Data-Driven Attribution Is The ‘New Thing’ For Google

If you manage Google Ads, you’ve probably received some emails and/or seen in your account that Google is starting to aggressively push data-driven attribution.

So I thought I’d do a quick article on what that means and how it compares with its less-smart cousin, position-based attribution. 

This is not an explanation of attribution models, that’s another post.

Position-Based Attribution In Google Ads

So briefly, understanding position-based attribution in Google Ads is important in realizing the direction of Data-Driven Attribution and how it tracks conversions.

PBA (position-based attribution) was one of Google’s early attempts to acknowledge that a user often interacts with your ads multiple times before converting.

Weighing each time the User interacts with your keywords and ads is important for understanding the full value of your campaigns.

In PBA, each ad click get’s partial credit for a conversion. This is how Google distributes credit for a conversion, from the mouths of the tech giant itself:

“Position-based” gives 40% of credit to both the first and last click, and distributes 20% among other clicks.

With emphasis on first and last click, Google helps share the love to all action-based factors that led to that conversion. Those who use PBA will note that many of their conversions read with a decimal point (i.e. 8.40 conversions).

But PBA falls short in the same way that all of the old Google Ads attributions fall short, they are arbitrary assignments with not concept of account level trends.

Data-Driven Attribution – In Google Analytics And Google Ads

The solutions to those aforementioned shortcomings? Automation! 

Here’s how Google talks about Data-Driven Attribution in Google Ads:

When possible, “Data-driven” uses your account’s conversion data to calculate the actual contribution of each ad interaction across the conversion path.

If you read further in Google’s documentation, you’ll come to understand that Data-Driven Attribution (DDA) looks for patterns in your account and if it starts noticing certain combinations of clicks lead to conversions, it will shift the assigned conversion value toward those conversions.

When Should You Turn On Data-Driven Attribution?

Google will let you know that your account is eligible for Data-Driven Attribution once it has enough conversion and click data within its designated time frame. Here are Google’s stated conditions:

Most conversion actions are eligible for data-driven attribution, regardless of conversion or interaction volume. Only eligible conversion actions can be upgraded to data-driven attribution.

Some types of conversion actions need at least 300 conversions and 3,000 ad interactions in supported networks within 30 days to be eligible.

That’s not that helpful, Google, but let’s look into this based on what we know about automated bid strategies in Google. 

With automated strategies, Google performs better when you’re either in a search market with tons of data due to high search volume and competition (like a calendar software or CRM or something like that), or if you have a ton of account data coming in each month (namely conversions).

Data-Driven Attribution will pair best with any conversion-based automated strategy you’re running, bringing Google Ads Automation full circle.

Something to note is that Google states that if you start losing the required volume for Data-Driven Attribution, that it will revert to a “Last Click” until volume is back up.

That being said, for my B2B clients, I still use Position-Based Attribution until I have 2-3 months of primary conversion data (30-40 primary conv. A month). This ensures that I have the required data to reduce wasted spend.

Something that I have toyed with was creating a duplicate conversion where I assign Data-Driven to one and keep Position-Based for the other and split a campaign, testing just the two conversions. I’ll let you know how that goes.

Conclusion

Even as I write this, I realize more and more how impactful this will be in terms of Google having a full picture of your customer.

In terms of implementation I still take new changes from Google slowly and ensure that I have ample data before I fully implement. 

Think about it, with looser match types, Google is deciding what terms you show for more than ever (as evident with broad terms performing better than ever).

Then Google’s Automation determines the ad copy it shows based on that individual search term. It’ll then decide what bid it assigns based on conversion data at the campaign, ad group, keyword, search term, audience, demographic, schedule, & location level (even more than that but you get the point).

And now, to wrap up that loop, Data-driven Attribution looks back at that process and determines which combination of these things over multiple clicks were most valuable in bringing in a conversion, and then it sends that data into the conversion column where it repeats the automation indefinitely!