake model, with federated identity for internal users and self-registration for external ones.

Why it matters for other organizations

The pattern repeats across CPG organizations that serve external suppliers: the desk lives on a legacy low-code platform, the supplier relationship starts by mail, routing is rigid and phantom tickets age the queue. Migrating the desk to an accelerator on the Operational Graph — preserving the multi-actor model, without rewriting from scratch or replacing the ERP — reduces maintenance cost and restores meaning to reporting.

Starting with a bounded multi-actor desk is also a low-risk point of entry: the same Operational Graph that governs the supplier desk later sustains a multi-domain service operation, contracts and per-supplier compliance.

How it scales — related solutions

The governed desk and its ticket corpus are reused on the same Operational Graph:

Evaluate my use case See the supplier-service-desk solution