Brand Suitability in Google Ads 2026: Maximum, Moderate, Limited and the 1 October Change

“What was my ad running next to?” stopped being a question only global brands ask. In 2026 Google reworked its brand suitability controls: the familiar inventory tiers were renamed to Maximum, Moderate and Limited, and from 1 October legacy digital content label exclusions start disappearing from parts of the stack. At the same time, independent verification of where impressions actually landed has become a more serious line item. This piece covers what genuinely changed, which tier fits which job, what over-filtering costs you in CPM and reach, and where Google’s own settings stop being enough.

This is not a guide to blocking individual channels by hand — that is its own discipline, covered in the breakdown of placement exclusions and brand safety. Here we work one level up: how the inventory filter itself is built and how to use it deliberately.

Brand safety and brand suitability are different problems

The terms get used interchangeably, and it costs advertisers real reach.

• Brand safety is protection from content nobody should be adjacent to: violence, hate speech, adult material, harmful misinformation. That is a baseline the platform enforces regardless of your settings.
• Brand suitability is fit with your brand specifically. What is fine for an energy drink may be wrong for infant formula. That judgement is always yours, and there is no universally correct setting.

The practical consequence: cranking every filter to maximum strictness “to be safe” does not buy you safety — you already have the baseline. It buys you a smaller addressable pool, and you pay for it in media cost.

The three inventory modes: what changed beyond the names

The tiers used to be called Expanded, Standard and Limited. In 2026 Google moved to Maximum / Moderate / Limited. It is more than a rename: the descriptions now define each tier by what it excludes rather than by an abstract notion of “standard”.

Mode	What is eligible	Best fit
Maximum	All inventory that meets the platform’s baseline monetisation rules. Widest reach and generally the best cost per contact	Performance campaigns, direct response, testing new verticals
Moderate	Excludes strong profanity, dramatised violence and sexually suggestive themes	Mass-market brands, everyday consumer goods, anything reaching broad family audiences
Limited	Additionally excludes moderate profanity — including in mainstream music videos	Children’s categories, pharma, conservative B2B, public sector

The detail people underestimate about Limited: it filters out a substantial share of popular music content. For campaigns aimed at younger audiences that is not “slightly safer placements” — it removes one of YouTube’s most-watched segments wholesale.

Performance Max defaults to Maximum. If your brand genuinely requires stricter inventory, you have to set it explicitly, or the campaign will serve across the widest available pool.

What changes on 1 October 2026

Google is consolidating its control surface. From 1 October 2026, digital content label exclusions and most sensitive category exclusions are removed from the Display & Video 360 API and Structured Data Files. Advertisers are steered toward inventory modes and content themes instead.

What that means in practice:

1. Old settings do not map one-to-one. Content themes are not a direct replacement for the previous categories — the correspondence is approximate. Assume an automatic migration lands somewhere other than where you were.
2. If your settings are applied programmatically, audit that code. Scripts and integrations writing to the deprecated fields will quietly become no-ops after the date. Same class of problem as version sunsets in the Google Ads API.
3. Document the current state now. Snapshot the settings per campaign before the date, or you will have no way to explain a reach shift afterwards.

Which controls exist where

Campaign type	Inventory modes	Placement exclusions	Notes
Video / YouTube	Yes	Yes	The fullest control set and the most detailed reporting
Demand Gen	Yes	Partial	Operates within the shared tier framework; less granular than video
Performance Max	Yes	Limited	Defaults to Maximum; channel visibility is covered in PMax channel-level reporting
Display formats	Yes	Yes	Rules converge after the Display migration into Demand Gen
Search	Not applicable	Partner network handled separately	See the guide to the Search Partners network

The price of caution

Google says it plainly: overusing these controls reduces reach and raises CPM. The mechanics are simple — you shrink the eligible pool, and competition for the remaining premium inventory is fiercer. Three effects show up in reporting:

• CPM rises. The magnitude varies by vertical and geography, but the direction never does. Treat any number from someone else’s case study as a rough guide: moving from Maximum to Limited on video is rarely free.
• Reach accumulates more slowly. The campaign takes longer to hit planned frequency, which feeds straight into planning — see frequency capping and Target Frequency.
• Learning slows. Fewer impressions means less data. For conversion-goal campaigns that can cost more than the CPM delta ever does.

