Intercompany Reconciliation Automation for Multi-Entity Groups

Intercompany reconciliation automation is a set of AI workers that pull each entity's ledger extract on a fixed schedule, pair the mirror entries between entities, and list every unmatched or mis-dated item with its entity, amount and age. The workers send confirmation requests to the counterpart finance contact and keep a log of replies. A finance lead reviews the exceptions, decides on currency and timing differences, and approves the eliminations. The aim is that the matching work runs without a person keying or comparing figures by hand, while the judgement calls stay with the finance team.
In short
- Intercompany matching is a fixed monthly routine, and it is a good candidate for scheduled automation.
- Most breaks come from timing, currency or coding differences, and each type needs a different decision.
- A finance lead keeps control of exceptions, currency and timing calls, and elimination approvals.
- Groups with few intercompany items may be better served by a shared schedule; the audit will show which applies.
- Confirmation requests and replies are logged, so each break has a dated trail for the close file.
What the service does
Every group with more than one legal entity carries balances between those entities. A subsidiary in Germany owes the UK parent for management fees. The parent holds a receivable, and the German entity books the matching payable. In a clean close those two figures are the same. In practice they often are not, and the difference has to be found, explained and cleared before the group accounts can be consolidated.
The service covers that matching work. It reads the ledger extracts from each entity, pairs the receivable in one entity with the payable in another, and sorts the results into three groups: matched, unmatched and mis-dated. Each unmatched item is listed with the entities involved, the amount in the reporting currency, and how many days it has been open. The finance lead then works from that list rather than from a spreadsheet of entity tabs.
Why the mismatches happen
Most breaks have a small number of causes. One entity records an invoice in the month it is issued, while its counterpart records it when it is paid. One side uses a spot rate on the transaction date and the other uses a month-end rate. The two entities post to different intercompany accounts, or one uses the wrong entity code. An accountancy firm's guide on intercompany accounting lists timing differences, currency and exchange-rate differences, and inconsistent coding as the common causes of mismatch.
The matching logic is built around these causes. A timing break shows up as the same amount appearing in two adjacent periods. A currency break shows up as a small gap that moves with the exchange rate. A coding break shows up as an amount that matches exactly but sits under the wrong account or entity code. Sorting breaks by type makes it easier for the finance lead to decide what to do with each group.
Why this matters before elimination
The IFRS Foundation's standard IFRS 10, Consolidated Financial Statements, requires that intragroup balances, transactions, income and expenses be eliminated in full when a group reports under IFRS. Under US GAAP the equivalent rule sits in ASC 810. If the two sides of an intercompany balance do not agree, the elimination entry either leaves a residual in the consolidated accounts or forces a plug. Datasights' 2026 guide on consolidation notes that an unbalanced consolidated trial balance often traces back to incomplete elimination entries, currency translation errors or data collection problems.
This is why the matching has to happen early in the close, before the elimination journals are drafted. The work is the same each month, which makes it a good candidate for a scheduled routine. The judgement about what a break means does not become simpler just because the list arrives sooner.
Who it is for
The service suits any business with several legal entities that trade with each other. That includes a five-person firm with a holding company and one operating subsidiary, and it includes a group with thousands of staff across many countries. The question is not size but volume and structure. If intercompany invoices run into the hundreds each month, or if entities report in different currencies, manual matching becomes slow and error-prone.
It is less useful where the group has one entity, or where intercompany trade is rare enough that a single reviewer can check a handful of items in an hour. In those cases a well-kept shared schedule may be the better choice, and the audit described below will say so.
