Secure and Friction-Free for the Holidays: 5 Tips to Simplify Your Checkout Process

The holiday shopping season is here. Is your business ready?

According to Gartner’s 2023 Holiday Marketing Guide for Retail, 19% of consumers planned to begin shopping for the holidays in October, while 29% more planned to begin shopping in November. That means the holiday shopping season is officially upon us! But merchants should beware that the economic uncertainty has impacted holiday shopping plans, with 80% of consumers expecting to spend less this holiday season than in previous years. In this economy, merchants need to make every customer count and a great place to start is the checkout process.

One issue that continues to hurt merchants is cart abandonment, where a customer does not complete a purchase for items they’ve put in their digital cart. There are many reasons for this, but some surveys are highlighting a particular culprit: checkout friction. One survey shows that 25% of consumers will abandon their carts if the merchant requires them to create an account before purchasing. Another 18% of consumers said they will abandon their carts if the checkout process takes too long or is too complicated.

A quick and easy checkout process seems to be a merchant’s best bet at keeping cart abandonment rates low. But how can your business balance a robust fraud prevention system with a quick and easy checkout?

5 tips for keeping your checkout process secure and friction-free.

Below are five tips from a fraud analyst on how to keep your checkout process both secure and friction-free for customers.

Forget the account – your fraud prevention team doesn’t need it.

Many merchants prefer or even require customers to create an account in order to complete a purchase on their platform. But a 2022 survey showed that 43% of customers prefer checking out as a guest. And that’s okay – your fraud team doesn’t need customers to create accounts in order to verify their purchase history. Modern link analysis strategies can make up for a lot, including linking a payment instrument, email address, or phone number to previous purchases made using those attributes. From a fraud team’s perspective, link analysis can give you all the same benefits as tracking an account’s purchase history without the customer friction.

This can be particularly useful if there are concerns that a repeat customer’s payment instrument has been compromised. If the customer has an account, tracked payment history can help a fraud analyst determine if a compromise has taken place; a customer’s purchase patterns (what, where, and how they’re buying) tend to remain consistent under ordinary circumstances. In the absence of an account history – if, for example, a customer prefers to check out as a guest – link analysis can fill the gap and provide much of the same information that a fraud analyst needs in order to make an accurate decision.

Cut down on the required info.

It’s tempting to want to get as much information from your customers as possible during the checkout process for security and marketing purposes. But an experienced fraud analyst will tell you that much of what merchants require isn’t always necessary to make a fraud/not fraud decision on a customer’s purchase. Of course, we never turn more data down, but many of the most important data attributes can be gleaned without the customer typing anything into a text box. For example, IP address and device fingerprint are both hugely important in a manual review, but both can be collected automatically during the checkout process.

That’s not to say you shouldn’t ask for any information to be input manually by the customer. There is still information you need to know in order to process the transaction. The customer must provide an email address, a physical shipping address (for physical goods being delivered), and obviously a payment instrument. There may also be other information your business model requires. But we’d recommend keeping the required text fields at a minimum. Your customers will expect to be able to complete the entire checkout process in three minutes or less.

Offer just the right number of payment instruments.

While your fraud team can and will find fraud on any payment methods you offer your customers, this is a situation where more isn’t always better. As the Payments Journal calls out in this article, customers who are offered too many payment options during the checkout process can have what we call choice paralysis, where there are so many options to choose from that the customer becomes unable to make a decision. This can result in cart abandonment. On the other hand, offering too few options can lead to a customer thinking your checkout process is inaccessible – and again, to cart abandonment.

What we recommend instead is offering between 3 and 6 payment options and tailoring them to your customer demographic. If your customers are mostly younger, they may be more open to (and in fact looking for) newer FinTech options, while older customers may want to stick to more established payment methods such as credit cards or even PayPal.

Use a machine learning model.

A machine learning (ML) model is the industry-approved first line of defense against fraud, and there are plenty of off-the-shelf ML models that you can use to monitor and approve your purchase traffic. You just need to define an acceptable level of risk, and the model will do the rest. An ML model can approve a legitimate customer’s transaction in under a second – so quickly that the customer won’t even know a security check has been performed. This will provide a friction-free checkout experience for many of your legitimate customers.

As for bad actors engaging in payment fraud, they’ll be blocked from making the purchase, which provides the security you need to keep your fraud loss down.

Use a human intelligence team.

An ML model is ideal for many of your customers, but what about the good customers who fall outside your risk threshold? An ML model can only approve or reject transactions, so legitimate customers can easily become caught up by the reject logic. This is where human intelligence comes in. Not only can a human intelligence team help you right-size your risk threshold, but it can also perform reviews of customers the ML model would have rejected – and it doesn’t even have to impact your customer’s checkout process.

The easiest (and most frictionless) way to make this work is to issue a conditional approval to transactions in this category (i.e. outside the risk threshold but not obviously fraud). From a customer’s perspective, this shouldn’t look any different from an unconditional approval.

They’ve completed their checkout process smoothly, and now they’re simply waiting for product delivery. In the meantime, your human intelligence team can immediately proceed with a manual fraud review on the transaction.

Manual review has a reputation for being time-consuming, but it doesn’t have to be; an experienced fraud analyst can perform a robust manual review in under three minutes. If the customer is legitimate, the fraud analyst can fully approve the transaction and send it for fulfillment – and you can consider it a missed false positive, since the ML model would have rejected the transaction. If the transaction was created by a bad actor, the fraud analyst can reject the transaction and stop the fraud from hitting your system after fulfillment – saving you on cost of goods, shipping, and other fees down the line.

When selling physical goods, there’s already a built-in processing delay that your customers will expect, during which time your human intelligence team can perform the review mentioned above. Even for same-day shipping, there’s an expectation that the product won’t ship instantaneously. Customers know that your fulfillment team will need time to pack and ship their purchase.

For digital goods, you can easily build in a processing delay as a part of normal customer experience, which is what Amazon does. Even a 30-minute processing delay after checkout is enough time for your human intelligence team to protect you from fraud loss – all without adding friction to your customer’s checkout experience.

If you’re interested in consulting with one of the many experienced fraud analysts on our Fraud Prevention team, we’d be happy to work with you. You can reach out to us here on our Contact page.