Intercompany Reconciliation: How AI Agents Automate It
Intercompany reconciliation is where multi-entity closes bog down: mismatches, timing differences, FX, and manual matching across entities and ERPs. Here is how AI agents automate it and prepare clean eliminations.
For any company with more than one legal entity, the close has a predictable bottleneck: intercompany. Every transaction between entities, a sale, a loan, a shared cost, an allocation, shows up on two sets of books, and the two sides have to agree before you can consolidate. When they do not, someone spends the back half of the close chasing why: a timing difference, an FX rate, a missing entry, a transfer-pricing adjustment.
Intercompany reconciliation is that matching work, and it is exactly the kind of high-volume, cross-system, rule-bound task an AI agent is built for. Here is what it involves and how agents automate it.
What intercompany reconciliation is
Intercompany reconciliation is the process of matching transactions between related entities so both sides agree before consolidation, resolving timing differences, FX, and transfer-pricing adjustments, and preparing the elimination entries that remove intercompany activity from the consolidated results. In multi-entity groups it is one of the most time-consuming parts of the close.
Why intercompany is so painful
It is also widespread: a Deloitte intercompany accounting survey found that 54% of companies still rely on manual intercompany processing and 30% report significant out-of-balance intercompany positions. The reason is structural: the two sides of an intercompany transaction often live in different systems, different ledgers, sometimes different ERPs, in different currencies, booked on different days. So a "mismatch" is usually not an error, it is a timing or FX difference that has to be identified and explained. Doing that means pulling both sides, matching them, isolating the real breaks from the noise, and chasing the counterpart entity to resolve them, across dozens or hundreds of transactions, on a deadline. It is slow, and it is where multi-entity closes routinely stall.
What AI agents do with intercompany
- Pull both sides. Gather intercompany transactions from every entity and ERP into one view.
- Match automatically. Pair the two sides, clearing the ones that agree and isolating the breaks.
- Explain the differences. Separate real mismatches from timing and FX differences, and attribute the cause.
- Flag and route. Surface unresolved breaks to the right entity owner with the detail to fix them.
- Prepare eliminations. Draft the elimination entries so consolidation starts from clean intercompany balances.
How Concourse automates intercompany reconciliation
Concourse agents connect across your entities and ERPs, match intercompany transactions, separate real breaks from timing and FX differences, and prepare elimination entries, with every match and flag traceable back to source and a person approving. It connects to 100+ systems including NetSuite and QuickBooks and is SOC 2 Type II certified. This is one of the workflows inside financial close automation.
Frequently asked questions
What is intercompany reconciliation?
It is matching transactions between related entities so both sides agree before consolidation, resolving timing, FX, and transfer-pricing differences, and preparing elimination entries. It removes intercompany activity from the consolidated results so the group numbers are clean.
Why is intercompany reconciliation so hard?
The two sides of a transaction often sit in different ledgers, ERPs, and currencies, booked on different dates, so most mismatches are timing or FX differences rather than errors. Identifying the real breaks and chasing the counterpart entity to resolve them, across many transactions on a deadline, is what slows the close.
How do AI agents help?
Agents pull both sides across entities and systems, match them automatically, distinguish real breaks from timing and FX noise, route the unresolved ones to the right owner, and draft the elimination entries, turning a manual chase into a continuous, traceable reconciliation with a human approving.
The bottom line
Intercompany reconciliation stalls closes because the two sides live in different systems and most mismatches are timing or FX noise that has to be sorted from real breaks. Agents do exactly that sorting at scale, matching both sides, explaining the differences, and preparing clean eliminations, so consolidation is not the part everyone dreads.
If you want agents matching intercompany across your entities, talk to our team.
