PMax & new customer conversion value
Google Pmax was very successful bringing us new customers. However about a month ago it started spending about 20% of our budget on YouTube. This caused a noticeable decline in new customer acquisition.
Our thinking is that this is because the ROAS target is higher than the algo feels comfortabel with. So it started showing more retargeting ads to existing customers. A quick fix would be to lower the ROAS target, but that’s not an option to maintain profitability. Our first time customers are already largely subsidised.
So what we came up with is to entice the black box to spend more on new customers by lowering the risk, by adding 15% of the average FTB value to the conversion value for new customers.
Has anyone tried something like this before?
The short answer is:
Yes, what you’ve come up with – adding a bonus value to the conversion value for first-time buyers – is a common and effective strategy, especially when running Performance Max (Pmax) campaigns with a strict Return on Ad Spend (ROAS) target, because it directly signals to the algorithm that a new customer is more valuable than its current calculation suggests, effectively lowering the perceived ROAS target for the valuable new customer conversions without changing the stated ROAS goal.
The best way to implement this reliably and cheaply is by using the Google Ads API in conjunction with Google Tag Manager (GTM) and a server-side solution like Stape or Google Cloud Platform to calculate and send the adjusted conversion value only for new customer purchases, overriding the standard conversion value sent with the purchase event.
The long answer is:
Your analysis is likely correct: when Pmax struggles to meet a high ROAS target with cold traffic (new customers), it leans into retargeting, which often includes cheaper, lower-funnel inventory like YouTube remarketing, to find easier conversions, even if those are from existing customers.
Modifying the conversion value is the right solution.
You want to increase the reported value for new customers only.
Since you can’t lower your ROAS target for profitability reasons, this method manipulates the input value for the algorithm.
By adding 15% of the average first-time buyer (FTB) value to the conversion value for new customers, you are telling the Pmax black box that these conversions are worth more to you, making it more willing to spend on acquiring them while keeping your overall profitability goals intact.
This strategy is excellent because it’s a surgical adjustment.
To implement this reliably and maintain data integrity, using the Google Ads API, GTM, and a server-side tagging solution like Stape or Google Cloud Platform is an excellent and cost-effective approach.
Here’s why: your browser-side GTM setup can easily identify a new customer (e.g., based on a first-party cookie or a lack of an existing customer ID).
When GTM detects a new customer making a purchase, instead of sending the data directly to Google Ads, it sends the purchase event to your server-side container (Stape, in this example).
In the Stape environment, you have a greater ability to manipulate the data before it’s forwarded to Google Ads.
A simple transformation can be set up to take the original conversion value, check if the customer is new, and if so, add your 15% bonus.
The server then sends this adjusted conversion value to Google Ads via the Google Ads API.
Using the API is crucial because it allows you to send conversion data directly to Google Ads without the data being filtered or blocked by browser restrictions, and it also allows you to update or override conversion values with a much higher degree of certainty and control compared to only using a client-side tag.
Stape and Google Cloud Platform are cheap solutions for this because they offer pay-as-you-go or free tiers (in Stape’s case for smaller volumes) that are far less expensive than other enterprise solutions.
This setup also centralizes your data collection, making it more resilient to future privacy changes and ensuring that the manipulated conversion value is only used for bidding in Google Ads, while your internal reporting systems can still rely on the original, unadjusted conversion value.
You are essentially creating a cleaner, more reliable pipeline to send the purchase Standard Event with an algorithmically preferred value for a new customer.
About The Author
