In a move that will delight some and puzzle others, Google will soon allow AdWords to choose which AdWords clicks are credited with conversions.
Currently, in AdWords the last click gets all the credit (and this has long been the default for Google Analytics and other web analytics platforms).
Google has now announced that, later this month, advertisers will be able to choose from a selection of attribution types: First Click, Linear, Time Decay, Position-based, Data-driven and, of course, Last Click. These are applied to each type of conversion tracked in AdWords but will only be available for Search campaigns.
While Linear attribution distributes conversion credit evenly across all clicks, Time Decay assigns greater value to the clicks closest to the time of the conversion. Position-based applies 40% of the credit to the first clicked ad and keyword and 40% to the last, distributing the remaining 20% evenly to the other clicks.
Data-driven attribution (which is still in beta) is limited to high volume accounts — those with at least 20,000 clicks and 800 conversions within 30 days. This attribution model, in Google’s words, “distributes credit for the conversion based on past data for the conversion action”. Data-driven attribution uses machine-learning in evaluating ad interactions, order of exposure, ad creative and other factors in apportioning conversion credit to keywords and clicks.
Select one of the new attribution models and conversion credit will be reassigned for Google.com search traffic. There will not be any post-processing of earlier data, and it’s important to note that:
- If you pick a new attribution model, choose wisely and stick with it.
- Last click attribution remains in effect for Search Partners traffic, and for Display and Video campaigns. It is also the default if you select Data-driven attribution and in any 30-day period lack sufficient traffic and conversions for Data-driven attribution to work.
Why should you care?
Sexy for some, but a trip to the dentist for others. Yet getting conversion attribution right for your business shouldn’t end up in the too-hard basket.
The longer the typical time lag between a visit to your website and a conversion, the more likely you need to be working with something better than Last AdWords click attribution. And the more often people interact with your site before converting, the more likely you’re overlooking important interactions that contributed to the conversion.
On the flip side, AdWords remarketing, in helping convert visitors who have already visited your site, may be getting more credit than it really deserves.
Funnily enough, there was a time when all web analytics platforms credited the first click with the conversion. The eventual move to last click came with recognition that most conversions didn’t happen on the first visit for most sites.
Fast forward to 2016 and not only are consumers visiting more websites as they research before making an important purchase, but they are often using a number of devices (smartphone, tablet and desktop computer) before converting.
We have all been able to compare attribution models in Google Analytics Multi-Channel Funnels reports for some time now. But Last Non-Direct Click is still used for all other reports in Google Analytics (although you can change this using the enterprise version, Google Analytics 360).
Spurred by ever-rising mobile usage, Google’s introducing other attribution models to AdWords, however, in order to provide more insight into the more complex journey taken by today’s customers. Importantly, this update will give AdWords advertisers a better picture as to the value of the early influencer keywords they are bidding on.
Google has published a guide to help AdWords advertisers understand whether or not they should change from last click attribution, and what alternative model works best for their business.
At SureFire, we’ll be assessing our clients’ accounts and making recommendations once the alternative attribution options become available.
Click here for more search marketing news.
If you found this useful, please tell your friends.