Platform · financial core

Consolidation that is not a spreadsheet

Most groups under $100M consolidate in Excel: export each entity, paste into tabs, eliminate intercompany from a schedule somebody maintains, translate currency at a rate somebody looked up, and produce a group P&L that is correct and entirely unauditable.

Permitted

  • Adjusting entries
  • Reclassifications

Blocked

  • New subledger activity
State is enforced by the posting engine, not by convention. An agent cannot post into a closed period at any authority level.
Continuous, not monthlyElimination from transactionsEvery figure drills to source

What it does

Six things, specifically.

Any ownership structure

Wholly owned, partial, joint ventures, and multi-tier holdings, with the consolidation method following the ownership rather than being decided per report.

Intercompany from transactions

Eliminated because the transactions are matched to each other, not because a schedule says they should be. Unmatched intercompany is raised as an exception rather than plugged.

Currency translation

Average rate for income, closing rate for balances, historical for equity, with the cumulative translation adjustment computed rather than balanced to.

Minority interest

Computed per entity per period on the actual ownership percentage, including where it changed mid-period, which is the case a workbook always gets wrong.

Continuous

The group position is current at all times rather than produced once a month, so a question about group performance in week two has an answer.

Drill through to source

Any consolidated figure opens into the entity balances behind it and then into the transactions. A group number you cannot trace is a group number you will be asked about.

The workbook is the risk

A consolidation spreadsheet is usually correct. What it is not is auditable, reproducible, or survivable — it has no version history, no test, no access control, and one maintainer who has never taken two consecutive weeks off during a close.

It is also slow at exactly the wrong moment. Consolidation happens at the end of the close, under time pressure, which is when manual paste-and-eliminate work is most likely to go wrong and least likely to be checked.

The consolidation workbook is usually right. It is the fact that nobody could prove it that makes it a problem.

Elimination that matches rather than schedules

The standard approach maintains a list of intercompany accounts and eliminates their balances. It works until the two sides disagree — one entity recorded $48,200 and the other $48,000, or one recorded in March and the other in April.

Matching transaction to transaction surfaces those as exceptions instead of absorbing them. In practice most groups discover a persistent intercompany difference they had been plugging for years, and the plug had a name and a home in the workbook.

Translation, done by the book

Income at average rate, balances at closing, equity at historical, with the cumulative translation adjustment falling out of the arithmetic rather than being the number that makes the balance sheet balance.

That last distinction is the one auditors test. A CTA computed as a plug is a plug, and it absorbs every translation error silently until somebody decomposes it.

Ownership that changed mid-year

An acquisition in May, a partial disposal in September, a step-up from 60% to 80% in November — each requires consolidation from a specific date at a specific percentage, and each is where hand-built consolidations reliably fail.

Holding ownership as dated periods rather than as a single percentage makes those ordinary. It also makes prior periods reproducible, which matters the moment anyone restates.

Limits

Where this does not help.

Not full statutory reporting

We produce consolidated statements and the supporting detail. Filing under a specific local GAAP or IFRS taxonomy needs your accountants.

Complex equity accounting needs a person

Associates, hyperinflationary economies, and unusual acquisition structures are handled with your auditors rather than automatically.

Entity data quality is the ceiling

A consolidation is only as good as the ledgers under it. Where one entity’s books are unreliable, the group figure inherits that and we will say so.

Questions

What people ask.

How many entities can it handle?
Multi-tier structures with dozens of entities. The constraint is the quality of the underlying ledgers, not the count.
Do all entities need to be on erp.io?
No. Entities can stay on their existing systems and be read into the graph. Mixed groups are the normal case rather than the exception.
How is intercompany eliminated?
By matching transactions to each other. Where the two sides disagree, it is raised as an exception rather than plugged.
Can we see the group position mid-month?
Yes. Consolidation is continuous rather than a month-end exercise, so week-two questions have answers.
What about mid-year ownership changes?
Ownership is held as dated periods, so an acquisition in May or a step-up in November consolidates correctly and prior periods stay reproducible.

Retire the consolidation workbook.

Send your entity structure and how you consolidate today. We will show you what it is plugging.