If you run an operating store, you have almost certainly seen this: your Performance Max campaign reports a gorgeous return, but total orders are flat. That gap is the single most common way PMax misleads a seller who already knows their numbers. This guide shows you how to shut it down and — more importantly — what the fix is worth in profit terms.
Why Performance Max eats your brand terms in the first place
Performance Max is built to chase the cheapest conversions it can find. Branded searches are the cheapest of all, because the intent is already built — someone typing your store name was going to buy anyway.
So PMax quietly serves against your own brand, "wins" conversions that organic and direct were already capturing, and books them as campaign performance. Your reported ROAS climbs. Your actual new-customer acquisition does not.
This matters more for retailers than most guides admit, because PMax is mostly a Shopping engine under the hood. The agency smec, which manages feeds for hundreds of retailers, reports that 74–97% of PMax cost goes to feed-based (Shopping-style) ads. The brand leakage rides inside that spend where it is hard to see.
What "exclude brand terms" actually does
A brand exclusion is not a negative keyword. Negatives force you to chase every misspelling and variation by hand, which is unwinnable.
A brand exclusion is an account-level list you apply to a campaign. Google's own documentation confirms the exclusion covers the exact brand term, its misspellings, and brand searches in a foreign language — so one list entry handles the long tail automatically.
These controls are not new or experimental. Account-level brand exclusions for PMax have been available since 2024, covering brand variants and misspellings, so there is no reason an operating account should still be running unprotected.
The number that matters: what brand leakage does to your reported ROAS
Most articles stop at "it skews your data." For an operator, that is not enough — you need to know the magnitude, because it changes every scaling decision you make.
Practitioner analyses put the apparent-ROAS inflation from absorbed brand traffic at roughly 15–30%, with about 8–15% of PMax budget leaking to brand queries in unprotected accounts. smec illustrates the raw cost mechanism bluntly: you end up paying "$1.50 for a click that you could have bought for $0.20". Treat both as field estimates, not laws — but they are directionally consistent and worth modeling against your own account.
A worked example
Say your store spends $4,000 a month on Google Ads, almost all of it in one PMax campaign, and it reports a 6.0x ROAS — $24,000 in attributed revenue.
Now assume the middle of that leakage range: 12% of spend goes to branded queries. That is 0.12 × $4,000 = $480 buying conversions that your brand name would have earned for free or near-free.
Strip the brand layer out. Suppose those cheap branded clicks are credited with about $4,800 of the attributed revenue (one-fifth of the total, which is conservative given how well brand converts). Your true non-brand picture is $24,000 − $4,800 = $19,200 in revenue on $4,000 − $480 = $3,520 of real prospecting spend.
That is $19,200 ÷ $3,520 = 5.45x on non-brand — versus the 6.0x the dashboard showed. The reported number was flattering you by roughly 10%, and in a leakier account the gap runs wider.
Here is why that matters for profit, not vanity. Figure a mug you sell at $34 with a $21 blended cost (base, shipping share, and processing fees), leaving a $13 contribution margin, or about 38%. Your break-even ROAS is 1 ÷ 0.38 = 2.63x. A 6.0x and a 4.8x both clear break-even — but one tells you that you have far more room to scale non-brand spend than the other does, and you will bet real budget on the difference.
If the arithmetic behind break-even ROAS is fuzzy, our guide to Google Ads economics for Shopify stores walks the full identity and the margin math underneath it.
How to exclude brand terms from Performance Max
The setup takes a few minutes. The patience comes afterward.
- In Google Ads, open Tools → Shared Library → Brand lists and create a list containing your store's brand (and any sub-brands or product-line names you own).
- Open each Performance Max campaign, go to Settings → Additional settings → Brand exclusions, and apply the list.
- If your exact brand is not already in Google's catalog, request that it be added — indexing a new brand can take several weeks, so do this first.
- Repeat across every PMax campaign. Exclusions are applied per campaign, not once per account.
