Entity as a first-class dimension
Not a separate company file. Every transaction carries its entity, so per-entity and consolidated views are the same query with a different scope.
Platform · financial core
The reliable indicator that a consolidation is being assembled rather than computed is the intercompany plug — the journal that makes the elimination balance because the two sides never matched. It only ever grows, and removing it means identifying intercompany at posting rather than reconciling it monthly.
184 consecutive days tied · variance $0.00
What it does
Not a separate company file. Every transaction carries its entity, so per-entity and consolidated views are the same query with a different scope.
Marked as intercompany when it is created, with its counterparty entity, rather than inferred at consolidation from account codes and hope.
Matched pairs eliminate automatically. Unmatched intercompany becomes a gated exception before consolidation rather than a plug afterwards.
Functional and reporting currency handled distinctly, with cumulative translation adjustment posting to equity rather than being absorbed somewhere convenient.
Minority interest and proportional consolidation, with equity-method investments carried as a single line rather than a full roll-up.
From a consolidated line to the contributing entity to the transaction. A consolidated figure that cannot be traced is a number a board will not trust twice.
In most mid-market groups intercompany activity is recorded informally — a management fee posted in the parent, a recharge entered in the subsidiary next month at a slightly different amount, a cost shared without anyone recording the other side. At consolidation the two sides do not net, and the difference gets plugged.
That is not carelessness; it is what happens when intercompany is a category you recognise afterwards rather than a property recorded at the time. The fix is structural: mark it when it is posted, with its counterparty, so the matching is arithmetic rather than archaeology.
The highest-value decision in a multi-entity implementation is whether entities share a chart of accounts or keep their own with a mapping. Both work. The failure mode is not choosing, because charts then drift apart and mapping happens ad hoc at consolidation, so no two months are comparable.
For most groups we recommend a shared chart with entity as a dimension, and local statutory differences handled by mapping at the reporting layer. More work in month one, and it removes a recurring reconciliation permanently.
Entities close on their own schedule with their own checklists, rolling into a consolidated close where intercompany elimination is a gated task. The group view is not produced until elimination passes, and what is blocking it is visible rather than being something the group controller has to go and find out.
Limits
We handle multi-currency and multi-entity structure. Statutory filing requirements across many jurisdictions is two decades of accumulated work we have not done, and it points to NetSuite.
Comfortable below that. Larger groups, particularly with many jurisdictions, are better served elsewhere and we will say so at the assessment.
Prior-period intercompany that was never recorded on both sides cannot be retrospectively matched with integrity. We can establish a clean starting point and will label what is derived.
Questions
Send your entity list and current process, and we will show you what computed consolidation looks like.