Tick & Tie Finance Blog

Automating Intercompany Transaction Reconciliations with AI-Machine Learning, RPA and Process

December 20, 2019

Intercompany transaction reconciliations are challenging on a number of levels - from intermingling to compressed timing for roll-up. This post dives into specific strategies and tactics to help minimize the pain and shorten the time to completion.

Intercompany (I/C) accounting can be a true cesspool of transactions and if done improperly, a driver of financial restatements. Companies in high growth mode that are actively acquiring and/or merging with other organizations or holding companies with multiple independently run companies create an environment with a multitude of teams, processes, systems and reporting cycles that must all connect and align in order to close the books accurately in the shortest period of time possible.

A 2016 Deloitte poll of more than 3,800 accounting and finance professionals suggests that disparate software systems in the different legal entities pose the biggest problem for intercompany accounting (21.4% of respondents), followed by intercompany settlement (16.8%), complex intercompany agreements (16.7%), transfer-pricing compliance (13.3%), and foreign exchange exposure (9.4%).

What Makes Intercompany Reconciliations Challenging?

Managing a fast, efficient and accurate financial close for one company is tough enough, but when adding more legal entities, multiple reporting layers of CEO’s, companies under the same stock symbol located in multiple countries and a multitude of systems - the challenge is exponentially more difficult.

Some Specific Challenges:

  • Rapid growth of an organization through acquisition. Results in new people, new financial control needs, data flows and new systems brought into the organization that needs to be aligned.
  • Multiple systems at the legal entity level. Different data formats, processes and output that need to roll-up to the overall corporate reporting level
  • Mistakes at one level cascade into issues at the next level (s) up
  • Disputes regarding intracompany settlements and how they are handled
  • Sales recorded in different financial periods
  • Different methods of accrual
  • Lost or incorrectly entered invoices
  • Invoiced amount discrepancies
  • Exchange rate differences
  • Different company level rules for revenue recognition
  • Cash in transit issues
  • Multinational Tax Laws and Legal Structures including BEPS impacts

Common Mistakes with Intercompany Reconciliations

Co-mingling I/C transactions with third-party transactions is often the first mistake companies make in this realm. Keeping a clean delineation between true I/C transactions and third-party transactions requires some best practices such as:

  • Establish separate and discreet accounts used solely for intercompany transactions. Each intercompany relationship will have it’s own set of accounts - AR/AP/Revenue/COGS.
  • Standardizing policies that govern critical areas across the organization globally. These policies need to be detailed enough to help F&A teams code transactions properly in their ERP systems across legal entities and international borders.
  • Establish a center of excellence made up of tax, finance, IT, and treasury experts from within the company who on a global level understand the accounting and technology involved in intercompany accounting.
  • Set up a master data management program to execute standardized global policies. A master data management program ensures that new and acquired accounts are set up in alignment with the policies and that intercompany transactions are processed in the same, standardized way.

The second mistake is using netting to settle I/C transactions versus full settlement. The result being an inordinate amount of close time and effort being consumed attempting to reconcile I/C. To achieve effective netting and settlement, which is critical for the treasury function, companies need multilateral settlements based on a cash management strategy that defines when settlements require cash transactions versus accounting entries.

How to Automate Intercompany Transaction Reconciliations

Automating your intercompany reconciliations requires three major activities:


A better process to eliminate the co-mingling of transactions and completely separate I/C transactions and accounts. It is always better to stop the problems from happening in the first place vs building protocols to fix later in the reconciliation flow.

In addition to establishing separate and discreet intercompany accounts, you want to ensure third party transactions don’t land in these separate I/C accounts. This requires starting at the end of the reconciliation process and identifying the transactions that are not intercompany and following those transactions back upstream to see how they became part of the intercompany transaction stream.

  • Was the transaction coded correctly?
  • If so, how did it become a problem in the reconciliation flow?
  • If not, who/what system was responsible for coding that set of transactions?
  • Is there a policy for how those transactions should be processed?
  • Are there a large number of manual procedures in the reconciliation flow where errors can be introduced? How can we minimize the errors and associated risk?


Utilize technology to evaluate and match intercompany transactions while highlighting exceptions that need further research. Depending on the number and complexity of your intercompany reconciliation process, you have a range of technologies to help in the transaction matching process:

  • Excel Templates - For small volume, low complexity reconciliations companies like Smartsheet provide free excel templates for various reconciliations including intercompany. As the data complexity grows and you need more intelligence - you will need to step up to using advanced Excel tools like Macros, Visual Basic and V-Lookups.
  • Robotic Process Automation (RPA) - RPA is the hot technology in many industries right now including the finance & accounting function. RPA automates specific manual tasks like downloading and basic data clean-up of bank statements, invoices and other documents. RPA can also follow along the rules you’ve created for basic account reconciliation - especially if you’ve built this reconciliation in a series of Excel spreadsheets. The big challenges with RPA are that it breaks quickly when data sources change or gets complex and it locks you into a process that may be inefficient to start with.
  • Machine Learning (ML) a form of Artificial Intelligence (AI) is a new technical approach to an old reconciliation problem. As the amount and complexity of data grows, rules based systems like built in Excel or with RPA loses accuracy and increases in cost to build and maintain. New account reconciliation platforms like Sigma IQ are built on ML and can be quickly deployed to new intercompany transaction reconciliation use cases and scale to handle lots of data as well as complex relationships.


Implement internal controls that enable group companies to reduce the risk of material misstatements in the financial information resulting from intercompany reconciliation issues.

Examples of such internal controls include reviewing posted intercompany transactions by senior specialists, approving of the prepared reconciliation spreadsheets by the financial officers of the individual entities, running automated matching processes in the system regarding debt claims and liabilities, accounts receivable and payable, etc.

Process, Technology and Controls Can Save the Day

Any business process that becomes unwieldy and a time suck can be improved through a combination of better process, the right technology and clear/consistent controls. The trick is to apply the time to do the hard work to clean up the upstream process of transaction coding and then use the right level of technology for your situation wrapped with proper controls to be consistently correct and then iterate for improvement.

If you are interested to learn how Sigma IQ is solving the account reconciliation process with Machine Learning, click here to watch a short video on how we do what we do.

Thanks for reading!

Related Posts & Resources

Reducing Time to Close through AI-driven Reconciliation

According to the APQC General Accounting Open Standards Benchmarking survey (2,300 companies participated) - Cycle Time for Monthly close ranges from 4.8 days or less for the top 25% of companies to 10 days or more for the bottom 25% of performers.

Learn How To Be a Top Performer
Cost Savings through AI-driven Account Reconciliation

According to a study by Robert Half & the Financial Executives Research Foundation (FERF), only 13% of F&A teams have utilized advancements in technology solutions, with the majority of CFO’s admitting they still struggle with painful aspects of account reconciliation.

Read About AI-Driven Cost Savings
Automating F&A and Operations through AI

It's time for finance and accounting operations to move beyond spreadsheets and fragile rules based systems. Next generation technology such as Machine Learning provide specific solutions to hard problems.

Read About F&A Automation with AI

Get the CFO Brief

Our newsletter is built for F&A professionals to deliver insight into new technologies, advice to advance your career and tools/tips to make your job easier so you get your evenings and weekends back.

Schedule a Demo

Click here to watch a short product video or set up time to see for yourself how simple account reconciliation can be.

Blog Categories

Blog Tags

Build Your Knowledge

Sign up for the Sigma IQ - CFO Brief and let us help you advance your knowledge and career by delivering articles, insights, and other resources geared towards Finance & Accounting teams.

© 2019 Sigma IQ. All Rights Reserved.
Privacy PolicyTerms of Service