Results: PayPal as a primary payment provider

It’s now been just over 6 months since I made PayPal the primary payment method on my ecommerce sites in an effort to reduce losses to fraud (see Will 53% of customers switch to PayPal and Betting on Igor). It takes a long time to see the effects of something like this, since chargebacks tend to occur weeks or months after the fraudulent transaction takes place. That was part of the problem — by the time someone realizes their credit card was used by someone else, the service has already been rendered, so the loss is not just the original payment but also the costs of the service sold. I’ve now been able to take a closer look at the long term effects of sending all new customers to PayPal rather than accepting credit cards directly.

There are a few metrics that were easy to compare: sales volume and number of chargebacks received. Unfortunately, my initial analysis didn’t hold up to the long term: conversion rate did drop slightly when the option to pay by credit card directly was taken away. Sales in the 4th quarter of the year were lower than the 2nd and 3rd even though December is usually the strongest month. The drop was not drastic, however, at about 7%. Even then, every month of 2007 had significantly higher sales volume than 2006, so I’m in good shape aside from the fraud issues.

In terms of raw number of chargebacks, the switch was successful. In the first six months of the year, I received 94 chargebacks with fraud-related reason codes, while only 34 came in the six months following the switch over. There were more than 5,000 payments during that period, so the chargeback rate fell quite drastically and is now below 1% as it should be.

In reality, the analysis is more complex, as PayPal uses a number of different ways to handle fraud in their system before chargebacks occur. The fraud rate did not actually drop significantly, it just changed in nature. There were over 100 transaction investigations conducted by PayPal in 2007, and 70 unauthorized payment claims. This is when a payment is held and the balance removed from the receiver’s account until PayPal completes a review of the transaction. These are a new source of work for me, as I have to deal with several per week. The funds aren’t available until I respond to the investigation through my account with information about the order associated with the payment and any communication with the payer. If things check out, the payment is released to me, and if PayPal believes there’s a risk issue or the payment is unauthorized, it’s reversed to the sender’s account.

It would seem from this that those who perpetrate payment fraud, using both stolen credit cards and PayPal accounts, are not at all intimidated by being sent to PayPal for payment, and are just as willing to test their cards there as directly on a smaller site like my own. The benefit to me, though, is that these investigations PayPal launches most often begin within hours of the payments arriving. With this, I can put a hold on the order until the investigation is completed. If the payment is reversed, I don’t pay any chargeback fee, and I even regain the payment fees on the original transaction.

Compared to chargebacks, where even if I somehow had yet to provide service, I lose the processing fees, a chargeback fee, and have my chargeback ratio increase, the investigations are worth the extra work. I also have the added benefit of once again qualifying for the 2.2% merchant rate at PayPal for processing $10,000-$100,000 per month every month. I was just under that limit when it was increased before making it the primary payment provider.

Because I’m losing the full cost of providing service to fraudulent orders less often, profits have actually increased despite the slight drop in sales volume. In the end, sending new customers to PayPal for payment was the right decision for my situation.