(Broaden Your Perspective)
A credit card processing operation made an expensive investment in upgrading its remittance system hardware and software to process customer payments. The new system was state-of-the-art and required a lengthy installation and training period before going live
Immediately, management noticed an increase in departmental performance and productivity. Six months after going live overall data entry operator’s productivity and total system throughput showed significant increases in productivity rates. The credit card company’s management shared the apparent success of their investment in trade publications and conferences.
After nine months of operation, management began to notice declining productivity and throughput. A tiger team reviewed the operation, and a few adjustments to the workflow process were made.
The decline in departmental performance continued to grow over the next several months. Disappointment rose as the gains from the processing and workflow adjustments began to disappear.
The vendor was contacted and informed that their system was no longer performing to specifications. The operation management team of the credit card company requested the vendor send a team of technicians to test the hardware and software for flaws or errors.
After an intense week of testing, the vendor’s team reported that they were unable to discover any problems with the system and declared it was operating to contractual specifications.
In the weeks following the vendor’s diagnostic testing, the customer continued to experience declining throughput. The customer was convinced that it wasn’t a management, workflow, staff issue, and it had to be a problem with the system or hardware.
A Problem With the System
Again, the vendor was contacted, and a team of engineers arrived on site for an additional week of intensive testing and analysis. At the end of a long week, the engineering team reported they still could not find any problems with the equipment or the software.
The customer became frustrated and was convinced that the problem was with the system. Tempers flared, tension rose, and threats of lawsuits were made if the problem wasn’t corrected.
The client began having buyer’s remorse, and the management team became anxious about their jobs. The vendor’s anxiety increased for fear of lawsuits, bad press, and loss of pending industry sales.
In a final act of desperation before legal action, both parties agreed to a review by an independent third party.
A Review by a Third Party
A mutually known and respected independent industry consultant was engaged to perform an operational autopsy. The first step in the analysis by the independent consultant was to gather historical data on processing volumes, a mix of work types, operator production activities, and rates.
The following background information is provided to educate the reader on the processing of credit card payments.
The amount of manual intervention required to process a credit card remittance is dependent on the type of payment transaction made by the cardholder. A cardholder can make one of three types of payment transactions:
- Payment of the total amount due
- Payment of the minimum amount due
- Payment of an amount different from the minimum or total amount due.
The total amount and minimum amount payments require a single keying of the paid amount. If the check amount keyed matches either the total or minimum amount due on the remittance coupon, the transaction is considered valid and accepted for payment. No further manual intervention is required.
When the payment amount does not match the total or minimum amount due, a second keying of the amount by a different operator is required to validate the payment amount. In the case of a difference in the amounts keyed by the first and second operators, a third operator will key the payment amount.
If the amount of the third keying matches either of the previous amounts keyed, the payment is accepted. A failure of the third amount keyed to match either of the two previous keyed amounts results in the payment transaction being forwarded for supervisor resolution intervention.
The consultant’s analysis discovered that in the time after the acceptance of the system, the mix of payment types had changed. Due to changes in the economy and higher interest rates, there was a change in cardholder payment behavior. Cardholders increasingly began to make payments of a higher dollar amount to reduce their balance and avoid additional interest fees on outstanding balances. Combined with the larger payment amounts was a change in the percentage mix of payment types, with an increase in the percentage of alternative transactions.
Before the system installation, the average amount of a payment was less than a hundred dollars. The changes in the economy that occurred after the installation triggered an increase in the average amount paid. The average payment amount increased to greater than one hundred dollars.
Upon completing the data analysis, the independent consultant reported to the relief of the vendor and anger of the client there were no functional problems with the equipment or the software.
The system was performing to contractual specifications. The system performance was exceeding contracted specifications.
The Root Cause
The root cause of the drop in throughput was the change in the mix of the type of payment transactions being processed. There were two primary reasons for the decline in overall system throughput. The first was an increase in the percentage of alternate payment transactions. The second reason was an increase in the average payment amount. Combined these two factors were accountable for the drop in throughput.
Increased Data Entry Work
In other words, more work and time were required to process the same volume but a different mix of work. The increase in the average dollar payment from under $100 to over $100 increased the number of keystrokes by 20%.
A payment of under $100 required a total of five keystrokes, four for the amount and one for the enter key.
A payment of greater than $100 required six keystrokes, five for the amount and one for the enter key. The one extra keystroke accounted for approximately a 20% increase in data entry. (1 extra keystroke due to the increase in average payment amount / 5 keystrokes required before the increase in the average payment amount = a 20% increase in keystrokes)
During the same period, the percentage of payments for alternative amounts (those not matching full or minimum payment due) had increased from 30% of the total volume to greater than 50% of the processed volume. Each alternative payment required a double keying of the amount for validation of the payment. The 20% increase in the percentage of alternative payments resulted in a 67% increase in keying requirements for alternative payments. (50% – 30% = 20%. 20% / 30% = 67% increase)
A combination of factors was responsible for the decline in department performance. The catalyst for the change was the economy. An increase in interest rates was the stimulus for a change in the percentage mix of payment types and growth in the average payment amount.
Combining these two factors was responsible for a thirty-plus percentage decline in overall department throughput due to the extra work required to process a payment transaction.
The decline was valid and quantifiable.
An interesting side note and discovery was that the average data entry speed of individual operators had increased as they became more familiar and confident in the processing system. Had their data entry declining productivity not improved, the system throughput decline would have been more significant.
This case study illustrates the value of following the guidelines for performing a root cause analysis:
Get Detached – Remove the emotion from the analysis.
Before seeking independent assistance, both the client and the vendor had become emotionally involved.
They adopted defensive positions and were unable to look beyond their positions for other causes to the problem.
Get the Facts – Focus on the facts and the variances
The problem did not exist before the installation of the new system. It was easy to place blame on the system. Neither party investigated what other changes had occurred during the same timeframe.
Face the Facts – Understand the cause of the variances
A review of historical payment data provided the information for the identification of the variances in the payment mix and work effort.
Manager Your Options – Move forward with the new understanding.
Once the identification of the root cause (changes in the economy) was identified effort was made by both the vendor and the client to put greater effort into the research and development of technology that would reduce the amount of manual effort. Improvements in Courtesy Amount Recognition and Handprint Recognition were accelerated and implemented in the following year. These two technologies quickly delivered the desired productivity gains and negated the production losses due to economic changes.