Reducing time, resources, and errors in payment reconciliation processes

With tighter cash flows, every day that passes without confirming all the outstanding receivables of a company adds to the risk of not being able to recover them. In addition, reconciliation processes are labor-intensive, time-consuming, and error-prone. Automation of such activities can immediately reduce costs, speed up resolution times and contribute positively to the bottom line.

KEY POINTS

Automation ensures a rapid transition from costly and imprecise manual reconciliation processes to automated reconciliation (requiring human intervention only for exceptions and errors in the source data).

b3lineicon|b3icon-gavel||Gavel

Main Pains

Extensive time and effort is required to process incoming payment details and correlate them with core insurance policy data
A team of up to 10 people could easily spend up to 3 days manually matching one day’s worth of payment data to policies (based on an average daily volume of more than 50 source files and 3900 payment transactions)
Re-work volume is also significant due to a 25-30% error rate
The process is excessively dependent on employees’ training, experience, accuracy, and attention to detail
There is a negative impact on other business processes waiting for input from reconciliation results

Solution

The system processes bank account statements looking for patterns related to policy numbers, paid amounts, currencies, payment dates, paying parties etc. The resulting data is checked against the core insurance information about payments due on policies, matched against the received amounts.

Data on allocated amounts resulting from the system matching is exported to a predefined export file that is later imported into the core insurance system.

Data entry errors and missing data from the source files are appended to tasks for manual handling.

Benefits

Once implemented, the solution provided significant time and cost savings because:

The automated processing now takes up to 3 hours, depending on the data load
98% of the processing is error free, only requires 0-2% data correction after the initial input
83% of the required resources have been freed – only 2 users needed to handle exceptions
All the daily information is processed in the same day
Most of the exceptions, usually due to simple data mismatches, can now be rapidly corrected manually based on automatically generated tasks