Home/Analytics/How to highjack eCommerce tracking from Google Analytics for non-eCommerce businesses?

How to highjack eCommerce tracking from Google Analytics for non-eCommerce businesses?

Blue2purple’s contribution to the GAUC BE

With this case we want to show the power of Google Analytics and how you can manipulate internal reports to reflect your business regardless if you are an eCommerce, eLead or Content-creator.

To get the most out of your Analytics, it’s essential that you know what information you want to get, what should you focus on? Don’t think on the how to get there, it will only block you.

In our case, we are working with an Affiliate, who promotes thousands of different products on his website and directs visitors to the best shops to buy the presented products. What we wanted to know was quite easy: how much money do we get from which click. If you look into the customer & data journey, it becomes clear that something sounding so easy isn’t easy at all.

The customer journey

To illustrate the difficulty of this case, we will start of with a brief explanation of the customer journey:

The visitor arrives on the website via paid or organic campaigns or is already a loyal visitor (direct traffic). He browses some products he likes (let’s say shoes for example) and finds 3-4 different shoes that look amazing to him. After reading some details of the products he decides to buy a pair of shoes that convinced him. So he clicks on the product and leaves the website of our client. At this point the customer journey on our client’s website is finished, but it continues on the Merchant Store’s website. There a sale will take place or not. Only if a sale takes place, our client will be paid a commission. The different affiliation networks will send us a report with the amount of conversions, the commission generated and a variable that we communicated to them. So our mission was to link the click to the Merchant Store with the commission generated via that click.

How did we do it

First we tracked via an event the click to store, with a fixed average value. To optimize paid campaigns based on this data, we decided to install an ecommerce tracking, so we could optimize on ROAS. The transaction also had a fixed value for each click. As this method doesn’t reflect the real value of the click, we decided to go further: Push back transaction-data given by the networks into Analytics.

To do so, we changed the ID pushed to the networks, so that it reflects the Google Analytics Session ID. Than we build a hit with the measurement Protocol to push a transaction based on all the information we gathered. Thanks to the Session ID stored in the Affiliation cookie, we were able to link a click with an actual transaction and so, we were able to calculate correctly the monetary value of each click.

Attention Points

Before we automated the process, we validated our hit via the hitbuilder. It is a handy tool to test your queries and validate the output you will get. Thanks to it easy to use interface, we didn’t have to try directly server side the pushes. We recommend to use it always before automatizing a hit via the Measurement Protocol.

It’s important to know that every push you do with the measurement protocol creates a new session, so all UTM’s are set (or not). We didn’t precise the source or medium in our push, so all transactions are reported with the source/medium direct. We wanted to validate first our transaction push method with the focus only on the transaction nothing else and thanks to the build in attribution modelling in Analytics, we are able to correctly attribute the transactions to the correct last-click source.

What did we do with this awesome information?

First, we created a dashboard that allows us to see which are the top products & shops that drive transactions.

We optimized the paid campaigns to focus on products that converted very well and so we optimized the ROAS.

And last but not least, we gained useful insights to optimize the UX of the website to get a better performing website.

 

 

Share with a friendShare on Facebook
Facebook
Share on LinkedIn
Linkedin
Tweet about this on Twitter
Twitter
By |2018-10-25T14:51:31+00:00September 1st, 2016|