Multi-Entity Consolidation Without a Corporate System

There is a particular spreadsheet that exists in most mid-sized groups. It has one tab per legal entity, a tab for intercompany eliminations, a tab for foreign exchange translation, and a consolidated output tab feeding the board pack. It was built by someone who has since been promoted or has left. It is opened once a month by a financial controller who understands roughly eighty percent of it and does not touch the rest. That file is the group's consolidation system. It is also, in most cases, the single largest undocumented control weakness in the finance function.

How Groups End Up Here

Nobody decides to consolidate in a spreadsheet. It happens by accumulation. A company has one entity and a general ledger. It opens a second entity in another country, and combining two sets of numbers in Excel is faster than configuring anything. It acquires a business with a different accounting system, and rather than migrate it — which would delay the deal benefits — the new entity reports through a template. Two more acquisitions follow. A holding structure is added for tax reasons. A joint venture requires equity accounting. Six years later there are eleven reporting units, four accounting systems, three currencies and a workbook with fourteen thousand formulas that nobody has audited. Each individual decision was rational. The aggregate is a group reporting process with no version control, no audit trail and no segregation of duties.

Why This Is Different From Ordinary Spreadsheet Risk

Spreadsheet error is a well-documented phenomenon — research into operational spreadsheets has consistently found errors in the large majority of production models, and the public examples of costly mistakes are numerous enough to fill a conference agenda every year. Consolidation, though, concentrates that risk at the worst possible point. The output is the statutory and board-level view of group performance. An error here does not produce a bad operational decision; it produces a misstatement, and depending on the group's obligations, a restatement, an audit qualification or a regulatory problem. The specific failure modes are consistent. Intercompany eliminations do not net. Entity A records a receivable of 500 and Entity B records a payable of 480, because of timing, currency or an unrecorded credit note. The difference is plugged, the plug is never investigated, and it grows. Currency translation is applied inconsistently. Balance sheet items at closing rate, income statement at average, equity at historic. A single cell using the wrong rate is invisible and can persist for years. New entities break the structure. A tab added for an acquisition is not included in a sum range. Every subsequent month understates the group by exactly that entity. Percentages are hard-coded. An ownership stake changes and the workbook still holds the old figure, because the change was communicated to legal and tax but not to whoever maintains the file. There is no trail from output to source. Asked to explain a consolidated figure, the controller can show the cell but not the chain of adjustments behind it. That is the definition of an audit finding.

What Auditors Actually Look For

External audit scrutiny of consolidation processes tightened significantly through this period, driven by regulatory attention to internal control over financial reporting. The questions are predictable and most spreadsheet-based groups answer them badly. Who can change the model, and how is that restricted? Is there a documented review of the consolidation before it is issued, by someone independent of the preparer? How are manual journals authorised and evidenced? Can the group reconcile from statutory entity accounts to the consolidated statements in a documented way? What happens if the file is corrupted or the preparer is unavailable? The honest answer in most cases is that control depends entirely on one person being competent and present.

Fixing It Without a Corporate System

A full consolidation platform is not always justified. Plenty of groups with a handful of entities genuinely do not need one. But the controls can be built regardless of the tool.

• Standardise the chart of accounts across entities. This is the foundation. Mapping tables between eleven different account structures is where most consolidation errors originate, and harmonising the chart removes an entire class of them.
• Fix the intercompany process, not just the elimination. Agree balances between entities before close, with a named owner on each side and a tolerance threshold. Eliminating mismatched balances is treating the symptom.
• Put the model under version control. Dated, locked copies of each period's file, stored where they cannot be overwritten. This alone answers several audit questions.
• Separate preparation from review. A documented, evidenced review by someone who did not prepare the file. Without this there is no control, only a competent individual.
• Lock and protect formula ranges. Input cells open, calculation cells locked. Most consolidation errors are overwritten formulas.
• Log every manual journal. Reference, reason, approver, supporting document. Manual adjustments at group level are the highest-risk entries in the entire process and are usually the least documented.
• Build an independent check. A simple proof — movement in group equity reconciled to profit, dividends and translation differences — catches a surprising proportion of errors before the numbers are issued.
• Document the process end to end. Which entity reports what, by when, in what format, with which adjustments applied and why. If the controller were unavailable next month, could someone else close the group?

When to Move to a System

The threshold is not entity count alone. It is the combination of entity count, currency complexity, acquisition rate and reporting obligation. Groups that are acquiring regularly should invest earlier, because each acquisition adds structural complexity to the model at the point when finance capacity is most stretched. Groups with external reporting obligations — listed, regulated, or with significant lenders — need the audit trail. Groups where the close takes more than ten working days are usually spending more in finance time annually than a system would cost. The cost that never appears in the comparison is the delay. A group that takes three weeks to know its consolidated position is making decisions on stale information, and in an acquisitive business that is an expensive way to save on software.

The Modern Twist

AI tools are now being pointed at exactly this problem — reading entity submissions, proposing eliminations, flagging anomalies in consolidated movements. Used well, they are a genuine improvement on manual reconciliation, particularly for spotting the intercompany mismatches that humans plug. Used on an undocumented spreadsheet consolidation, they add a layer of unexplainable adjustment on top of an unexplainable process. Every assisted adjustment needs the same three things a manual one needs: a reason, an approver and a trail back to source. The underlying requirement has not changed since group accounting began. Someone must be able to explain how the consolidated number was produced, and that explanation must survive the departure of whoever produced it.

Common Questions

Why is spreadsheet-based consolidation risky?

Because it concentrates high-error-rate manual processing at the point of statutory and board reporting, with no version control, no audit trail, no segregation between preparation and review, and dependency on one individual's knowledge.

What are the most common consolidation errors?

Unreconciled intercompany balances that get plugged, inconsistent currency translation rates, new entities omitted from sum ranges, hard-coded ownership percentages that are never updated, and overwritten formulas.

What controls can be added without buying a system?

A standardised chart of accounts, pre-close intercompany agreement, version-controlled locked files, an evidenced independent review, protected formula ranges, a manual journal log, and an independent equity movement proof.

When should a group buy a consolidation system?

When acquisition activity is regular, external reporting obligations require an audit trail, or the close takes more than about ten working days — at which point the finance time consumed usually exceeds the cost of the software.

---

Consolidation Process Review — Outpace audits the workbook your group reporting actually depends on, adds the controls auditors ask about, and tells you honestly whether you need a system or just a better process.