Hence the working principle: strict inventory is a purchase you make on purpose, not a default you inherit. If you cannot name the specific risk you are buying protection from, you are probably just paying more for the same media.

Exclusions: what works and what only feels like control

Content types

Excluding embedded video, live streaming and gaming content is blunt but occasionally justified. Live streams are the most frequent source of awkward adjacency simply because real-time moderation is a harder problem.

Topics and keywords

Keyword exclusion operates on the text surrounding the content, and the failure mode is well known: the word “shot” appears in crime reporting and in basketball highlights alike. An over-long exclusion list carves away good inventory while barely touching genuinely risky placements.

Placements and channels

The most precise instrument is excluding specific channels and sites, ideally maintained at account level so you are not duplicating one list across ten campaigns. Adjacent to this sit brand lists and brand exclusions when the goal is avoiding association with particular brands.

What does not work

Trying to close every risk at once: strict inventory mode, plus a long topic list, plus a thousand negative keywords, plus a manual block list. The result is a campaign with an unpredictably narrow pool that nobody can diagnose, because there is no way to tell which filter ate the reach.

Third-party verification: why bother if Google has settings?

Platform settings describe what should have happened. Independent verification describes what did. At scale those are different statements, and the gap between them is worth measuring.

The independent measurement market runs on a handful of vendors — DoubleVerify, Integral Ad Science and Zefr. They classify content and supply segments that activate inside buying platforms. In 2026 the category took a visible step forward: YouTube in-stream and Shorts received MRC brand safety accreditation, meaning the measurement methodology passed an independent audit.

Verification earns its cost when:

• video and display budgets are material to the company and leadership or a client needs an independent report;
• you buy across several platforms and want one consistent methodology rather than each platform grading its own homework;
• the brand has formal placement obligations, for instance commitments written into a contract.

It rarely earns its cost for small and mid-sized budgets, pure direct-response campaigns, or where no external obligation exists. Verification adds a line item and a layer of latency; for a campaign judged on cost per lead, that maths usually does not close.

Reporting: proving the filters did anything at all

Configuring without verifying is the most expensive habit in this area. The minimum reporting set:

What to check	Question it answers	Cadence
Placement report (where ads served)	Actual distribution across channels and sites	Every 2–3 weeks initially, then monthly
Excluded placement report	What got filtered and how much reach it cost	After every settings change
CPM and reach before / after	The price of the control, in money	Two weeks after a change
Format and device breakdown	Whether a filter removed an entire format, such as Shorts or CTV	Monthly

A useful discipline: before tightening anything, open last month’s placement report and find concrete examples of placements you object to. If there are none, the risk exists only in the abstract and nothing needs changing. Roughly half of all brand-safety discussions should end at this step.

Three scenarios from practice

The brand demands maximum safety everywhere

The request usually arrives from above and arrives absolute. The productive answer is not an argument but a measurement: show the CPM and reach-velocity difference between tiers on a single campaign over two weeks. The conversation moves from “stricter is better” to “here is what it costs and here is the risk it removes”. The usual landing point is Moderate for brand formats and Maximum for performance.

Conversions dropped after tightening, not just reach

The typical cause is slower learning: the pool narrowed, impressions and conversions per unit of time fell, and the model takes longer to stabilise. The fix is to revert the tier, let performance recover, then tighten surgically with specific channel exclusions instead of a global filter.

An unsuitable channel showed up in the placement report

Excluding that one channel at account level is almost always better than changing the inventory mode. One channel costs you a fraction of a percent of reach; a tier change costs tens of percent. The rule: surgical first, global only once the surgical list becomes unmanageable.

A seven-step configuration protocol

1. Write the risk down in words. Not “we want to be safe” but “adjacency to content about X is unacceptable for us”. If you cannot write the sentence, stay on Maximum.
2. Start at Moderate for brand campaigns and Maximum for performance. A sensible default, not a commandment.
3. Capture a baseline first: CPM, reach, frequency, cost per conversion. Without it you cannot price the control you are about to apply.
4. Change one filter at a time. An inventory mode and an exclusion list applied on the same day cannot be separated in analysis afterwards.
5. Open the placement report after two to three weeks. Look at where impressions actually went, not where you assumed they went.
6. Consolidate exclusions at account level so you are not maintaining ten copies of the same list.
7. Review quarterly. Inventory changes, and exclusion lists accumulate until they throttle reach without anyone noticing.

An October 2026 checklist

A short list to close before the changes land.

