Preventing Fraud

Nobody wants to see fraudulent orders, and Givy tries hard to prevent them from impacting your business - find out how.

Updated over a week ago

Givy employs a variety of techniques to detect and prevent fraud using our app. Several of these settings are configurable, allowing you to decide how best to handle potential fraud.

Order Risk Level

The order risk level is a rating which is calculated by Shopify when a customer completes an order. It looks at things like address mismatches, suspicious IPs and locations, use of VPNs, and several other factors to determine whether an order might be fraudulent.

Shopify then assigns a value of LOW, MEDIUM, or HIGH and flags the order for Givy to see. Depending on the level you choose, we either flag High risk orders only, Medium and High risk orders, or no orders at all (this is best when testing on a development store).

Gift Cards Used

Typically, there isn't a good reason to use a gift card to purchase another gift card. This is sometimes used by malicious customers who have found a loophole in one of your processes where they can get many gift cards for small amounts.

It's also possible for customers to use gift cards to purchase gifted products. While there may be a valid use case for this, we recommend keeping this setting turned on.

We recommend keeping both of these options turned on.

Discount Codes Used

We recommend that merchants update their discount codes to not be valid for the gift card product Givy creates on the store. The one exception to this is if you want to sell gift cards at a discount as an incentive to buy them.

This setting will flag any orders where the gift card was discounted using a discount code.

How to Handle Flagged Orders

When a new order is generated which is triggered by your fraud settings, you will receive an email (if enabled), and the order will be flagged with a red fraud tag on the Gifts page in Givy.

We recommend reviewing the order in your Shopify admin, and fulfilling the order if it appears to not be fraudulent.

Did this answer your question?