Google Ads Brand Lists 2026: Exclusions Guide

Sooner or later every automated account faces the same question: how much are you paying for traffic that would have arrived for free? Performance Max is happy to take brand queries — they are cheap, they convert, and they make the report look excellent. Broad match in Search pulls in the same direction. The summary view looks great, while in reality half the “result” is people who already knew your name and would have clicked the organic listing anyway.

Google Ads brand lists are the mechanism for managing that boundary explicitly instead of maintaining endless negative keyword lists. Below: what brand exclusions, brand inclusions and brand restrictions actually are, where each of them works, how to apply them, and how to measure the effect so you do not cut your own volume in the name of tidiness.

A brand list does not “turn off brand”. It gives you a choice about which campaign receives a brand query — and that is the only way to see the true cost of acquiring a new customer.

What a brand list is and why it beats negatives

A brand list is an account-level entity: a set of brands (your own, competitors, partners) that is then applied to individual campaigns. The key difference from a negative keyword list is that it operates on the brand as an entity rather than on a string. The system matches spelling variants, misspellings and other-language versions on its own — you do not have to enumerate twenty transliterations by hand.

Criterion	Brand list	Negative keyword list
Unit of work	The brand as an entity	A specific string and its match type
Misspellings and variants	Handled by the system	Added manually
Other languages	Matched automatically with the newest version of broad match	Require separate entries
Where it applies	Search and Performance Max, depending on the setting	Search, Shopping, partly other types
Excluding one specific query	Not its job	Yes, the primary tool

That last row matters: brand lists do not replace negatives, they complement them. Surgical removal of junk queries stays manual work — the method is in our article on negative keywords and search term analysis.

What actually counts as a brand

Not every word qualifies. Google expects a brand to have at least one of the following: a logo, a trademark, a domain name (including one that redirects), a product name that dominates the packaging or the site, or a business name — the last being particularly relevant for smaller companies. Generic category words will not be accepted as brands, and that restriction is deliberate: otherwise the feature would become a way to exclude entire topics.

Three tools that are easy to confuse

Tool	What it does	Where it is available	Typical scenario
Brand exclusions	Stops the campaign serving on queries containing the listed brands	Search and Performance Max	Remove brand from PMax so it stops harvesting cheap conversions
Brand inclusions	Restricts the campaign to queries containing the listed brands	Search campaigns	Build a clean brand campaign without a mile of exact keywords
Brand restrictions	Narrows serving by brand relevance inside a product context	Campaigns with a product feed	Show only your own brand’s products on brand queries

The practical distinction is simple. Exclusions answer “where brand must not go”, inclusions answer “where only brand may go”, restrictions answer “which products belong in a brand context”. The first two together deliver the whole point of the exercise: splitting brand and non-brand traffic into separate campaigns with separate goals and separate economics.

Why separating brand and non-brand matters

Mixing the two inside one campaign creates three concrete problems, each of which costs money.

One: distorted acquisition cost. Brand conversions are cheap — the person already knows you and converts at a high rate. Inside a blended report they pull the average CPA down, and you scale a campaign believing it is efficient while its non-brand half loses money.

Two: distorted bid learning. The automated strategy sees that queries containing your name convert brilliantly and pushes bids there. It optimises toward the path of least resistance, because you are paying it to. The mechanics are covered in our breakdown of the Smart Bidding learning period.

Three: incrementality becomes unmeasurable. While brand and non-brand share one pot, you cannot answer the central question — how many sales you would have got without the ads. The correct way to answer it is a geo experiment, described in our piece on incrementality and geo experiments.

The separation blueprint

Create a brand list with your own brand. Add the core name plus any variants that share no words with it — sub-brands or abbreviations the system will not infer.
Apply it as an exclusion to Performance Max. From that point PMax stops harvesting brand search and works on acquisition.
Build a dedicated brand Search campaign. The construction logic is in our article on brand defence in search: modest budget, high impression share, simple ads.
Give it brand inclusions. That removes the need to maintain a huge list of exact keywords and misspellings.
Check non-brand campaigns for leakage. Brand queries should not appear in their search terms report; if they do, apply the exclusion there too.
Split goals and budgets. The brand campaign should carry a materially lower target CPA than the non-brand one. A blended target serves neither.
Record the change date. Every subsequent period comparison has to respect that boundary.

