If you run a store with real sales history, you have probably noticed Performance Max reporting a return that feels too good to be true next to your bank balance. Usually it is. The gap almost always traces back to one mechanic the top-ranking guides explain how to toggle but never cost out for you.
This article covers the same setup steps those pages do, then goes where they stop: the per-order math on what branded traffic inside PMax is really doing to your margin.
What brand exclusions actually do
A brand exclusion is an account-level list that tells Performance Max not to serve on queries containing your brand name. Once you add it, Google automatically covers variants and common misspellings using its own brand index, so you do not have to list every version yourself (AdNabu). The control has been available since 2024 and is now a standard account hygiene step (Astraloop).
One limitation matters before you rely on it. Inside Performance Max, brand exclusions apply to Search and Shopping ads only — not Display, YouTube, Discover, or Gmail placements (Google Ads Help). For a retail account that is fine, because the branded leakage you care about is almost entirely on Search and Shopping anyway.
Why Performance Max drifts to your brand
Performance Max is built to chase the cheapest conversions it can find. Branded searches are the cheapest conversions in any account, because the shopper already knows your name and is most of the way to buying.
So the system does exactly what you told it to: it finds those searches and claims the conversions. The reported ROAS climbs, everyone nods at the dashboard, and the branded orders that organic, direct traffic, or a tidy brand Search campaign were already capturing quietly move onto the paid tab.
This is why a PMax campaign can post a spectacular number while your total business orders stay flat. It is not growing the store — it is re-pricing conversions you already had. Practitioner analyses put the apparent-ROAS inflation from absorbed brand traffic at roughly 15% to 30%, with an estimated 8% to 15% of PMax budget leaking to branded queries in unprotected accounts (GrowthSpree; ClickTrends). Treat those as field estimates, not laws.
The profit problem the other guides skip
Here is the calculation none of the setup posts walk through. Say your store spends $4,200 a month through a single Performance Max campaign.
Take the middle of that leakage range — about 11% of budget drifting to branded searches. That is roughly $462 a month flowing to clicks from people typing your store name into Google. Those shoppers were already on their way to you.
Now price the alternative. A dedicated brand Search campaign buys those same clicks cheaply, because nobody outbids you on your own name at scale. The agency smec describes the waste bluntly: Performance Max pays about "$1.50 for a click that you could have bought for $0.20" (smec). At that gap, your $462 buys about 308 branded clicks (462 ÷ 1.50) that a brand campaign would deliver for roughly $62 (308 × 0.20). The other $400 is margin you handed to Google for orders you had already earned.
That is the hidden cost. The visible cost is worse, because it corrupts your decisions. If PMax reports 6.0x but 15% to 30% of that is absorbed brand traffic, the campaign's true pull on net-new orders is closer to 4.6x (6.0 ÷ 1.3). You scale a 6.0x; you would think twice about a 4.6x — and for a thin-margin print-on-demand catalog, that difference decides whether you make money.
Run the margin, not the ROAS
ROAS is not profit, and this is where the brand question gets sharp for print-on-demand. Your break-even ROAS is simply 1 ÷ your contribution margin.
Say you sell a mug for $24. Base cost, shipping share, and processing fees run about $15, so you net $9 — a 37.5% margin (9 ÷ 24). Your break-even ROAS is 1 ÷ 0.375 = 2.67x. Every dollar below that on real acquisition loses money.
Branded clicks convert so well that they can single-handedly drag a campaign's blended ROAS up past break-even while the non-brand work underneath it quietly runs at a loss. Exclude brand, and you finally see the number that decides scaling: what it actually costs to put your products in front of someone who has never heard of you. If you want the full framing, our guide to Google Ads economics for Shopify stores builds the break-even model out end to end.
How to set up brand exclusions
The mechanics take a few minutes. In Google Ads, open Tools and Settings, go to Shared Library, and create a Brand List. Add your store name; Google resolves it against its brand index and folds in variants and misspellings automatically.
Then attach that list under your Performance Max campaign's settings, in the Brand Exclusions section. The brand must exist in Google's database to be selectable — if yours is new or small, you may need to request it be added before the list will hold.