While you are in there, pair the exclusion with a dedicated brand Search campaign. Excluding brand from PMax without a brand campaign to catch those searches hands the click to a competitor bidding on your name.
After you exclude: what to watch
The useful signal is not inside PMax — it is the total-business view. Watch your overall orders against total ad spend for the next two to four weeks; if PMax's ROAS falls but total orders hold steady, the fix worked exactly as designed.
You can now verify leakage directly, because full search-terms reporting for PMax arrived in 2025. Pull the PMax search-terms report and confirm branded queries have disappeared from it.
Expect your PMax reported ROAS to drop. That drop is the point — you are trading an inflated number for an honest one. The classic companion move is to also pull non-converting SKUs out of PMax so the campaign's budget concentrates on products that actually earn their impressions.
The honest limitation: exclusion is a measurement fix, not a growth lever
Excluding brand terms does not create new demand. It reveals the truth about the demand you already pay for, so you stop over-crediting PMax and over-starving your brand Search campaign.
That is exactly why it belongs in a portfolio approach rather than a set-and-forget one. Optmyzr's study of 24,702 PMax campaigns found that 82% of advertisers run PMax alongside Shopping or Search, and split-budget accounts posted the strongest returns in that sample — the hybrid is the norm now, not a hack. Our Performance Max case study walks a real restructure of this kind end to end.
The deeper problem is that none of this shows up as profit until you subtract product cost, fulfillment, and fees from every order. ROAS — even an honest, brand-excluded ROAS — still ignores all of that.
That is the job Victor, the AI employee inside PodVector AI, was built for. Victor runs your Google Ads as a full operator and connects to Shopify, Meta Ads, Printify, Printful, Gelato, and Klaviyo, so it computes true per-order profit across the whole account — and because every write action is approval-gated, Victor can flag branded leakage and queue the brand-exclusion change for you to approve before anything executes. Put Victor on your store and see what your campaigns actually net.
For sellers who would rather hand the whole feed-and-bid problem to specialists, our breakdown of what a Google Shopping ads agency does covers that path and where it pays off.
FAQs
Does excluding brand terms from Performance Max hurt my sales?
It should not reduce real sales — only reallocate where the credit lands. The branded searches PMax was absorbing still happen; they just get caught by organic, direct, or your brand Search campaign instead. If total orders drop after you exclude, that is a signal your brand Search coverage has a gap, not that the exclusion was a mistake.
How long does a brand exclusion take to work?
Applying an existing brand list takes effect quickly. The wait is when Google has to index a brand it does not yet recognize — that can run several weeks, which is why you request new brands before you need them.
Will my Performance Max ROAS go down after I exclude brand terms?
Almost always, yes — and that is the correct outcome. You are removing the cheapest, easiest conversions from the campaign's reported results, so the number that remains reflects genuine prospecting performance. Practitioners estimate branded absorption inflates apparent ROAS by roughly 15–30%, so a drop in that range is expected, not alarming.
Should I exclude competitor brand terms too?
You can, and the control supports it, but the economics differ. Bidding on competitor terms is a deliberate conquesting strategy with much lower conversion rates, so exclude them only if you have reviewed the search-terms report and decided that traffic is not worth its cost for your margins.
Can I just use negative keywords instead of brand exclusions?
For PMax, no — campaign-level negatives did not historically exist and still do not scale to every misspelling and language variant. The account-level brand exclusion list is the purpose-built tool, and Google confirms it covers variants and foreign-language searches automatically. Use the exclusion list; use negatives for the surgical cases it misses.
How do I know how much brand traffic PMax was really taking?
Pull the PMax search-terms report before and after you apply the exclusion and compare branded-query impressions and conversions. For the fuller picture, reconcile total business orders against total ad spend over the same window — if orders held steady while PMax's reported conversions fell, that delta was the brand traffic PMax had been quietly claiming.