{"id":132357,"date":"2017-01-03T11:30:22","date_gmt":"2017-01-03T16:30:22","guid":{"rendered":"https:\/\/www.practicalecommerce.com\/?p=132357"},"modified":"2022-11-04T16:56:10","modified_gmt":"2022-11-04T20:56:10","slug":"Using-Google-Analytics-to-Identify-Fraudulent-Orders","status":"publish","type":"post","link":"https:\/\/www.practicalecommerce.com\/Using-Google-Analytics-to-Identify-Fraudulent-Orders","title":{"rendered":"Using Google Analytics to Identify Fraudulent Orders"},"content":{"rendered":"

Fraudulent transactions can occur for most any ecommerce company. Larger companies can invest in sophisticated fraud protection services. But smaller merchants sometimes attempt\u00a0to detect fraud themselves.<\/p>\n

Back in 2014 we addressed do-it-yourself fraud detection. The article<\/a>, from columnist Erica Tevis, owner of an online gift store, has helped many smaller merchants.<\/p>\n

But\u00a0Google Analytics can help detect fraud, too. I’ve used it for my own niche ecommerce site (selling imported food products), which occasionally encounters fraudulent orders.<\/p>\n

In this article, I’ll explain how to use\u00a0Google Analytics to examine\u00a0three characteristics of suspect transactions, to help detect fraud.<\/p>\n

Use Google Analytics to Identify Fraudulent Orders<\/h3>\n

Characteristic 1: The transaction was not recorded in Google Analytics.\u00a0<\/strong>If you suspect fraud, see if the transaction was reported in Google Analytics. If not, it could be that the purchaser is using a browser add-on or he\u00a0disabled JavaScript to block the sessions and transactions in Google Analytics.<\/p>\n

To find a transaction in Google Analytics, go to \u201cConversions > Ecommerce > Sales Performance\u201d and view the list of Transaction IDs.<\/p>\n

\"Go<\/a>

Go to \u201cConversions > Ecommerce > Sales Performance” to view Transaction IDs. Click image to enlarge.<\/em><\/p><\/div>\n

Select a date range that should include the suspect order. \u00a0If you see potential transaction matches, enter the Transaction ID in the search box for more details.<\/p>\n

\"If<\/a>

If you see potential transaction matches, enter the Transaction ID in the search box for more details. Click image to enlarge.<\/em><\/p><\/div>\n

Characteristic 2:<\/strong> The geo-location of the transaction does not match the billing address.<\/strong> Follow the instructions above to find the transaction in Google Analytics. If it is present, click on the Transaction ID to display the products in the order, then click on \u201cSecondary Dimension\u201d and enter \u201cCountry,\u201d \u201cRegion,\u201d and \u201cCity\u201d to identify where the purchaser was located.<\/p>\n

\"Selecting

Selecting country of the transaction can identify the location of the purchase, to match against the billing address he provided.<\/p><\/div>\n

—<\/p>\n

\"Example<\/a>

Example results when \u201cCity\u201d is Selected as the Secondary Dimension. Click image to enlarge.<\/em><\/p><\/div>\n

Google Analytics uses IP addresses to identify geo-location. It is not precise.\u00a0But it can provide clues to the actual location of the purchaser at the time of the transaction.<\/p>\n

People frequently purchase gifts, so that their billing address is different from the shipping address(es) in the order. That’s why it’s important to compare the billing address, not the shipping address, to the Google Analytics geo-location.<\/p>\n

Erica Tevis, in her aforementioned article, stated that expedited delivery requests can be suspect. Couple expedited delivery requests with a geo-location in Google Analytics not matching the billing address and you may want to flag the order.<\/p>\n

Geo-location information in Google Analytics that is \u201c(not set)\u201d could also be a suspect. In Google Analytics, \u201c(not set)\u201d means the location of the\u00a0IP address cannot be identified. But\u201c(not set)\u201d could also be a legitimate sale. So \u201c(not set)\u201d alone should not be the sole determining factor.<\/p>\n

\"These

These transaction IDs with \u201c(not set)\u201d as the Region are, in fact, legitimate.<\/p><\/div>\n

Characteristic 3:<\/strong> Purchaser spent much\u00a0time on the website \u2014 especially on the payment page.<\/strong> If you find a suspect transaction, create an advanced segment for that Transaction ID, as follows.<\/p>\n

First, click on \u201cAll Users\u201d at the top-center of the reports to open the advanced segments window.<\/p>\n

\"Click<\/a>

Click on \u201cAll Users\u201d at the top-center of the reports to open advanced segments.<\/p><\/div>\n

Next, click on \u201cNew Segment,\u201d give it a name (such as the Transaction ID) and click on \u201cConditions\u201d on the left-hand side. Select \u201cTransaction ID\u201d for the Filter and enter the Transaction ID in the box, then save.<\/p>\n

\"Select

Select \u201cTransaction ID\u201d for the Filter and enter the Transaction ID in the box, then save.<\/p><\/div>\n

There are two locations to view a\u00a0user\u2019s activity.<\/p>\n