Manual ReviewSlower, More Expensive… So Why Do Merchants Still Rely on Them?

April 24, 2023 | 10 min read

This image was created by artificial intelligence using the following prompts:

A professional red-headed woman is sitting at her desk and typing diligently on her computer, looking at her monitor, she is in deep thought. The background is an office, with other people working on their computers as well, people are walking and working all motion blurred, business. Pops of Red and teal, high camera angle, dynamic composition, depth of field, well lit.

Manual Review

In a Nutshell

If a computer can do a job faster, that’s the way to go… right? Well, faster isn’t always better. There are still a few areas where old-fashioned human insight can win out over artificial intelligence. Case in point: manual fraud reviews. In this post, we’ll talk about manual reviews, including how they work, and why AI won’t make them obsolete any time soon.

Manual Review Can Identify & Stop Potential Fraud That Automated Tools Might Miss

Manual review is a pretty simple concept. Retail orders are checked over by a human for signs of fraud before being processed. If the order looks good, the transaction goes through. If the order raises too many red flags, though, it can be rejected.

This manual fraud review process has mostly been replaced by computers. These automated fraud tools speed up the process. But, are they as accurate? Will a time ever come when we can finally put manual reviews out to pasture? Let’s find out.

What is Manual Review?

Manual Review

[noun]/man • yoo • uhl • ree • vyoo/

A manual review is the process of subjecting a transaction to human oversight. It can occur at any point in the transaction process at which the transaction is manually reviewed, as opposed to automatically reviewed using computer algorithms.

The phrase “manual review” could apply to any number of situations. In this case, however, we’re talking about going through credit card orders on an individual basis and looking for signs of fraud before the transaction is approved.

Once upon a time, this was the only pre-payment option for identifying fraudulent transactions. While it may have worked for local stores handling their own credit accounts, it’s not practical anymore. There were over ten trillion credit card purchases completed in 2023. So, checking every order individually simply won’t work any more.

Sophisticated, rules-based detection programs are exponentially faster, but they’re not perfect. Relying on automated fraud detection alone can be problematic, as computers may overlook warning signs that a human would catch, or reject legitimate orders by mistake. Conducting manual fraud reviews as a supplement to automated tools can help address these shortcomings.

How Does a Manual Fraud Review Work?

Manual reviews require a team made up of dedicated, trained experts who can check incoming orders for warning signs associated with fraud. Agents review order details, then make “yes-or-no” judgment calls on the legitimacy of individual transactions. 

There are a number of different methods of validating transaction data. Sometimes a single step can demonstrate the order is fraudulent. More typically, reviewers use a combination of techniques and indicators to analyze information made available by the consumer, including:

  • Performing reverse lookups on name, address, phone number, or email address.
  • Contacting the buyer through phone, email, or SMS text messaging to validate their identity.
  • Comparing cardholder information provided by the buyer against information the bank has on file.
  • Checking historical purchase records to see if there is any other activity from this consumer.
  • Attempting to validate shipping address, credit card number, email addresses, etc., based on past records.
  • Reaching out to request identity verification before approving and submitting the transaction for processing.
  • Rechecking transactions initially declined by automated systems and attempting to validate them.
  • Reaching out to buyers to request additional transaction information, or even alternate payment methods.
  • Conducting investigation via social media or other public records to attempt to validate a user’s profile.
IMPORTANT!

Open-source intelligence (OSINT) refers to collecting legally obtained verbal, written, or electronic information, including data found on social media sites, public records, and so forth. It is perfectly acceptable to use such information for informed decision making, but the time it takes to gather the data can make it impractical.

As you can see, there is a lot of time and energy invested in the manual review process, even for a single transaction. Despite all the information, many decisions are still finalized based on the reviewer's experience and “gut feeling.”

While that may sound a little random and unprofessional, it’s actually the true reason for manual reviews. A human making a decision entirely based on a strict rule set would be no different from a computer. Human insight and reasoning is the entire point of the manual review.

Why Manual Fraud Reviews are Still Relevant

Most providers of fraud prevention automation tools boast about AI capabilities. That makes sense; after all, an automated system can do in seconds what manual reviewers can take several minutes or more to do.

But, there are many situations in which relying on an automated system doesn’t make sense. People with few credit card orders, for example, or those with high average tickets may be better served by reliance on manual reviews.

Consider this scenario: a merchant sells custom, hand-made guitars. These artfully produced instruments cost upwards of $10,000 or more. At that price point, they only sell between five and ten instruments in a given month.

Having a high average ticket amount can put your business at high risk for fraud and chagebacks. Talk to us about developing a comprehensive fraud prevention strategy.REQUEST A DEMO

In the above example, automated fraud detection would not be the best choice. Manual review is a far better option because the small number of orders don’t justify the cost of buying the automated tool. Also, with that much money at stake, the merchant would want a hands-on, highly detailed analysis of each buyer, plus the payment used, the timing, and so on.

Of course, this is the exception, not the rule. Relying only on manual review to vet every single buyer simply isn’t possible for most merchants. As we’ve already discovered, however, a totally automated solution isn’t ideal either.