What happens to the numbers afterwards

The first thing you will see is Performance Max “getting worse”. That is expected and correct: the campaign just lost its cheapest slice of conversions. What matters is that total account output usually does not fall — it is simply distributed honestly now.

Metric	What happens	How to read it
PMax CPA	Rises	This is your real cost of acquiring a new customer
PMax conversion volume	Falls	The departures were conversions you already owned
Blended account CPA	Barely moves	Conversions relocated, they did not disappear
Brand impression share	Becomes controllable	You can now hold it at a deliberate level
Non-brand volume	May increase	PMax budget stopped draining into easy queries

To distinguish redistribution from genuine loss, do not compare month over month — use an experiment, or a geo split where structure makes experiments impractical. Otherwise the first seasonal wave will convince you the split “did not work”.

Competitor brands: a different conversation

Brand lists work on other companies’ names too, but the logic changes. Advertising on competitor brands is legitimate with caveats: it produces expensive traffic with low conversion rates and almost always triggers a retaliatory campaign, after which both sides pay more.

Excluding competitors from general campaigns is almost always right: there, their queries just pollute the statistics and consume budget.
A dedicated competitor campaign is a deliberate decision with its own budget, its own ads and a knowingly higher CPA. Judging it against blended targets is meaningless.
Using someone else’s trademark in ad text is a policy question, not a targeting one; brand lists control serving only.

To see who actually overlaps with you in the auction, use auction insights and impression share.

Where brand lists will not help

Being honest about a tool’s boundary matters more than listing its features. Brand lists do not solve the following:

Removing brand from every campaign completely. Some formats and surfaces are governed differently; always verify actual search terms after configuring.
Separating brand traffic where there are no search queries. In display and video formats the concept of a brand query does not exist.
Catching heavily mangled versions of your name. Strongly distorted variants may not be matched — pick those up with negatives.
Replacing search term work. Non-brand junk is still junk and still requires manual review.
Managing brand traffic on search partners separately from core Search. Network logic is different — it is covered in our article on the Google Search Partner Network.

Six mistakes with brand lists

Excluding brand everywhere at once. If no brand campaign exists yet, you are simply handing brand traffic to the competitors who bid on it.

Forgetting sub-brands and abbreviations. Name variants that share no words with the core brand must be added explicitly — the system will not infer them.

Judging after three days. Changing traffic composition is a medium-impact edit: strategies need time to relearn and conversions need time to back-fill.

Not verifying actual queries afterwards. The only proof an exclusion works is the search terms report, not the fact that the setting saved.

Keeping the same target CPA for brand and non-brand. After the split the targets must diverge, otherwise you strangle acquisition and overpay for brand.

Comparing before and after by calendar month. Different seasonality, different working days, different promotions. Only a controlled test answers the question.

A 30-minute checklist

Open the search terms report for 30 days and calculate the share of spend on your own brand.
Calculate that same share inside Performance Max separately.
Create a brand list with your brand and every variant that shares no words with it.
Check whether a dedicated brand Search campaign exists; if not, build it before applying exclusions.
Apply the exclusion to PMax and to non-brand Search campaigns.
Separate the target CPAs of brand and non-brand campaigns.
After 14 days re-check search terms — there should be no leakage.
Log the change date in the team change log.

How to measure your brand spend share

Before excluding anything, measure the scale. The procedure is simple and takes about fifteen minutes.

Open the search terms report for the last 30 days across all Search campaigns.
Filter to queries containing your name and its main variants.
Sum cost and conversions for that subset and for the full set separately.
Repeat for Performance Max — PMax offers a lower-granularity search terms view, but it shows the order of magnitude.
Compare brand’s share of spend with brand’s share of conversions. If brand delivers, say, 15% of cost and 45% of conversions, you are paying for the easy half of demand and fooling yourself with a blended CPA.

