Audits are a necessary part of the accounting process, and, while there are clear benefits resulting from audits, the process also places resources constraints on controlling processes. This analysis focuses on how to manage an audit to require fewer resources while maintaining effectiveness of the audit. This article is meant to provide a framework for financial statement audit preparation in order to allow controllers to efficiently prepare for the arrival of internal and external audit. The controller should build the process for ensuring accurate financial reporting (normally in the order listed):
The controller will find that every audit question fits into one of the above categories. By addressing each point above when preparing GL balances through internal documentation and procedures, the controller prepares to efficiently manage a financial statement audit. Throughout the article, “documentation” means narratives in either word documents or excel. The standard of documentation should be such that a new user can use the documentation to prepare the procedures in a way which is materially correct (COSO 2013).
Policy
The policy development is always the first step. A good policy is very technical and includes discussion of any situation which might occur. One good tool is to discuss why certain complex situations are either not likely or not material and include data which demonstrates this. This will allow the controller to avoid implementing complex processes for immaterial transactions. A good policy also prescribes how the company will execute ASC 606 accounting and ensure financials are materially correct.
Inputs
Inputs are metrics or data which often result from actions from non-accounting departments, for example shipping information entered by the logistics team. Accordingly, it is important to have internal controls on the process of developing inputs or a process for confirming that there is little to no risk that input data will be incorrect or inaccurate. Operational controls should be in place to ensure that the data entered by non-accountants is sufficient and accurate enough for accounting reporting. Testing sample data is a good approach for allowing a controller to ensure the data is accurate.
It is also best to use a data set which will not often fluctuate. This will have implications on future conclusions and assumptions used in accounting as well as call into question past accounting and reporting decisions and ensure data can be locked for past periods to maintain the integrity of the data. Locking data is also necessary for audit purposes as a typical audit procedure is to re-pull the data to confirm the same results.
Assumptions
Assumptions are applied to inputs prior to performing final calculations of GL balances. The fewer assumptions, the better. To paraphrase Occam’s Razor, the fewer the assumptions, the more likely it is that you have the correct result. Each time an assumption is added to the process, there needs to be robust documentation supporting the assumption and an analysis of the sensitivity of the calculation to the assumption. Support for the assumption should use historical data which has been tested internally whenever possible and documented. If the assumption has a significant impact on the calculation, then there needs to be a narrative supporting the basis of the assumption. If the assumption can change within a wide range and not impact the calculation much, normally it is sufficient to document that the calculation is not sensitive to the assumption and leave the documentation simple.
Calculations
Calculations often are applied to inputs and assumptions in order to develop the final GL balance or entry for booking. In some cases, the calculation is as simple as a sum of records, but could also be quite complex, for example, a regression analysis. If the data being used are kept in Excel, then the best approach is to lock calculation cells or pages so that they cannot be edited. The best solution, though, as discussed later, is to configure an ERP or interface to perform the calculations. This allows the audit process to focus on testing the configuration of the system (application controls) which is much more efficient. Clear documentation of the support for the calculation method should be maintained and easily accessible in order to facilitate audit procedures.
Reconciliation
Of course, it goes without saying that in the end of the process, there needs to be a reconciliation from the GL calculations to the general ledger. For excel calculations, the most efficient is to have this in a tab within the calculation. When an ERP or other automated tool performs the calculation, then the controller should build a process to check the subledger data from the ERP to the general ledger. Document within the reconciliation clear, quantitative thresholds for further investigation and amounts which are immaterial and thus do not require additional investigation or consideration. Example of adequate language is “The difference is neither qualitative nor quantitively significant is the causes are not pervasive and the amount is immaterial to the financials per the company’s policy on reconciliation difference.”
Automation
Automation can take many forms. The simplest form of automation is excel models. Building detailed, robust reporting is another form of automation in order to facilitate excel models or GL entry generation. More complex forms include building ERP functions or build-on applications that do automated calculations and GL entries without user intervention. Automating processes and calculations goes a long way to decreasing risk and reducing processing time, thus decreasing cost. It can also significantly decrease audit work performed as audit can then focus on only the few failure risks and rely on the rest of the process. The more manual/error-prone a process is, the more the auditors will question the process and results.
Reconciliations and review procedures can also be automated using tools like Blackline which automatically population GL balances to ensure the proper balance is being reconcilied and also keeps electronic signatures and time stamps of preparation and review for reliable audit evidence that controls are performed.
The Big Picture
Remember to keep it simple. Document the big, most material points. Use data which is reliable and can be locked. Automate processes and draw the auditor’s attention to the reliability of inputs, assumption and most importantly, process. For any other potential questions on revenue, find a data set and analysis which proves the amount is immaterial, and recalculate the potential amounts quarterly to confirm no further consideration is required.
Leave a Reply