Several tips and tricks have already been presented this year whether A/B Testing, Google Analytics parameters or yet again Data Studio. This week, I will broach a subject relative to tracking, an issue that concerns practically all your e-commerce clients, even if most of the time, they don’t know it.

I came across this problem a little while ago now when one of my clients questioned several amounts for shipping that appeared in the sales performance report. In fact, he who only offered one shipping cost regardless of the order, let’s say 3€ for instance, observed higher figures (6€, 9€, 30€). First concern: is it a tracking problem only or is there a bug on the site where we’re really billing the user for shipping costs?

Digging a little, we realized that on these orders, the abnormal shipping costs were multiples of the original price and that the duplication did not concern shipping costs only but all sales revenues generated by the concerned orders. Second lesson, these duplicated orders were observed for the most part on mobiles.

futurama_phone

In fact, this is what happened: on mobiles, applications are rarely closed but rather left in the background, and on the browser application, the tabs are also rarely closed after browsing. When the browser is re-opened, the pages which are still present, are reloaded and the transaction hits are resent, as a order confirmation page of course.

Natively, Google Analytics will de-duplicate transactions with the same ID in the same session. However, when the same transaction is resent after expiration of the first session, the de-duplication does not take place.

macgyver

To get around the problem, one need only put a little hack in place that consists in memorizing the last transaction sent from the browser (using local storage) and to send the transaction only if it has not already been accepted.
You’ll tell me that if the site has different shipping costs depending on quantities, destinations, etc., it will be more difficult to find this order duplication than in the customer case I previously mentioned. To do so, one need only create a simple personalized report that contains the transaction ID and the number of associated transactions. In that way, it will be easy to identify whether certain orders were recorded several times.

transactions_report

Have a great weekend and come back soon for more tips & tricks!

Anthony Le Crom

Consultant Analytics & Conversion

More Posts

Follow Me:
TwitterLinkedIn