• Snapshot the settings. Export or screenshot the inventory mode and every exclusion for each active campaign. That is your reference point.
• Audit the automations. Find everything that applies these settings programmatically — scripts, launch templates, integrations — and check which fields they write to.
• Capture baseline metrics. Record CPM, reach, frequency and cost per conversion for the fortnight before the date, so you can separate the change’s effect from seasonality.
• Prune the exclusion lists. Delete anything added “just in case” more than a year ago and never reviewed since.
• Align with the brand team. If there are formal placement requirements, agree how they map onto the new framework — the old categories do not port across directly.
• Diary the review. Set a reminder for mid-October to compare actual placements against September’s.

Mini-glossary

• Inventory mode — the campaign setting that determines how broad a pool of content is eligible to serve.
• Content exclusion — a filter on the characteristics of the material itself: type, topic, surrounding keywords.
• Placement exclusion — a surgical block on a specific channel, site or app.
• Content themes — semantic groupings of content that Google is moving advertisers toward in place of the legacy categories.
• Third-party verification — independent measurement of where impressions actually ran, performed by an outside vendor.
• MRC accreditation — confirmation that a measurement methodology has passed an independent audit.

Common mistakes and what they cost

• Limited applied to everything. Chosen out of caution, paid for with higher CPMs across every campaign, including the ones with no real exposure.
• An exclusion list nobody reviews. Two years in it holds hundreds of entries, a good share of which now block perfectly reasonable inventory.
• Changing settings without measurement. You tighten inventory, see a dip a month later, blame seasonality. With no baseline the causality is unprovable.
• Expecting verification to do something it cannot. An independent measurer tells you where the ad ran. It does not guarantee your CMO will like the answer.
• Ignoring newer formats. Shorts, CTV and shoppable video behave differently — see the breakdown of shoppable CTV in Demand Gen.

How this connects to everything else

Inventory mode is not an isolated toggle. It interacts with reach planning (see video campaign groups and Reach Planner), with budget distribution in Demand Gen, and with how clean the data reaching the algorithm is. If you are simultaneously fighting product feed quality, the piece on Merchant Center automatic improvements follows the same logic: understand the platform’s defaults before configuring on top of them.

FAQ

How does Maximum differ from the old Expanded tier?

Conceptually it is the successor: the widest pool within the platform’s baseline rules. What changed is the name and the framing — tiers are now described by what stricter modes remove.

What does Performance Max use by default?

Maximum. Stricter inventory has to be configured explicitly.

How much does Limited raise CPM?

It depends on vertical, geography and season, so treat any figure from someone else’s case as indicative. The only reliable way to know yours is a before-and-after measurement on one campaign.

What exactly is being removed on 1 October 2026?

Digital content label exclusions and most sensitive category exclusions leave the DV360 API and Structured Data Files. Advertisers move to inventory modes and content themes.

Do I need to act right now?

Yes: snapshot your current settings and audit anything that applies them programmatically. After the date those integrations can run silently without effect.

Are content themes a replacement for the old categories?

Not one-to-one. The mapping is approximate, so verify actual placements after the switch rather than trusting a “similar” setting.

Do keyword exclusions actually work?

They work, but bluntly: the filter reads surrounding text and frequently catches harmless inventory. Long lists tend to cost more reach than they prevent risk.

Does a small advertiser need third-party verification?

Usually not. It is a tool for large budgets and external obligations, and it rarely pays back on direct-response work.

What does MRC accreditation actually mean?

It confirms the measurement methodology passed an independent audit. It is about trusting the numbers in the report, not a guarantee that unsuitable impressions are now zero.

Where can I see where my ads actually ran?

The placement report at campaign level. That is the first thing to open before changing an inventory mode — frequently the risk you are guarding against does not appear in the data at all.

Can different campaigns in one account use different modes?

Yes, and they should. Brand campaigns can run stricter while performance runs wider. One account-wide setting is nearly always a compromise against your own interest.

How often should exclusion lists be reviewed?

Quarterly, plus after any significant change in the formats you run. Lists accumulate and quietly strangle reach over time.

Inventory control is not a “stricter equals better” dial. It is a trade of reach and cost per contact against a specific, named risk. Until the risk is named, every extra filter is simply more expensive advertising.

If you need stable ad accounts behind your video and Demand Gen work, see PPC Rebels agency accounts.