Brand share of spend	What it usually means	First action
Under 10%	Under control	Review quarterly
10–25%	Normal for a recognised brand, but campaigns should be split	Move brand into its own campaign
25–40%	Automation is visibly skewed toward easy conversions	Apply exclusions in PMax and non-brand campaigns
Above 40%	Reporting barely reflects acquisition at all	Full split and a rebuild of targets

Treat those bands as orientation, not policy: for a business with a strong brand and high repeat purchase, 30% may be a deliberate choice, while for an unknown startup 15% already suggests something is misconfigured.

Three application scenarios

E-commerce with a strong brand

The classic case: Performance Max with a product feed absorbs both branded and non-branded demand. The fix is to exclude brand from PMax and keep a separate brand Search campaign with high impression share and a moderate budget. It is also worth checking how traffic distributes across product types, and whether budget is flowing to bestsellers that sell themselves.

Lead generation with no brand recognition

If there is effectively no brand, brand lists add little — there is nothing to exclude. The inverse task is useful though: excluding competitor brands from general campaigns so automation stops spending on someone else’s low-converting audience. The real work here sits elsewhere — in structure and keyword match types.

Multi-brand groups

The hardest case: several owned brands competing inside one account. Here brand lists become routing infrastructure — each brand gets its own campaign and an exclusion in all the others, so campaigns stop bidding against each other. The necessary complement is an explicit agreement about who owns generic non-brand demand, otherwise nobody funds it.

Turning brand lists into a routine

A one-off setup degrades: new sub-brands, new competitors, new products appear. The minimum process looks like this:

Monthly: check non-brand campaigns’ search terms for brand leakage; recalculate the brand share of spend.
Quarterly: review the contents of every brand list — new products, new competitors, retired names.
On every new campaign launch: applying the correct brand list is a checklist step at launch, not a month later.
Ownership: each list needs one named owner, otherwise three people edit it and nobody knows its current state.

This fits naturally into the broader Google Ads account audit checklist, alongside adjacent checks: account-level negatives, placement exclusions and conversion goal correctness.

Where PPC Rebels fits

Splitting brand from non-brand is data-hygiene work, and it only pays off on an account that runs consistently. If you need a working account with history, we offer Google Ads agency accounts, and if you want to connect structure, bidding and measurement into one system, there is Google Ads training. Related pieces from this series: primary and secondary conversions on what the algorithm learns, and Google Ads automated rules on noticing early when something goes wrong.

FAQ

How is a brand list different from a negative keyword list?

A negative keyword is a string with a match type. A brand list is a brand entity to which the system attaches misspellings, variants and other languages. They solve different problems and are used together.

Can I fully exclude brand from Performance Max?

Brand exclusions are available for PMax and remove the bulk of brand queries. There is no absolute guarantee — always verify actual search terms after setup.

Do brand inclusions work in Performance Max?

No. Brand inclusions are a Search campaign feature. For PMax, exclusions are the available control.

How many brands can go into one list?

You will rarely reach a practical limit. What matters more is keeping your own brand and competitor brands in separate lists so you can apply them to different campaigns.

Do I need to add misspellings and foreign-language spellings?

Usually not — the system matches variants itself with the newest version of broad match. Add manually only those variants that share no common words with the main name.

Will conversions drop after excluding brand from PMax?

In that campaign, yes. Across the account, usually not: the conversions move to the brand campaign. Judge the total, not one campaign.

Should I pay for my own brand at all?

It depends on whether competitors bid on it and how visible your organic listing is. The answer comes from a holdout test in selected regions, not from a general rule.

How do I know brand is eating too much budget?

Calculate the share of spend going to brand queries. If it is materially above your share of brand demand and climbing month over month, automation is optimising into easy conversions.

Do brand lists affect Quality Score?

Not directly. Indirectly yes: once traffic is split, ads and landing pages can be written tightly to intent, and that does affect relevance.

Can one brand list be applied to several campaigns?

Yes — that is the point of an account-level entity: one list, many applications, changes propagate everywhere at once.

What about partner and reseller brands?

They usually go into a separate list and get handled case by case: sometimes serving is appropriate, sometimes it conflicts with a partner agreement. There is no universal rule, only an agreement.

How long before I evaluate the result?

Not sooner than two weeks, ideally a month: the strategy needs to relearn and conversions need to back-fill. And evaluate with a test, not with a calendar comparison.

Google Ads Brand Lists: Controlling Brand Traffic in Broad Match and Performance Max