Best practices new vs returning customers

1. Campaign setting: "Bid more for new customers" (NCA)
This is the New Customer Acquisition (NCA) objective in Google Ads and is currently the absolute industry standard for Performance Max and Search campaigns.

• How it works: You tell the algorithm how much extra a new customer is worth (for example: a purchase is worth €50, but a new customer receives a fictional bonus value of €30, total conversion value = €80).

• Advantages: The algorithm (Smart Bidding) keeps all your conversion data in one campaign. This prevents data fragmentation and gives the machine learning maximum fuel to optimize. You can choose between 'Bid higher for new customers' or 'Bid only for new customers'.

• Disadvantages: You are dependent on Google's ability to recognize a new customer. If users clear cookies or purchase via a different device, Google may incorrectly identify them as new.

• Verdict: The winner. This is by far the best and most scalable method, provided you feed the algorithm with the right data (see option 3).

2. Steering via a separate conversion point for new customers
With this method, you create a separate conversion point (e.g. via your own backend or Tag Manager) that only fires if the customer is not yet in your database.

• Advantages: It is 100% accurate based on your own backend data. You have full control.

• Disadvantages: You fragment your conversion data. Smart Bidding performs best with as much conversion volume as possible. By splitting existing and new customers into separate goals or campaigns, you give the algorithm less fuel. Moreover, the technical setup is complex.

• Verdict: Not recommended. In the past this was useful, but with today's AI-driven bidding strategies, splitting conversion volume works against you.

3. Via Audience Signals (Audience Signals & Exclusions)
Audience signals and audience lists play a crucial but supporting role in the modern Google Ads landscape.

• How it works: You upload your own customer database (Customer Match) of existing buyers, or use website audiences (e.g. "Buyers last 540 days").

• Application in PMax: In Performance Max these are signals, meaning you point the algorithm in the right direction, but it is not a hard exclusion.

• Application in Search/Shopping: Here you can add existing customers as an exclusion (negative audience) to your campaign, so your ads are shown literally only to unknown users.

• Verdict: The perfect complement. On its own, it is not enough in PMax to drive purely for new customers (because signals are not hard targeting), but it is essential to make Option 1 work perfectly.

The Ultimate "Best Practice" Setup
If you want to get the maximum out of your budget for new customers, combine option 1 and 3 as follows:

1. Turn on the NCA objective (Option 1) and choose "Bid higher for new customers" in your key campaigns.

2. Determine a realistic value: give the new customer an extra conversion value that makes sense based on your Customer Lifetime Value (CLV). Don't make this unrealistically high, or you'll disrupt your ROAS objective.

3. Upload your customer data (Option 3): upload a dynamic Customer Match list of all your existing customers. Google uses this list in the NCA objective to know with almost 100% certainty who is an existing customer and who is genuinely new.

What does this mean for Metrion?
For Metrion, the advice is also to use the NCA setting for every campaign instead of splitting conversions. Even though it would be no problem for us to split the conversion based on our own user identifier or the user identifier in a CMS, advertising at the cutting edge benefits most from high conversion volumes.