With the arrival of e-Commerce, customers were able to access merchant’s goods at all hours of the day from all over the world. Shopping and transacting from the comfort of their home and having products delivered to their doorstep. Today, this is commonplace and second nature to most of us. However with the development and expansion of e-Commerce with its ease of use came the ability for anonymous users to profit from stolen identities, again from the comfort and safety of their homes. This new wave of fraud required a new way of checking their customers to make sure the transaction was not going to be disputed. Looking for a solution to the growing fraud issue merchants turned to rules and rules engines, a concept which had been in use for other reasons to help.

What is a Rules Engine?

Rules engines are a backend software solution that takes predetermined actions based off of a set of criteria and parameters which are created by the merchant. For example, a merchant may not want someone to make an order of over $1000 in a single purchase. The risk level may be too high for the processor and a processor that accepts high risk companies will likely charge higher fees and rates. One way to combat this was with a rule that any order with a total value of greater than $999.99 is declined or sent to review depending on the merchant’s desire or limitations by technical capabilities. Rules engines allow merchants to control the environment that users interact with on their system system.

Of Note: When creating and implementing any new rule or rule strategy it’s important to monitor the reactions of the transactions to the new rule. It’s always possible that an unintended reaction could occur which would decline good transactions. Better yet, if you’re able test your rule out before it goes live, do so. If not, set the rule to simply, “flag” the order as opposed to out and out declining the transaction. This way you minimize customer support issues and still convert good transactions. Most rules engines allow merchants to see which transactions would have triggered a specific rule before actually taking the rule live. This becomes much more important as you increase the sophistication of the rules your system is running off of.

Best practices for merchants learning and implementing a new rules engine are to first and foremost identify the problem you’re trying to solve. By understanding what the specific problem is you’re able to more accurately implement a solution. Some general ideas and question to ask yourself and to consider when implementing new rules:

• How many times do you expect your merchants to transact on your site within a given timeframe?
• What is the average dollar order of your transactions?
• What products are or have been the target of fraud?
• What information do I expect to see for a transaction?
  • What is the first and last name, is the information gibberish?
  • What is the address, is it formatted correctly?
  • Is there valid email address
  • Is there a valid phone number?
• How often will you see the same payment information linked to different people (names) or addresses?

Of Note: A best practice for implementing new rules is by reviewing and analyzing past transactional and chargeback data. By doing this you’ll increase the probability of implementing a successful rule to prevent fraud.

As nice as it would be to implement a rules engine and then let it run and manage itself, that’s not the case. Rules engines do require some maintenance as the fraudsters can learn the parameters that have been set up and engineer ways around your safeguards and the purchase patterns of your valid customers may change over time so it’s best to keep track and monitor the activity of your rules engines.