PURCHASE TO PAY

We help organizations with digital transformation and process optimization from purchase to pay.

TECHNOLOGY

We use various cloud solutions to suit more sizable organizations.

INTEGRATIONS

We work with several P2P solutions that interface with leading ERP systems.

2 min read

Save money by allowing more errors

When implementing a purchase-to-pay system, the conditions under which an invoice may be matched against an order are defined. "These conditions are usually set with 'common sense' and are often based on years of experience in the accounts payable process," says Vincent Wouters. Sometimes there is even a clear policy in place that the difference should never exceed a certain amount.


All invoices whose discrepancy with the order/receipt differs by more than amount X should not be processed automatically. The invoice matching system is then set up so that every time an invoice does not meet the criteria, it is presented for manual processing

Vincent: "This means that an invoice is taken separately and picked up by an accounts payable employee. In most cases, this employee cannot assess the price difference. He will forward the invoice to the person who placed the order. The ordering person will look at the invoice and conclude that the difference is explainable."

He continues: "The prices in the ordering system have not been updated, for example, or for other reasons the price difference turns out to be correct after all. After this, the invoice is approved by one or more people and forwarded for payment. If the price is not explainable, the ordering party will contact the supplier and request a credit invoice."

What have we learned from this bill matching scenario?

We have a habit of keeping principles and rules if they worked well in the past. But what if we could use the data from these scenarios to verify that bill matching conditions are defined correctly?

 

In the image, a data analysis of a Dutch organization where the invoice should not deviate more than 10 euros and no more than 1 percent of the total invoice amount. What we see is the deviation of the invoice (the further from the center, the larger the deviation), the number of invoices with this deviation (height of the bar) and the total amount of the deviations (red line).

Examination of the data reveals that 97 percent of invoices with a discrepancy between 0 and 40 euros are approved without communication with the supplier and without a request for credit. Surprising perhaps, but the practice.

But if the tolerances are increased from 10 to 40, 100 percent instead of 97 percent will be processed automatically. We will then miss 3 percent being wrongly approved. This seems disadvantageous, of course, but this adjustment means that this organization will have to review 7,000 fewer invoices!

If the accounts payable clerk and buyer spend an average of at least 15 minutes per failed invoice, this is 1,750 hours of work. So this can be a significant cost savings versus 3 percent wrongly paid suppliers. And the suppliers who structurally over-invoice are also very easy to identify. Limited risk right?

Conclusion

"My conclusion, then, is that in the financial domain, past principles and rules need to be critically examined," Vincent said. "Within the current automated processes, an adjustment in invoice matching can achieve a lot. With (big) data analysis and the right analytics, the math is quick. Good luck."

Never miss a blog again?

Sign up for the newsletter and receive the latest blogs automatically in your mailbox every month. 

Sign up