If you have ever built a long negative keyword list to keep a competitor's brand out of your campaigns, only to watch a misspelling or a reworded query slip through and buy a click, you have hit the structural limit of negative keywords: they match text strings, not brands. You block acme crm and the query acme c.r.m. or akme crm walks straight past it. Google's brand inclusions and brand exclusions were built for exactly this problem, and in an account where matching keeps getting looser they are increasingly the right instrument for brand-level control — not a longer negatives list.
This post is about when to reach for brand inclusions and exclusions instead of hand-built brand negative keywords, and where each one belongs. The short version: brand controls operate on Google's brand entity index rather than on the literal strings you type, so they catch the misspellings and variations a negative keyword list leaks, and they hold up under AI Max's intent matching in a way text-string negatives do not. Negatives still own the non-brand work. Getting the division right is the difference between a lean, durable setup and a list you are forever patching.
What brand inclusions and exclusions actually are
Brand exclusions prevent your ads from serving on queries associated with brands you name. Search Engine Land's rundown of the feature is blunt about the scope: brand exclusions stop ads on "queries & misspellings for certain brand names you don’t want to match to – across all match types and Dynamic Search Ads". That phrase — and misspellings, across all match types — is the whole point. Because Google maps the excluded brand to its query knowledge, one exclusion covers the variations you would otherwise have to enumerate one string at a time.
Brand inclusions are the mirror image. Applied to a broad match Search campaign, they make your ads match, in Google's own words, only to queries related to your brand name in that campaign. Instead of blocking a brand, an inclusion fences a deliberately broad, AI-driven campaign so it stays on your own brand rather than wandering into unrelated queries. Inclusions require a Smart Bidding strategy because broad match is designed to run with automated bidding; you cannot bolt an inclusion onto a manual-CPC broad campaign and expect it to behave. The two controls answer opposite questions — keep this brand out versus stay on this brand — but both work at the level of the brand entity, not the string.
Why brand negative keywords leak, and brand controls don't
A negative keyword blocks a text string. That is its strength for surgical, one-off exclusions and its weakness for brands. As covered in the negative keyword match types guide, negatives do not expand to plurals, synonyms or misspellings the way positive keywords do, so excluding competitorbrand leaves competitor brand, competitorbrand reviews, and every misspelling of it still eligible. Google's own documentation confirms the gap: "Negative keywords don’t match to close variants", which means the brand-negative list you thought was airtight has as many holes as there are ways to misspell the brand.
This gap widens under AI Max and intent-based matching. A text-string negative can only block queries that contain the string; AI Max infers the meaning behind a query and can serve on a paraphrase that never contains it. PPC Land, describing the newer brand controls inside AI Max, notes that traditional negatives targeted specific text strings but that AI Max’s intent inference could route around them, and that the entity-based controls instead work at "a higher level of abstraction" using Google's brand index rather than manual lists. That is precisely why a brand exclusion holds where a brand-negative list springs leaks: it is matching the brand, not chasing the strings people use to search for it. This is the brand-specific case of the broader paraphrase problem with negatives.
When to use a brand exclusion instead of a negative list
Reach for a brand exclusion whenever the thing you want to block is a brand as a concept rather than a single phrase. Competitor brands are the obvious case: if you do not want to pay for clicks on a rival and its misspellings, one brand exclusion is more complete and lower-maintenance than a negative list you keep extending every time a new variation shows up in the search terms report. It is the direct answer to the "search term whack-a-mole" that a competitor-brand negatives list turns into, and it frees you from re-auditing that brand's spellings week after week. Where you are weighing whether to block a competitor at all, the negate-or-bid decision on competitor terms still applies — the exclusion is simply the cleaner mechanism once you have decided to block.
The case is even stronger in Performance Max, where you have less search term visibility and a lower negative keyword ceiling to work with, which is the argument laid out in the Performance Max brand exclusions versus negative keywords comparison. Brand exclusions were available in PMax before they came to Search for exactly this reason: when you cannot see every query and cannot list unlimited negatives, matching on the brand entity is the only control that scales. The migration path is the same in both campaign types — set the exclusion, verify in the search terms report that the brand's queries and misspellings have gone quiet, and only then retire the now-redundant string negatives so you are not maintaining two systems that do the same job.
Brand inclusions: keeping broad match on-brand
Brand inclusions solve the opposite failure mode: a broad match campaign that spends your brand budget on queries that have nothing to do with your brand. Broad match plus Smart Bidding is powerful and genuinely wide, and left unfenced it will chase intent far from where you meant it to. An inclusion constrains that campaign to serve only on brand-related queries, so you get broad match's ability to catch the long tail of ways people phrase a brand search without also inheriting its tendency to drift into unrelated territory. If you run a dedicated brand campaign on broad match, an inclusion is what keeps it a brand campaign.
The important caveat is that an inclusion is a scope control, not a substitute for the rest of your hygiene. It ensures the campaign stays on-brand; it does not decide which non-brand junk to block in your other campaigns, and it only functions with Smart Bidding attached. Think of inclusions as the way you tame broad match on a brand campaign specifically — a sharper, entity-level version of the general problem of running broad match without a paired negative list. On non-brand campaigns you are still doing ordinary search-term review and negation; inclusions are a targeted tool for the brand-campaign case, not an account-wide setting.
The AI Max branded-search toggle
As accounts move to AI Max, brand controls have been folded into a single branded-search setting. PPC Land describes the toggle as offering three choices: show ads on all relevant searches, control branded searches with brand inclusions and exclusions, or show ads only on unbranded searches. That middle option is the surgical one — it preserves the ability to allow some branded terms while excluding others, for instance permitting your own brand while blocking competitors. The "unbranded only" option is the blunt instrument for a pure prospecting campaign that should never touch brand queries at all.
For a practitioner, the shift is that brand handling under AI Max is a settings decision made once at a higher level, rather than a running battle in the negative keyword list. That does not make your search terms report obsolete — you still read it to confirm the toggle is behaving and to catch the non-brand waste that no brand control addresses, the same discipline as the visibility-ratio audit. But it does mean that if you are still trying to control brand traffic purely through negatives under AI Max, you are fighting the platform's matching with the wrong tool. Move brand control to the branded-search setting and keep your negatives for what they are good at.
The division of labour: what stays a negative keyword
Brand controls replace the brand portion of your exclusion work, not all of it. Negative keywords remain the right tool for everything that is not a brand entity: category terms you do not sell, research-intent modifiers like free, jobs, salary and how to, and the specific junk strings your search terms report surfaces that do not map to any brand. Those are text-string problems and text-string negatives handle them well, especially when grouped into durable patterns rather than one-off blocks, as covered in the pattern-grouping guide. The mistake is using negatives for brands, where they leak, and neglecting them for the non-brand waste, where they are irreplaceable.
So the practical division is straightforward. Brands you want to block — competitors, partners, adjacent names — go to brand exclusions, which catch the misspellings and variations for you. A brand campaign you want to keep on-brand gets a brand inclusion with Smart Bidding. Everything else — the non-brand irrelevant queries that are the bulk of most wasted spend — stays in a well-structured negative keyword list, organised the way the account, campaign and shared list structure guide lays out. Set the brand controls, verify with the search terms report, retire the redundant brand negatives, and you end up with a leaner setup that does not leak brand traffic and does not forget to block the non-brand junk.
Related reading
For the Performance Max angle on the same trade-off, see brand exclusions versus negative keywords in PMax. For why text-string negatives miss variations in the first place, see the negative keyword match types guide, and for deciding whether to block a competitor at all before you set the exclusion, see negate or bid on competitor search terms.