What is the cause of the invisible "falling basket"? The Relationship Between EMV 3-D Secure (3D Secure 2.0) and authorization Approval Rates

The basket drop is not necessarily due to the customer's convenience

As an improvement measure for the e-commerce business, "cart abandonment countermeasures" is always one of the priority issues. Whether it's improving product pages, reviewing shipping terms, or simplifying input forms, UI/UX improvements are certainly effective approaches.

However, there may be many cases where incompleteness that occurs "after you press the payment button" may be overlooked.

In recent years, with the introduction of EMV 3-D Secure (hereinafter referred to as "3D Secure"), if you feel that "sales growth is somehow slower than before" or "the number of completions is not increasing even though the number of additions to the cart has not changed", it may be due to the system's judgment in the payment flow.

In this article, we will focus on Credit card payment among payment method​and break down the cart abandonment structure. On top of that, we would like to think together about a realistic approach to improvement.

Disassembling the Structure of a Basket Drop: 4 Steps

It's hard to talk about "abandoned basket" all at once, it's actually a few different layers stacked up. Let's try to organize the journey from a customer to a payment in their cart in four steps:

Step1: Leaving due to UI/UX

It is the most widely recognized area. This is the stage where the friction of the interface reduces the customer's desire to buy, such as the large number of input items, forced membership registration, psychological resistance to shipping and handling fees, and inadequate smartphone support​.

There is a lot of information on measures for this group, and I think that many businesses are already working on it.

Step2: 3D Secure Leave the authentication process

This is a case where the customer interrupts the operation during the authentication process of the 3D Secure. You may have trouble receiving a one-time password, feeling anxious about transitioning to the credit card company's authentication screen, or unable to remember your password.

There is also an aspect of leaving due to "customer will", but the frequency of occurrence may vary depending on the ease of understanding and transition design of the authentication screen.

Step3: The identity of the invisible basket -3D Secure Incomplete due to authentication judgment

From here on, this is an area where the customer is trying to complete the operation, but is stopped by the system's decision.

3D Secure basically calculates a risk score for each trade and determines whether the risk level is low, medium, or high. If the attribute data passed to the card issuer (issuer) is not sufficiently linked in this determination, there may be a lack of material to judge the reliability of the transaction, and it may be more likely to be evaluated as a high-risk transaction than medium-risk.

This is a mechanism to prevent malicious fraud, but there is also a possibility of "false positives" that can stop legitimate customers who are originally willing to make a purchase. This is also an issue that the industry as a whole is continuously trying to improve its accuracy.

Step4: authorization Incomplete due to "Judgment"

3D Secure Even if you pass verification, you may be stopped by subsequent authorization operations. The main causes include insufficient credit card limits, expired cards, and incorrect card numbers.

As in Step 3, this is also a case where the transaction is not completed on the system side regardless of the customer's will.

[Reference value] payment Know the "guideline" until completion

Here are some references for various processes in Credit card payment. In fact, please note that the numbers vary greatly depending on the industry, product, time, customer base, etc.

*Reference values are based on the average value of GMO-PG's online payment service "PG Multi-Payment Service" merchants for a specific period.

Organize the flow of "yield"

Credit card payment goes through two main decision steps from the time the customer presses the payment button to the time they complete. And the number of cases that pass the previous step is the denominator of the next step. In other words, it has a "funnel" structure in which the number of cases is narrowed down little by little at each step.

The image is as follows.

*100 cases × 3DS certification success rate 92.7% ≈ 92 cases
*92 cases × authorization Approval rate 95.7% ≈ 88.0 cases

List of reference values by process

Point: The denominator of the authorization approval rate is not the number of Credit card payment runs, but the number of 3D Secure passed*. Therefore, it is difficult to understand the final payment completion rate by looking at the 3DS authentication success rate and the authorization approval rate in isolation. Only by multiplying the two numbers can we see the overall picture of "what percentage of customers who press the payment button are payment complete".
* When requesting 3D Secure for all transactions

In addition, as a breakdown of the 92.7% success rate of 3DS authentication, about 50% are processed without additional authentication due to the 3DS authentication low-risk judgment. The remaining approximately 42% are approved through challenge certification (with additional certification). In 3DS authentication risk determination, the higher the low risk rate, the lower the customer's operational load, which may be an indication that the customer is able to achieve a payment experience.
*Each risk determination in 3DS certification is based on the figures of GMO-PG member stores in the same period.

You can learn more about funnels across payment flows here.

Reference estimates: How meaningful is an improvement in approval rates?

This is just an example, but please refer to it to capture the numbers with a sense of reality.

Assumption: 1,000 attempts per month, e-commerce site with payment an average customer price of 10,000 yen

*The number of payment completed per month is an approximate value calculated based on "payment number of executions × 3DS authentication success rate × authorization approval rate".

In terms of advertising costs, if you try to compensate for the loss of opportunities of 190,000 yen per month with advertising, it may cost more than the same amount. On the other hand, improving payment flow can be expected to have a continuous effect in a single effort, so it may be an approach worth considering from the perspective of return on investment.

(*The actual impact varies greatly depending on the product and customer base business scale)

Steps to Improvement: Organize Information and Improve Accuracy

So, what are the possible approaches to reduce payment incompleteness due to "3DS authentication false positives" in Step 3?

Data linkage as "trust building" with card issuers (issuers)

The accuracy of risk-based authentication in 3D Secure is passed to the card issuer (issuer) through the payment systemQuality and quantity of attribute dataIt tends to be greatly influenced by.

For example, by linking the following information, you may build up evidence that the transaction appears to be a legitimate customer, increasing the pass rate without additional authentication.

• Cardholder's name
• Email address
• Phone number (mobile, home, work) and country code
• Shipping address(match or mismatch with billing address)
• Purchase history and account information(first-time purchase or existing customer)
• Characteristics of cartcontents (price range, product category, etc.)

The 3D Secure specification states that more than 100 attribute items can be linked. However, in many cases, it is not feasible to have everything at once from the perspective of development resources.

A Realistic Way to Get Started: Organize Your Implementation Priorities

It is realistic to start with the reliable linkage of "member attribute information" such as name and contact information mentioned above, and then gradually expand to link more advanced behavioral data such as "shipping address history" and "past purchase Actual".

In particular, some international brands strongly encourage and make it mandatory to link such attribute information, and these are "data held only by merchants" that cannot be automatically completed by the payment system. Where to start and what your company's current cooperation status is depends on the operator's system environment. First of all, visualizing "what data is not being sent" is a big step towards improvement.

Summary: Design the "best payment experience" together

payment doesn't end with introductory. Customer purchasing behavior and fraud trends are changing every day, and the measurement parameters need to be continuously reviewed accordingly.

If you feel that the number of cart drops has increased since the introduction of 3D Secure, it may be a sign that you are reviewing your system settings and integration data.

GMO-PG analyzes authorization approval data and 3DS authentication data and provides accompanying consulting to find the optimal parameters and linkage items according to the situation of e-commerce businesses. Even if you are at the stage of "I don't know how to read the numbers", please feel free to contact us.

payment Improving the experience is a critical component of business growth that is directly linked to the completion rate of purchase. Why don't you start with visualizing your company's "invisible walls" together?

Related Articles

• authorization What is the approval rate? The funnel from cart entry to sales closure and 3 improvement steps to maximize profits
• 3D Secure One year since it became mandatory. Guidelines 6.1 show both fraud control and sales improvement
• 10 questions answered about EMV 3-D secure! Explanation of the mechanism and benefits
• What is authorization (authorization Nation)? A thorough explanation of how Credit card payment works