What is included
- A mapping of each entity's ledger accounts, currencies and intercompany account codes
- A scheduled extract from each entity's ledger, with the cut-off date and run time agreed in advance
- Automatic pairing of mirror entries, such as a receivable in one entity against a payable in another
- An exception list with entity, counterparty, amount, currency, period and age for every unmatched or mis-dated item
- Draft confirmation requests sent to the named counterpart finance contact, with a dated log of each request and reply
- A review queue for the finance lead, with each exception marked as a timing, currency, coding or unresolved item
- A written record of each reviewed item and the decision taken, kept for the close file
- Draft elimination schedules prepared from the agreed matches, for approval by the finance lead
How the service runs
1. Audit
The first step is a look at what exists today. The team reviews the chart of accounts for intercompany codes, the ERP or ledger tools each entity uses, the currencies involved, and how many items cross entity boundaries each month. The audit also looks at past close files to see which breaks recur. The output is a short findings note and a recommendation on whether automation fits this group.
2. Process map
Next, the team maps the close as it runs now, and as it would run with the service in place. The map shows who sends what to whom, which data is pulled, where confirmations are chased, and which decisions stay with the finance lead. Owners sign off the map before any build starts, because the approval points are where the control sits.
3. Build
The build connects the agreed ledger extracts, sets the matching rules and configures the exception categories. Confirmation templates are written in plain language and tested with a few counterpart contacts. Access is limited to the extracts and the review queue. Nothing posts to the ledger during the build.
4. Run
In the run phase the workers pull the extracts on the agreed schedule, pair the entries, produce the exception list and send the confirmation requests. The finance lead works the review queue, records a decision on each item and approves the elimination schedule. A parallel run against the manual close for at least one full cycle is the usual way to check the output before the manual process is switched off.
5. Support
After go-live, support covers changes to the chart of accounts, new entities, changes to ERP exports, and rule adjustments when a new type of break appears. Support also includes a periodic review of the exception log, so that recurring causes can be addressed at source.
Typical timeline
The durations below are planning estimates for a group with a few entities and one reporting currency per entity. The audit confirms the real plan.
| Phase | Typical length | Main output |
|---|---|---|
| Audit | 1 to 2 weeks | Findings note and go or no-go view |
| Process map | 1 week | Signed-off map of the close with control points |
| Build | 3 to 5 weeks | Extracts, matching rules and review queue |
| Parallel run | 1 full close cycle | Comparison of automated and manual results |
| Run and support | Ongoing | Monthly exception log and rule updates |
What the client provides
The client provides read access to each entity's ledger extract, the intercompany chart of accounts, the exchange rates used at each close, and a named finance contact for each counterpart entity. A finance lead must be available to review exceptions in the close window. Someone who knows why each entity posts the way it does also needs to be available during the audit. The quality of the output depends on the quality of these inputs. If two entities post the same transaction to different codes with no shared reference, the workers can only flag the break, not explain it.
What this service does not cover
Automation does not decide what a break means. A timing difference, an exchange-rate difference and a genuine error can look identical in the exception list. The finance lead makes that call, and the service records the decision. The service also does not approve eliminations, set transfer prices, judge whether an intercompany charge is arm's length, or resolve a dispute between two entities' management. Those remain human decisions. The service does not replace the ERP or the consolidation tool, and it does not post entries to the ledger. Counterparties who do not reply still need a person to follow up. Results depend on the group's own figures, which the free audit measures.
This explains the rule in general terms; it is not legal or tax advice for your case.
How AiStaffo would automate this
AiStaffo would connect to each entity's ledger extract, map the intercompany accounts, and set up the pairing rules for receivables and payables across entities. On the agreed schedule, the AI workers pull the extracts, pair mirror entries, sort breaks into timing, currency, coding and unresolved groups, and send confirmation requests to each named counterpart contact. A person still reviews the exception queue, decides on currency and timing differences, and approves the elimination schedule before anything is posted. Book a free automation audit to see how this would fit your entities and close calendar.
Questions people ask
What is intercompany reconciliation automation?
Why do intercompany balances fail to match between entities?
Does automation remove the need for an accountant to review intercompany items?
Which accounting standards require intercompany balances to be eliminated?
How long does an intercompany reconciliation setup take?
What does the client need to provide?
Book a free automation audit
Thirty minutes. We look at one process you run every week and tell you exactly what an AI worker would take off your desk, and what it would not.
