That raises the question: how do you find a balance?

There are benefits and downsides to both positions, obviously. But, as we saw in our example, whether any one factor is a plus or a minus largely depends on the merchant’s needs.

Manual reviews will necessarily deliver a personalized experience. This could be more important to a “mom-and-pop” store than a giant retail chain. For a global corporation processing millions of transactions, though, speed and cost may trump other considerations, meaning AI is the right way to go.

When Should A Business Use Manual Reviews?

Given all the variables, it’s almost impossible to create an apples-to-apples comparison between manual and automated reviews. If there is one universal answer here, it’s that the best solution is a combination of the two in most situations.

To capture the most potential fraud, automated review filters need to be set with fairly strict parameters. Yet doing so can cause a dramatic increase in false positives. A smarter approach — one that most merchants take — is to implement automated reviewing for the cut-and-dry cases, while manually reviewing transactions that fall in a gray area.

Looking to build out or optimize an existing manual review process? Here are a few questions you should ask:

What Do You Need?

This seems obvious, but it’s still important to ask.

Automated fraud filters are designed to identify third-party (or “criminal”) fraud. That’s outside parties using stolen payment credentials to make unauthorized purchases. With modern eCommerce, however, more threats are coming from post-transaction, first-party sources.

Automation, backed by manual fraud review, can help you determine whether your disputes are coming from genuine criminal fraud or not.

Can You Afford It?

There is no way around the fact that manual transaction reviews demand a lot of time and resources. You’ll need to take a hard, honest look at your capabilities, and your company’s capacity for handling the workload.

You may have to hire more staff, or even outsource some process. Will the extra cost have a reasonable return on investment, though? Don’t forget that in-house staff will require additional training, as will each new hire.

Is Your Fraud Rate in the “Danger Zone”?

Your fraud rate gauges the number of fraudulent transactions you allow to be processed, as a portion of your total sales in a given period. An excessive fraud rate could land you in a fraud monitoring program, which would be costly and could also restrict your operations.

If your fraud rate is approaching the thresholds allowed by Visa, Mastercard, or another card brand, then conducting manual reviews backed by automation could help you keep that issue in check.

How Will Manual Review Impact Your Customers?

Modern consumers have grown accustomed to instant responses. Will your customers accept the extra time required for authentication? Or are they apt to get frustrated and try their luck with another vendor? These questions will give you direct insight into how your customers would respond to delays caused by manual review processes.

If you’re not sure, it may be worth a small survey of your clientele.

Is There a Better Option?

Putting other verification methods in place may work better for you. For example, two-factor authentication can dramatically lower the odds of successful fraud purchases, while keeping checkout friction in check.

Again though, most merchants find that any single fraud prevention solution is likely to be ineffective on its own. Using manual review as one piece of an overall fraud strategy will usually deliver the best results.

A Comprehensive Strategy for Better Results

When considering fraud screening, it’s really not a simple question of “human intelligence vs. artificial intelligence.” You’ll get much better results from using the tools in tandem, maximizing effectiveness while minimizing risk.

In most situations, the highest percentage of transactions can be easily analyzed and decided using an algorithm. By setting acceptance parameters, the computer can accept or reject the clear-cut cases, while still flagging the few that require human attention.

The role that manual fraud reviews play is dependent on intelligent fraud and chargeback source detection, as well as a comprehensive chargeback management system. Only a reliable, end-to-end fraud detection strategy can help you:

This is the only way to see true revenue recovery and sustainable growth. Ready to learn more? Click here and get started today.

FAQs

What is a fraud manual review?

A manual fraud review is a fraud prevention tactic that consists of a human being reviewing incoming fraud data to determine a specific response or course of action. Manual fraud reviews are an essential part of any fraud prevention strategy, as automated technologies are only capable of a certain level of nuanced thought or selection.

How long does a manual review take?

Most manual fraud reviews can take place in just a few minutes. However, depending on the complexity or severity of the incoming issue, a thorough review and investigation may take several days in extreme cases.

How does a manual fraud review work?

Manual reviews are typically conducted by trained analysts or data experts with access to pending transaction logs. When used in tandem with an automated process, any transactions flagged are examined by the reviewer, who may use other data to make an informed decision on whether to accept the order.

How do you use the results from a manual fraud review?

In the short term, the results from a manual review can be used to validate or decline the associated transaction. For long term use, information gained from a manual review should be fed into your automated systems for improved automated decisioning.

Why are manual reviews important in fraud detection?

Simply put: computers can’t think like humans.

Despite the breadth and scope of emerging technologies, manual fraud review will continue to be an asset for the foreseeable future. Machine learning and AI have come a long way in the past few years, but a machine is largely incapable of seeing and diagnosing gray areas, particularly those that revolve around human decisions.

Like What You're Reading? Join our newsletter and stay up to date on the latest in payments and eCommerce trends.
Newsletter Signup
We’ll run the numbers; You’ll see the savings.
triangle shape background particle triangle shape background particle triangle shape background particle
Please share a few details and we'll connect with you!
Revenue Recovery icon
Over 18,000 companies recovered revenue with products from Chargebacks911
Close Form