Over the past few weeks, Google advertisers may have received an email informing them they are now able to take advantage of a relatively new type of attribution modelling in Google Ads known as ‘data-driven attribution’.
The communication from Google states:
“Data-driven attribution is a machine-learning model that attributes credit for your conversions based on how people search for your business and interact with your ads.
“Compared with the last-click model, data-driven attribution looks at the entire journey that leads a user to convert and allocates credit to each step in that journey”.
But what exactly does that mean?
Google Ads attribution models
There are lots of different attribution models available, and how suitable each of them is will very much depend on your business and its goals. Until now, the default attribution model has been ‘last click’.
This means that all of the credit of that conversion is attributed to the last ad click and so effectively ignores any credit that previous ad interactions via paid search may have had.
For example, with last click, someone may have:
Searched for ‘men’s Nike trainers’, and clicked on a Search ad within an ad group containing that keyword
Searched for “men’s Nike SB skate shoes”, and clicked on a Shopping ad
Searched for “men’s Nike SB trainers size 9” and was served an RLSA ad, and made the purchase.
In this example, all of the credit goes to the RLSA ad, even though we can see that there were other interactions that influenced this purchase.
Can I use the data-driven attribution model?
The new data-driven model that Google Ads is recommending advertisers now use was previously only available to those who met certain conditions, as per Google’s support page:
Some types of conversion actions need at least 300 conversions and 3,000 ad interactions in supported networks within 30 days to be eligible. Once you’re using data-driven attribution for these types of conversion actions, you won’t be able to continue using this model if your data drops below 2,000 ad interactions in supported networks or below 200 conversions for the conversion action within 30 days. You’ll receive an alert when your data drops below this level. After 30 days of continued low data, your conversion action will be switched to the ‘Last click’ attribution model.
However, we noticed that some of our clients who did not meet this eligibility criteria were also being prompted to make the switch from last click attribution.
We reached out to Google Ads support querying this for a client, and they stated:
“As for the data-driven attribution model – you’re eligible. They recently got rid of the requirements.”
This confirms that any account is now eligible to use this attribution model.
How does the data-driven attribution model work?
In response to our query, Google Ads support further confirmed:
“To explain this attribution model a bit – to identify which touchpoints are the most influential, algorithms examine all of the numerous paths in your client’s account (both converting and non-converting). The amount of ad interactions, the timing and sequence of exposure, and the creative assets utilized in each conversion path are all considered when determining results. The algorithms use a counterfactual method to compare what happened with what may have happened to identify which ad clicks are most important for a conversion. Because the model is dynamic, it responds to changes in the account (e.g. expansion of generic keywords).”
In plain English, this means that:
- The model compares the paths of customers who convert to the paths of customers who don’t
- It then identifies patterns among those ad interactions that lead to conversions
- There may be certain steps along the way that have a higher probability of leading a customer to complete a conversion
- The model then gives more credit to those valuable ad interactions on the customer’s path
Switching to the data-driven attribution model
Advertisers are being notified that if they do not make the switch on or around 28 August 2022, they will be automatically switched to this new model.
When logging into Google Ads and reviewing conversion action settings, advertisers will be shown a prompt giving the option to learn more about the data-driven model, and the opportunity to see how it compares to the existing attribution model they have selected.
What are the benefits of using a data-driven attribution model?
If you are running a lot of campaigns across Search (including Shopping), YouTube, and Display, and you want to know how each touchpoint is influencing conversions through the customer journey, data-driven attribution is likely to be a great choice. This is because it helps to determine which keywords, ads, and campaigns have the greatest impact on your business goals.
For more information on attribution modelling, download this datasheet.