Performance Max Brand Exclusions and What They Miss

What brand exclusions actually do

A brand exclusion tells a Performance Max campaign not to serve on traffic that Google has classified as belonging to a brand you selected. You pick brands from a brand list — entities Google already recognises, or ones you submit for verification with a name and a website. The exclusion is applied to the campaign, and it covers Search and Shopping inventory inside PMax.

The important word is classified. You are not typing text. You are pointing at an entity in Google's index and saying "not this one". Google then decides which queries belong to that entity. That decision is made by Google's systems, it is not published, and you cannot inspect it line by line. You can see the consequences in your search terms report, and that is all.

This is a different mechanism from a negative keyword, which matches text. A phrase negative "nike" blocks any query containing the token nike, full stop. A Nike brand exclusion blocks whatever Google thinks is Nike traffic — which may include misspellings and product line names you never listed, and may exclude a query you would have expected it to catch. Broader in some places, narrower in others. Never exactly the same shape.

Two campaign types accept them: Performance Max and Search. In Search, brand exclusions exist mainly so broad match keywords stop drifting onto brands you do not want. In PMax they exist because there are no keywords at all, so before brand exclusions arrived the only lever was campaign-level negatives — and for a long time even those needed a support rep.

Brand inclusions are the other half

Alongside exclusions, PMax accepts brand inclusions: serve only on traffic Google classifies as these brands. Same machinery, opposite sign. A brand-only PMax campaign built with inclusions is a legitimate structure — it lets you separate branded and non-branded spend inside PMax rather than fighting the algorithm for control of the split.

The catch is identical. An inclusion narrows to a classification, not to a word list. "Only my brand" still leaves room for whatever Google decides is adjacent to your brand, and that regularly includes long-tail informational queries carrying your name. More on that below.

Why every article on this setting stops too early

Most guides to PMax brand exclusions cover the same four steps: open the campaign, find brand settings, pick your brand from the list, add competitors if you like. That is accurate and it takes four minutes. Then the article ends, as if the brand question were the whole waste question.

It is not. Brand traffic is one slice of what a Performance Max campaign buys. In the accounts we audit, the terms that burn money without ever converting are mostly not branded at all — they are research queries, job searches, DIY instructions, wrong-geography queries, and product words that happen to overlap with something you do not sell. None of those name a brand, so no brand exclusion in the world touches them. Where Performance Max wasted spend hides goes through the categories in detail.

The second thing those guides skip: the reason people reach for brand exclusions is usually a reporting problem, not a targeting problem. PMax eats branded search, the CPA looks fantastic, and the advertiser cannot tell how much of that performance was real incremental demand. Excluding brand fixes the measurement. It does not, by itself, make the non-brand half of the campaign profitable — it just stops the brand half from hiding the loss. If you exclude brand and your PMax CPA triples, nothing broke. You just saw the number for the first time.

Say that out loud before you flip the setting, because someone will ask.

What brand exclusions do not catch

They do not catch anything that is not a brand. That sounds obvious and it is where most of the money goes.

Here is the kind of list we pull out of a PMax search terms report for a mid-sized retailer, with brand exclusions already in place on the advertiser's own name and three competitors:

• how to fix a dishwasher that won't drain — informational, no purchase intent, no brand
• dishwasher repair jobs — job seeker
• dishwasher salary uk — wrong country and wrong intent at once
• free dishwasher council — someone looking for a government scheme
• dishwasher parts diagram — DIY repair, not a new machine
• second hand dishwasher facebook marketplace — a different marketplace entirely

Not one of those contains a brand name. Every one of them can take real clicks at real cost. A brand exclusion list is completely blind to them. Only campaign-level negative keywords stop them, and Performance Max caps you at 10,000 of those per campaign — generous, but it is per campaign, and there are no shared lists to copy across campaigns. Negative keywords in Performance Max and the 10,000 cap covers how quickly that limit becomes real in a multi-campaign account.

Brand-adjacent queries that slip through

Even within brand traffic, exclusions leave residue. Common patterns we see surviving a correctly configured brand exclusion:

• Misspellings far enough from the brand that Google does not tie them to the entity
• Brand plus a third-party word: yourbrand on amazon, yourbrand tesco
• Brand plus a service you do not offer: yourbrand returns policy, yourbrand customer service number
• Your brand inside a longer generic phrase where the generic part dominates
• Sub-brands and product lines that were never registered as separate entities

Support queries are the expensive ones. Someone typing yourbrand contact number is an existing customer with a problem. They click your ad, land on a product page, bounce. You paid for a support call and got nothing. A phrase negative on contact number, customer service, returns policy and track my order handles that in a way no brand setting does — because the query belongs to your brand, which is exactly why the exclusion lets it in.

Competitors are only partly covered

Adding competitors to a brand exclusion list works, within the limits of Google's brand index. Two things break it. Small competitors are often not in the index, and getting them verified means submitting a name and a site and waiting. And competitor traffic arrives in shapes that are not brand queries: alternative to [competitor], [competitor] vs, cheaper than [competitor] — some of which you may want to keep. Whether to block competitor names at all is a real strategic question, not a hygiene task; competitor searches in Google Ads argues both sides.

The measurement trap behind the setting

Brand exclusions are most often sold as a way to stop PMax from stealing credit from your brand Search campaign. That is a real effect and worth fixing. But the fix has a cost that guides rarely name.

Here is the arithmetic in round numbers. A PMax campaign spends $10,000 a month at a $20 CPA, so 500 conversions. Say $4,000 of that spend sits on branded queries at a $10 CPA — 400 conversions — and $6,000 on non-brand at a $60 CPA, 100 conversions. Blended, it looks great. Split, the non-brand half is probably losing money.

