Why Do We Have So Many Backdated Adjustments? | Payroll Best Practices - A System Implementer’s Viewpoint

Is it payroll that’s complex or just your payroll software?

Payroll Implementation Best Practice #3: Why do we have so many backdated adjustments?

Payroll process is fairly simple but what makes it complicated and time-consuming is the backdated adjustments, even if you have automated the calculations for these adjustments enough. Backdated adjustments may be triggered by a multitude of factors such as lack of relevant technology, lack of discipline, policy gaps etc. This creates more work for back office, to check and reconcile such adjustments. Making the process even more tedious are the backdated pay rate changes.

Some of the ways to tackle this problem include:

• Make pay rate changes on start of pay cycle – This takes away the need of proration so that payroll analysts don’t need to eyeball such variances. This practice comes in even more handy when you have multiple pay frequencies, for e.g. weekly, semi-monthly, monthly etc. and staff movements trigger a pay frequency change in mid of pay cycle
• Introduce automation to ensure timely and error free data flow from attendance systems to payroll as per the processing schedule
• Leverage technologies such as facial recognition, mobile apps & chatbots to get attendance/ absence inputs and approval on time. Some companies in services and education sector are also experimenting with blockchain technology to simplify their attendance process
• Limit past pay cycle related data changes to 2-3 pay cycles to minimize retro processing
• Change management to create awareness amongst employees and managers about entering/ approving transactions on time so adjustments are contained well within 5 -10%

Despite all this, there could still be issues and we must have pre-processing checks to ensure that the payroll is ready to run with minimal exceptions. While you may have a robust integration between Time & Payroll, it is recommended that the payroll system allows you to configure reasonability checks around the data inputs to ensure that any fallouts are flagged in advance. For instance, if you expect ~10,000 time records every cycle and in a certain cycle this number drops to 8,000 it may be worth investigating and correcting this before the payroll numbers are churned out.

Similarly, for any ad-hoc inputs that are to be added to payroll, it’s important to have system checks in-built to ensure that a user does not make a mistake. Like, if a person enters 2000 as an input when generally it never exceeds 500 for any employee in the past, the system should be able to detect this outlier.  In addition to allowing such controls on ad-hoc inputs, Ramco Systems has recently introduced a Machine Learning based tool for detecting anomalies in payroll that a user might miss. This allows the system to flag out discrepancies based on data across several pay cycles as compared to the usual payroll reconciliation checks that only compare last cycle and current cycle.

Another time consuming task for payroll officers is collating inputs from managers across sites and branch offices. This process is usually based on manual spreadsheets and data may not be complete at the time of payroll cut off thereby resulting in backdated payments or deductions. It is best to use manager self-service options for such inputs using an approval workflow so that the data is available for payroll prior to the cutoff.

In a nutshell, a systematic approach using both process and technology enhancements is important to reduce the volume of backdated adjustments. For companies working towards payroll transformation, reduction of backdated adjustments should be one of the foremost priorities that should be identified right in the beginning. And it should be continuously monitored throughout the transformation journey.

Want to know what the other best practices are?

Read the 1^st part here: https://bit.ly/2XneWuF

Read the 2^nd part here: https://bit.ly/2Wercfn