Billing Errors — the Unhappy Marriage of Too Many Systems
If you are unable to accurately bill your customers, then there is not much point to being in business.
For many companies, the time and effort required to accurately bill is more costly than the good or service being provided to the customer. This is especially true for companies with heterogeneous data and process platforms. Consider, for example, a company with separate billing, accounts receivable, order fulfillment and inventory management systems. While this might seems extreme, it is actually fairly typical, especially in more established organizations. The silo technology still may fill a need, and either discarding or re-engineering the business applications may be too expensive an option.
Thus, these systems must periodically be reconciled. Indeed, proper audit control demands a regular accounting of data and associated input processes. If users must enter similar data into numerous systems, then control processes must be in place to ensure accuracy and conformance across the varying platforms. Ideally, of course, these control processes could be used as a basis for system interconnects, thereby eliminating duplication of effort in data input and management. In lieu of such automation, manual review is required.
Creating such a reconciliation process is relatively straightforward, albeit initially time consuming. First, identify all of the pertinent systems. Not every data source in an organization need be part of the reconciliation. This should be an iterative process. Initially, use only those that can be directly associated with the variance. Second, following identification, document all of the data that is input into each system. Key calculations made by each system should also be included. Third, evaluate all of the processes which impact the data. This might include a review of the physical contract, the order fulfillment processes, including shipping and returns as well as inventory accounting procedures. Fourth, identify break points in the process. Incorrect bills, customer complaints and reports from the various systems can all provide test cases for identifying problems with the existing procedures and controls. Finally, the process break points must be addressed and the resolutions tested. If the root problems are not addressed, then the system variances will continue. The goal of this process is to eliminate or minimize errors, and thereby provide an accurate bill to your customer.
Once all of the supporting processes have been identified, evaluated, corrected and tested, the data should be cleaned. There are a number of ways to approach this task, and, depending on the amount of data and the complexity of the relationships, it may be best to hire a consultant that specializes in data migration. If the data sets are relatively small or simple, then either a spreadsheet or simple database structure can be used to identify variances.
Cleaning the data is always a challenge. One key problem is identifying which system actually has the correct data. If customer information, for example, is entered into a billing, inventory management and order fulfillment system, then the possibility exists that all three systems have incorrect data. This is why identifying all of the processes that support each data source is so important. If all three systems are in agreement, then one can be fairly certain that the given customer details are correct. If, however, as is often the case, at least two of the systems are not in agreement, then an external data source (e.g. a physical contract) must be used to validate the data.
The system that stores customer billing data (e.g. product of service purchase details, charges and payment history) should always be used as the primary system to which all other data sources should reconcile. This system is, after all, the system that generates bills and provides revenue data to your organization. Assume, for example, that the order fulfillment system shows three products being delivered to the customer, but the billing system shows charges and payments for only one. In the absence of any supporting external information, the billing system would be assumed to be correct. Basically, the customer should be getting what they are paying for. Since in this case the customer is paying for 1 product, then the order fulfillment system should be modified to provision only one of the products for the customer.
After the initial variance reconciliation is complete, it is important to revisit all of the supporting processes. The users may have discovered additional information or a new method to better manage the data. It is essential that the reconciliations be scheduled and regularly completed. If the initial reconciliation found an inordinate amount of errors that resulted in substantial revenue loss, then the process should be completed at least quarterly until such variances are minimized between the audits.