How to Use Chargeback Data to Prevent Disputes

Data analysis is a key component of a successful chargeback management strategy.

With data, you can make educated, informed decisions. Without data, you are left with hunches and guesses. You don’t know why disputes happen, how to effectively stop them, or if you’ve solved the problem.

Let’s take a look at how to use data analysis to prevent chargebacks and disputes.

Each chargeback you receive provides a ton of useful information. If you analyze that information, you can gain valuable insights — like the reason the cardholder chose to dispute the transaction. Then, you can use those insights to solve problems at their source — which, in turn, prevents future chargebacks from happening.

When it comes to creating and managing a chargeback prevention strategy, you basically have two options: a strategy with data analysis or a strategy without. Here’s what each looks like.

Data analysis can be a powerful chargeback management tool — but it can also be confusing and difficult to manage without any assistance.

Here’s a step-by-step guide on how to collect and analyse your data.

Before you start doing the actual analysis, you’ll need to complete a couple of administrative tasks to get the data ready.

Start by creating a calendar. How often will you review your data? Daily? Weekly? Monthly?

And who will be responsible for collecting and reviewing the data?

Make sure responsibilities and expectations are clearly outlined so you don’t miss valuable opportunities.

There are a lot of data elements you’ll want to collect. The more data you have, the more comprehensive your analysis will be.

Here’s a list of data elements we suggest you gather for each chargeback you receive.

• The card brand (Mastercard, Visa, etc.) that filed the chargeback
• The merchant account ID (MID) used to process the disputed transaction
• The chargeback amount (which may be different from the original transaction amount)
• The chargeback reason code
• The marketing campaign that generated the disputed purchase
• The cardholder’s country
• The issuer’s bank identification number (BIN), the first 6 to 8 digits of the account number
• The product name or stock keeping unit (SKU)
• The date the chargeback was filed

Your data is probably spread out across several different platforms. Here are some places you will probably need to check and the data you’ll want to collect from each.

• Your shopping cart or order management system will have information like which item was purchased, whether or not the order was the result of a marketing campaign, and where the purchase was made.
• Your merchant account portal will have your transaction and chargeback data.
• Your prevention solution provider portal will have your prevention case data like Ethoca alert outcomes and RDR counts.

Once you have all your data, you’ll want to consolidate it — probably in a platform like Excel or Google Sheets — so you can cross reference different variables.

Your reporting will look like this.

If you use prevention solutions — like prevention alerts and RDR — you’ll want to repeat this process with your prevention data. Note the same variables for each prevention case — card brand, merchant account, amount, marketing source, country, BIN, product type, and date.

Once you’ve collected all your information, you’ll have about a dozen data elements for each chargeback.

Your data alone probably won’t tell you much. You need to analyse it in context to start gaining insights.

Break down your data by different variables — marketing source, country, price point, etc. — and then calculate your chargeback-to-transaction ratio for each associated element. For example, the chargeback-to-transaction ratio for Canada, U.S., and Australia.

Usually, tables of numbers aren’t super easy to review. Charts, on the other hand, are more visual, making it easy to spot issues at a glance.

Ideally, you’ll get as granular with your analysis as possible. For example, break down all your data by MID, processor, and/or card brand.

If you use prevention solutions, you’ll want to repeat this process for your prevention case data.

Here’s where all your hard work starts to pay off!

You have your data. You’ve extracted it from your different platforms and consolidated it. You’ve broken it down by different variables. And you have lots and lots of pretty charts.

But what does it all mean?!

As you regularly analyze your data, you’ll start to see norms and patterns for your business. You’ll also detect anomalies and situations that differ from the norms — signs that there are issues that need to be resolved.

Here are some examples of pattern anomalies and what they could mean for your business.

• A high chargeback ratio for low-value items could mean criminals are running cards (making test purchases to see if stolen payment information is still valid).
• High-value items might be enticing criminals because of their resale value.
• Customers might be filing chargebacks if certain items have a ‘no refund’ policy — like customized merchandise.
• Your merchandise might be lower quality than you expected.

• Issuers might auto-chargeback transactions if you are repeatedly making processing errors.
• Auto-chargebacks might happen if your business has a bad reputation or banks think your practices are fraudulent.
• A data breach at a bank might cause an influx of chargebacks from a single BIN.

• If you are automatically adding dynamic currency conversion to orders instead of letting customers opt-into the service, you might see an increase in chargebacks.
• Region-specific preferences might mean certain products are inappropriate.
• Expectations for size or material might vary by region.
• Your billing model might not be popular in a particular region, especially if you offer something like free trials or subscriptions.

• Chargebacks labeled as ‘canceled recurring billing’ could mean your billing model and policies aren’t clear.
• Your support team might not be handling refunds properly if you have lots of ‘credit not processed’ chargebacks.
• If you receive an influx in ‘unauthorized transactions’ chargebacks, you might need to adjust your fraud detection strategy.
• Your product descriptions and ads might be misleading, inaccurate, or incomplete if you get a lot of ‘not as described’ chargebacks.
• A spike in ‘merchandise not received’ chargebacks could indicate your delivery methods aren’t quick enough, your inventory is not able to keep up with demand, or your fulfilment team is short staffed.

Once you understand what your data means, it is time to adjust your policies and procedures to solve the issues you’ve uncovered.

If you make updates but chargeback activity remains abnormal, you probably have other issues that need to be addressed.

An abundance of accurate, helpful data is the foundation of an effective chargeback management strategy. If you don’t have access to the data you need, it might be time to work with a different solution provider.

Reach out to AltoPay if you are interested in creating a more data-driven chargeback strategy.