Your Performance Max campaign shows a strong ROAS. The numbers look good on a Tuesday afternoon when you check the dashboard. But a question keeps nagging: are these actually new sales, or is Google charging me for customers who would have bought regardless?
It is a reasonable question, and for a significant number of stores, the honest answer is: a meaningful portion of the budget is going to people who were already going to buy.
This is not a flaw in the way your campaign is set up. It is a structural tendency of how Performance Max decides to spend money, and it requires deliberate action to manage.
How Performance Max Decides Where to Spend
Performance Max is a goal-based campaign type. You give it a conversion goal (purchases), a budget, and a set of assets (product feed, images, headlines). Google’s algorithms then decide where across its network to show ads: Search, Shopping, YouTube, Display, Gmail, Maps, all from one campaign.
The algorithm’s objective is to maximize conversions within the budget and ROAS targets you set. It learns over time which audiences, placements, and times of day produce conversions. And here is the key thing to understand: it is optimizing for conversions, not for incremental conversions.
An incremental conversion is a sale that would not have happened without the ad. A non-incremental conversion is a sale that was going to happen anyway and the ad just appeared before it.
The algorithm cannot distinguish between them. From its perspective, a click that led to a purchase is a successful conversion, whether that person was going to buy in the next five minutes regardless or not.
Branded search terms, retargeting pools, and people close to a repeat purchase are all high-probability converters. The algorithm finds them and spends budget on them because they convert. The ROAS looks great. The actual contribution to new revenue may be much lower.
The Branded Search Problem
When a customer searches for your store name, or your brand name, or a specific product name they have bought from you before, they are already your customer. They are navigating to your store.
Performance Max, without explicit controls in place, will bid on these searches. They convert at high rates. The campaign shows strong conversions. But you did not acquire a new customer. You paid Google to be the middleman between your existing customer and your store.
The cost of this is the bid amount for each of those clicks, which is real money, and the opportunity cost of that budget not being spent on reaching someone new.
How common is this? I have reviewed Performance Max accounts where 40 to 60 percent of Shopping and Search conversions came from branded queries. At that level, the effective cost of new customer acquisition from the campaign is substantially higher than the headline ROAS implies.
The fix is brand exclusions: a setting that prevents Performance Max from bidding on your own brand terms. This requires configuring a brand exclusion list in the campaign settings. It is not turned on by default, and Google’s interface does not make it obvious that this is necessary.
The Retargeting Problem
Performance Max uses your audience signals and Google’s own behavioral data to find people likely to convert. People who have visited your site in the last 30 days, people who have already purchased from you, people who added to cart but did not check out: these segments convert at high rates.
The algorithm allocates budget toward high-probability converters. This means retargeting activity, showing ads to people who already know you, becomes a significant portion of what Performance Max does.
Is retargeting worthless? No. It has real value for cart abandoners and recent visitors who were genuinely considering a purchase. But there is a point of diminishing returns. Showing ads to someone who purchased from you three times in the last year, every time they visit any website, is expensive and provides little incremental lift.
The more a Performance Max campaign leans into retargeting for its conversions, the more the reported ROAS is a retargeting ROAS, not a new customer acquisition ROAS. These are different things with different implications for business growth.
How to Tell If This Is Happening to You
A few signals to check:
If your Performance Max ROAS is significantly higher than what you can find benchmarks suggesting for your product category, branded traffic and retargeting are likely a significant component. Unusually strong ROAS deserves scrutiny, not just celebration.
If your account has audience insights enabled, check the composition of your converters. Look at what percentage of conversions come from returning customers versus new visitors. If returning customers dominate, the campaign is doing more retention than acquisition.
If you can compare your Google Ads conversion volume to your actual new customer acquisition rate from Shopify, a wide gap between reported conversions and new customer count is a signal that existing customers are being counted.
Pull the brand terms from your search terms breakdown if you can access it. (In Performance Max this requires digging into insights or using third-party tools, since Google limits search term transparency for this campaign type.) If you see your brand name or store name appearing as a query, those conversions are branded.
What to Do About It
Set brand exclusions. In your Performance Max campaign settings, configure brand exclusions that prevent the campaign from bidding on your own brand name and variants. This is the highest-impact single action for a store where branded cannibalization is occurring.
Consider a separate branded Search campaign. If brand protection matters (and it does, because competitors can bid on your brand name), run a separate, very small branded Search campaign with a controlled budget. This captures branded intent without allowing Performance Max to route its main budget there.
Adjust audience signals to emphasize new customer acquisition. In the audience signals section of your campaign’s asset groups, weight toward prospecting audiences (interest and in-market segments that represent potential new customers) rather than existing customer lists. This does not eliminate retargeting entirely, but it gives the algorithm a stronger new-customer bias.
Set a new customer acquisition goal if your account has the option. Google has been rolling out the ability to set a goal specifically for new customer acquisition within Performance Max. If your account shows this option, using it tells the algorithm that new customers are more valuable than returning ones, which changes its allocation behavior.
The Broader Question of Incrementality
The underlying issue this post is pointing at is incrementality: whether the sales Google Ads is taking credit for would have happened without the ads.
This is ultimately a hard problem to measure precisely without formal incrementality testing. But you do not need a controlled experiment to understand that a large portion of branded conversions and existing-customer retargeting are probably not incremental. Common sense and the structural logic of how Performance Max operates are sufficient to warrant action.
The stores that get the most real value from Performance Max are the ones that have done the work to exclude branded terms, configure meaningful signals, and monitor the audience composition of conversions on an ongoing basis.
Getting This Fixed
Configuring Performance Max to drive actual new customer growth rather than just harvesting existing intent is one of the most impactful changes I make when auditing and restructuring accounts. It requires specific settings changes and ongoing monitoring.
If you want to understand what your Performance Max campaign is actually buying, reach out at adnanagic.com/#contact.
Part of the Google Ads for Store Owners series, written for ecommerce owners who want their ad spend to actually work.
Related Posts
- Performance Max: What to Actually Do With It
- Your Google Ads ROAS Looks Great but Your Bank Account Disagrees: Why
- Branded vs Non-Branded Campaigns in Google Ads
Related Posts
Why Scaling Your Budget Made Your Google Ads Results Worse, Not Better
9 min read
What Happens to Your Google Ads When You Switch Platforms or Redesign Your Store
10 min read
The Hidden Reasons You Are Wasting up to 30 Percent of Your Shopping Budget
10 min read
Need Help With Your Google Ads?
I help e-commerce brands scale profitably with data-driven PPC strategies.
Get In Touch