Google says goodbye to linear, time-decay and first-touch attribution models.

London, 10. May 2023

Isabella Colli, Activation Lead at SYZYGY gives her thoughts on the impact of Googles decision to remove rule-based attribution models in favour of data-driven and last click.

On April 6th, 2023, Google announced the from June this year the withdrawal of rules-based attribution models within Google Ads and GA4.

So, how does this affect us media buyers?

For media strategists, planners, and buyers, this change comes with both challenges and benefits. Without the ability to weight user touchpoints according to business and media strategy, flexibility may be reduced for those using rule-based attribution, resulting in less control.

However, when considering the implications of flexibility and control on your paid media strategy, this change could be advantageous. That is because of changes to the customer journey.

The customer journey is not as linear as perhaps it once was, as we have observed at SYZYGY while working across a number various industry verticals. The more complex and evolving the journey becomes, the less effective it is to define rules and adhere to a rigid attribution model.

Therefore, some of the existing models such as linear, first interaction or time decay start to feel less applicable in the modern era of media buying. This is in comparison to data-driven attribution which can consider thousands of customer touchpoints to begin to tell us the picture of conversion contribution.

Can we trust Googles data-driven attribution model?

In recent years, the data-driven attribution (DDA) model has been the most effective measurement approach in supporting buying decisions and reporting across mid and upper customer funnel.

Unlike looking through the lens of other models, DDA will provide a more accurate view on the value of certain channels and tactics such as generic search terms in PPC or YouTube campaigns.

Additionally, defining the weight and value of certain channels in previous attribution models would fail at truly capturing all the signals, customer touch points and data points. Meaning important decisions could be made without the full picture being considered.

The DDA model isn’t all positive. The largest criticism echoed across the media buying world is not knowing the algorithms logic. We have no idea what weighting or decision-making Google is applying behind the scenes. We simply must trust it is correct. The DDA model needs a significant amount of data that has a large amount of complexity to it. So, trusting Google can feel like a big leap of faith into very unknown waters.

Is the change related to the demise of third part cookies?

Maybe. We know the eventual abolishment of third-party cookies will have an impact on the quality and quantity of touchpoints we can see, capture and store. It could be possible that Google is simply sticking with last click and DDA models because they provide the highest confidence and lowest risk measurement models.

So, where does all this leave us?

There is an inevitable conclusion that if you use Google Analytics as your go to measurement tool then you will be moved toward the direction of the data-driven attribution model. The only alternative will be the last-click model. This is fine for some channels or tactics but ultimately won’t give you a full picture of marketing performance.

So regardless of how much we trust it, DDA is the future.

On this page