If your brand is not yet indexed, the fallback is campaign-level negative keywords for your brand terms, which Performance Max also supports now that it exposes search terms. It is coarser than a true brand list, but it closes the biggest leaks while you wait.
Brand exclusions alone are not the fix
Turning brand off inside PMax without catching that traffic elsewhere just hands it to competitors who bid on your name. The complete play has three parts.
First, add the account-level brand exclusion so PMax stops eating brand. Second, stand up a dedicated brand Search campaign to capture those cheap, high-intent clicks deliberately and defend them. Third, audit your PMax search terms on a schedule, which you can now do because Google ships full search-term reporting for the campaign type.
The hybrid structure this implies is not a hack — it is the norm. Optmyzr's study of 24,702 Performance Max campaigns found 82% of advertisers run PMax alongside Shopping or Search, and the split-budget accounts posted the strongest returns (Optmyzr). Brand control is one reason the portfolio beats a single catch-all campaign. The same discipline applies to how you structure asset groups and feeds — the Performance Max visual and asset guide goes deeper on that side.
When not to exclude brand
There is a real case for leaving brand in. If you are a tiny account with almost no branded search volume and you are scraping to feed Performance Max the conversions it needs to exit learning, excluding brand can starve the campaign of the data that stabilizes its bidding.
The honest test is volume. If branded searches are a rounding error in your account, the cannibalization is too small to matter and the lost conversion signal costs more than it saves. Once your brand generates meaningful search volume — the point where competitors start noticing you — exclude it and move that traffic to a dedicated campaign. Benchmarks help you judge scale here: retail Search clicks average around $4.14 per LocaliQ's 2026 data (LocaliQ), while ecommerce CPCs broadly sit in the $2 to $4 range (WebFX) — branded clicks run a fraction of either.
Where this leaves your account
Brand exclusions are not a growth lever. They are a measurement fix that happens to save money, and that is exactly why they matter for an operating store.
Once brand is out of Performance Max, the ROAS you see is finally the ROAS you can act on. You can set a target off real break-even math, scale the non-brand work that actually grows the catalog, and stop paying a premium for orders you had already won. If you would rather not police search terms and reconcile per-order profit by hand, Victor — the AI employee inside PodVector AI — operates Google Ads and computes true per-order profit from your live store data, with every write action gated behind your approval. For the handoff from in-house management to a partner, see our take on choosing a Google Shopping ads agency.
FAQs
Do brand exclusions in Performance Max also cover misspellings and variants?
Yes. When you add your brand to a brand list, Google automatically includes common variants and misspellings using its own brand index, so you do not have to enumerate them (AdNabu). You still want to spot-check your search-terms report, because edge-case queries occasionally slip through.
Will excluding brand make my Performance Max ROAS drop?
Almost certainly, and that is the point. The campaign was banking cheap branded conversions that lifted the reported number; remove them and you see the true cost of acquiring new customers. The drop is not a performance loss — it is the inflation leaving the metric, which is exactly what lets you make honest scaling decisions.
Where do the branded searches go after I exclude them?
Ideally into a dedicated brand Search campaign you control, where you capture those high-intent clicks cheaply and defend against competitors bidding on your name. Some will also return to organic and direct traffic that was being cannibalized. If you exclude brand everywhere and build nothing to catch it, a rival can step in and buy your own shoppers.
Does this apply to every Performance Max placement?
No. Inside Performance Max, brand exclusions apply only to Search and Shopping ads, not Display, YouTube, Discover, or Gmail (Google Ads Help). For a retail account that is usually fine, because branded leakage concentrates on Search and Shopping, where the intent lives.
How do I prove brand cannibalization is happening in my account?
Open your Performance Max search-terms report and look for your store name and its variants. Then check whether your brand Search campaign's impressions dropped while PMax spend rose — that pattern is reallocation, not growth. The cleanest test is comparing total business orders against total ad spend over the same window, rather than trusting any single campaign's ROAS.
What is the single most common mistake with brand exclusions?
Treating a high Performance Max ROAS as proof the campaign is working before brand traffic is excluded or segmented. A PMax campaign eating branded search can show a stellar return while adding zero incremental orders. Judge it on non-brand performance and total-business results, not the headline number.