Google Attribution is Evolving
Marketing is typically an expensive process and it is only really possible to generate results when you have data as to who, what and where your ads are going. And what the ROI is for that.
With ecommerce websites it is possible to track sales and revenue, however it is not that easy to attribute this activity to particular marketing channels and campaigns. Some data cannot be tracked traditionally, such as sales from phone calls. There are issues in cross-device tracking, and in the attribution of different channels when more than one is involved.
Which Model – First, Last or Assisted?
To illustrate that you can see in Google Analytics in the Conversions -> Attribution area there is different models, based on the order of the clicks. For most advertisers, including us, we use the ‘last click’ model. This means that if a customer clicks an ad, comes to the site, leaves, then searches and comes back by an organic listing, or remembers the url and goes direct, and then they convert, it doesn’t get chalked up as a PPC conversion. This happens only if the ad was the last click in the buying process.
Comparing the different models, PPC as the *first* time someone found your site usually has much higher conversion rates and ROI. So Last Click models seem to be underestimating the value of our advertising. Nevertheless, its the most direct means of attribution, and the least contestible, so its the standard that is widely used.
It is hard for us to evaluate cross-device conversions, where a person may have come across a business on one device, and later gone back on another device and converted. So this means we may not know the true ROI of advertising on different platforms and devices. Mobile advertising usually shows much lower ROI for this reason.
Now Google has been busy beavers and announced improvements in its tracking abilities. Unfortunately it requires a leap of faith to trust some of it, since it uses machine learning.
Google’s New Attribution Service
It also is improving the tracking of offline sales, for example, by emails signups who later purchase.
Google is going further than this though with offline tracking. It’s actually collecting data from major credit card companies. It now collects data from 70% of credit and debit card transactions in the United States. This potentially will give us the ability to estimate the conversion value of phone calls generated from ads.
To protect privacy Google does not collect details on items purchased, and all data exchanged is encrypted.
These tools are now available to markets on the new Google Attribution platform, and they are free.
However, this tool only evaluates the impact of paid clicks generated from the AdWords platform and not from other paid media.