Explainer

Multi-entity management: how it works and what to require

In short

Proconomy models a holding company and its direct subsidiaries as separate organisations. Each one configures its own approval routing, sourcing policy, supplier document requirements and locations, and can read the holding company’s shared code lists, contract clauses and templates rather than rebuilding them. The holding company gets registers across the subsidiaries for contracts, obligations, suppliers, parts and purchase orders, and a card for each company.

See Multi-entity management

What each company keeps. And what the group writes once.

The division that decides whether a group platform survives its second company.

Configured by each company

  • Its approval routing, as a graph with its own conditions
  • The value above which a sourcing event becomes mandatory
  • The minimum number of bids an event has to attract
  • How a supplier off the approved list is handled
  • The award mode, by category or by spend type
  • Which documents and certifications a supplier must provide

Also its own

  • Its legal entity name, currency, timezone and financial year start
  • Its plants, warehouses, sites, storage locations and docks
  • The address, manager, currency and timezone of each one
  • Its cost centres, and which location each belongs to
  • Which locations require an approval step on an intake request
  • Its own users and their approval limits

Written once by the group

  • Shared code lists, read by the subsidiaries
  • The contract clause library
  • Contract templates
  • A supplier master that works in both directions
  • Industry defaults that preload modules on creation
  • Compliance settings that come with those defaults

Visible across the subsidiaries

  • Contracts and their obligations
  • Suppliers, parts and purchase orders
  • Part-approval submissions and bills of materials
  • Worklists and service entry sheets
  • A card per subsidiary: open purchase order value, pending requisitions
  • Each entity’s spend beside the others, top suppliers merged

Configuration where it belongs. Libraries where they help.

Six properties, and the first one is the whole argument.

MechanismWhat it has to do
Each company is its own organisationA holding company and its direct subsidiaries, each with its own legal entity name, currency, timezone and financial year start — not departments of a single configuration wearing different labels.
Its administrators configure itApproval routing, sourcing policy, supplier document requirements, locations and cost centres belong to the organisation they apply to, and changing one changes that one.
Routing on the conditions that matterValue, category and its parent categories, requester role, urgency and emergency, budget utilisation, sourcing method, line type, contract type and contingent labour — and, on an intake request, new supplier, risk tier, cross-border, renewal and data classification.
Libraries the group writes onceShared code lists, the contract clause library and contract templates are read by a subsidiary rather than rebuilt, and the supplier master works in both directions.
Places that transactions point atPlants, warehouses, sites, storage locations and docks, each with an address, a manager, a currency and a timezone, referenced by requisitions, orders, receipts, invoices, budgets and cost centres.
A live view across the subsidiariesGroup registers for contracts, obligations, suppliers, parts and purchase orders, and a card for each company — the same screens the companies themselves use, not a pack assembled at month end.

A description of what the practice requires, not a feature list.

Same structure. Different reasons companies differ.

Why a group ends up with companies that do not buy alike.

Engineered parts

Long qualification lists and a low threshold for going to market, because the cost of the wrong supplier is a line stoppage.

Services and contingent labour

A contingent-labour condition on the routing, a shorter document list, and a different award mode.

Recently acquired

Starts from the group’s code lists, clauses and templates rather than a blank page, and is configured for how it already buys.

Maintenance and site upkeep

A higher mandatory-event threshold, and locations that matter more than categories do.

Five things to ask for. Configured live, in the session.

Each one is quick, and none of them can be staged in advance.

  1. Have two subsidiaries configured differently in front of you — different thresholds, different routing conditions, different document lists — and check that changing one leaves the other alone.
  2. Ask what a new company inherits from the group on day one, and what it has to set up itself. Both answers should be specific.
  3. Open the group contract register and go into an individual contract from it.
  4. Ask which conditions an approval can actually route on, and expect a list rather than "anything you like".
  5. Ask how the group view handles companies reporting in different currencies, and listen to the whole answer.

Definitions. Asked and answered.

It is how a group of companies runs on one platform without being forced into one process. The holding company and each of its direct subsidiaries is its own organisation, with its own legal entity name, currency, timezone, financial year, approval routing, sourcing policy, supplier requirements and locations. The holding company gets a view across the subsidiaries.

Each company. Approval routing, sourcing policy, supplier document requirements, locations and cost centres belong to the organisation they apply to. What the group provides instead of a mandate is a set of libraries — shared code lists, the contract clause library, contract templates and a supplier master — that a subsidiary reads rather than rebuilds.

Registers across the direct subsidiaries for contracts and their obligations, suppliers, parts, purchase orders, part-approval submissions, worklists and bills of materials. A group dashboard carries a card for each subsidiary with its open purchase order value and pending requisitions, and analytics puts each entity’s spend next to the others with top suppliers merged. From the group contract register you open the contract itself.

It becomes another organisation in the group, set up the way any company is — its own routing, policy, requirements and locations — and it reads the holding company’s libraries from the start. Once it is linked to the holding company it appears in the group pages immediately, with zeros against it until it has activity of its own.

Connectors are configured per organisation, and one organisation can hold more than one, each with its own address and credentials. A group running different systems in different companies sets each connector up under the company it belongs to, which is also what keeps the data attributed to the right one.

It is running several companies on one procurement platform without forcing them through one process. Each company configures its own approval routing, sourcing policy and supplier requirements, while a holding company can see across them. The useful question for any vendor is narrow: what exactly does a parent impose on a subsidiary, and what does a subsidiary simply read if it wants to?

It is the obvious design and it rarely survives the second company. A policy written for one operation becomes something every other company has to be an exception to, and exceptions are where governance quietly stops meaning anything. The alternative is to give each company its own configuration and give the group libraries worth reading — code lists, clauses, templates, a supplier master — so that the centre is useful rather than imposing.

Enough to answer a question without starting a reporting round: the contracts and obligations in force, the suppliers and parts in use, the purchase orders open, and each company’s position beside the others. Ask specifically which modules the group view covers, because it is rarely all of them, and ask what happens when companies report in different currencies.

See it on your own workflow. Not a prepared scenario.

Reference material explains the model. A demonstration on one of your own processes is what settles the internal argument.

Someone from client success replies, not a sales sequence. If we are not a fit we will say so on the first call.