Exclude brand. The $4,000 moves to your brand Search campaign, where it likely converts at a similar CPA because the demand already existed. The PMax campaign is now $6,000 at $60. Same business outcome, different reporting. You have learned something true and changed nothing yet.

The change comes next, and it is the harder part: decide whether $60 non-brand CPA is acceptable, and if it is not, attack the search terms driving it. That is negative keyword work, not brand settings work. Our note on how to calculate wasted ad spend honestly explains why the blended figure fools people so reliably.

One genuine side effect: taking a chunk of volume away from a Performance Max campaign can push it back into a learning period. Smart Bidding recalibrates on a narrower, harder set of conversions. Expect a couple of weeks of noise. If the campaign was thin on conversions to begin with — under roughly 30 a month, the level at which we start seeing bidding wobble in accounts we audit — think twice about removing its easiest volume source mid-quarter.

Brand exclusions and negative keywords are not interchangeable

Use both. They solve different problems and they fail in different ways.

The row that matters most is the last one. Everything else is detail.

There is also a match type trap inside negatives that does not exist in brand settings. Negative broad, phrase and exact do not behave like their positive counterparts — a negative does not expand to synonyms or close variants, and negative broad in particular behaves unpredictably across match types. We wrote it up in negative keyword match types don't work like positives. If you are adding brand-adjacent negatives after setting up brand exclusions, use phrase. Exact is too narrow for query text that varies, broad will surprise you.

And if you run Standard Shopping alongside PMax, remember the asymmetry: Standard Shopping does accept shared negative keyword lists, PMax does not. The same block has to be created twice, in two places, with two workflows. Shopping campaign negative keywords work differently covers the rest of the difference.

How to set them up without breaking reporting

Order matters. Do it in this sequence and you will be able to tell what changed.

1. Export the search terms first. Pull campaign_search_term_view for the PMax campaign, last 30 days, before you touch anything. Sort by cost. Tag each term as brand, competitor, or neither. You now have a baseline and a count of how much of the problem brand exclusions can actually solve. In most of the accounts we look at, it is a minority of the wasted spend — usually well under half.

2. Add your own brand. If it is not in Google's list, submit it. Verification takes a few days. Add obvious sub-brands and your .com if it is recognised separately.

3. Decide about competitors separately, later. Do not bundle the two decisions. Your brand is a measurement fix. Competitors are a strategy call, and you want to see the effect of the first change before you make the second.

4. Make sure brand Search can catch what PMax drops. If your brand Search campaign is budget-capped, excluding brand from PMax just loses you the traffic. Raise the cap before, not after.

5. Wait two weeks. Then pull the search terms again. Compare. The terms that survived are your negative keyword list. This is where the actual money is, and it is the step most people skip because the brand setting felt like completion.

6. Add campaign-level negatives for what remains. Phrase match, campaign by campaign, because PMax will not take a shared list. Keep a record of what you added and when — you will need it when someone asks why volume dropped.

A note on the search terms report itself: PMax reporting is thinner than Search. Low-volume queries are aggregated away, and you will not see a full one-to-one mapping between spend and terms. What the search terms report hides explains where the gap comes from. You are working with a sample, not a census. Act on what you can see and accept that some spend stays invisible.

Where search themes fit, and where they do not

Search themes are an input signal, not a targeting control, and they interact with brand exclusions in a way that catches people out. A search theme telling PMax "running shoes for flat feet" is a hint about what to go and find. It does not override an exclusion — exclusions and negatives always win — but it will happily pull in adjacent non-brand junk if the theme is loose.

So a common failure looks like this: the advertiser adds brand exclusions, congratulates themselves, and leaves four vague search themes running. Brand traffic drops. Junk traffic rises, because the asset group is now hunting harder for volume somewhere else. Net waste is flat or worse. We laid out the division of labour in search themes vs negative keywords — themes steer, negatives block, brand exclusions filter one specific class of traffic. Three tools, three jobs, no substitutions.

If you want to know whether this is happening to you, look at what changed in the search terms report after the brand exclusion went live, not at the headline CPA. The CPA will move for reasons that have nothing to do with quality. How often you should check search terms has a rhythm you can keep to.

What PPCleaner does here, and what it does not

PPCleaner does not set brand exclusions. It has no access to Google's brand list, it does not add or remove brands, and it will not decide whether your competitors should be excluded. That is a strategy decision with revenue attached, and it stays with you.

What it does is the step after. Every night it reads campaign_search_term_view across Search, Shopping and Performance Max, finds the terms that took spend and were never going to convert, and adds them as campaign-level negative keywords — including inside PMax, where the 10,000-per-campaign cap and the absence of shared lists make manual upkeep tedious. It never touches budgets, bids, ads or keywords. It never blocks a term that converted in the last 90 days. Every block is reversible in one click for 90 days, so the cost of a wrong call is one click, not a month of lost volume.

The honest limitation: Display, Video and App report no search terms at all. If your PMax campaign is spending heavily on those surfaces, neither brand exclusions nor negative keywords reach that spend, and no tool can show you queries that were never recorded. Placement exclusions are the lever there, and they are a different article.

The other honest limitation: a term-level system can only act on what appears in the report. Aggregated low-volume queries are invisible to us for the same reason they are invisible to you. We work the visible spend, which in most accounts is where the recoverable waste sits anyway. If you want to check how much that is before deciding anything, the hour-long audit checklist will give you a number.