Brand Safety Playbook: What to Block at Account Level (and What Not To)

Stop wasting time firefighting placements — centralize brand safety without killing scale

Marketers and agencies are pressured to launch campaigns fast, squeeze CPMs, and trust automated formats like Performance Max. But when ads start appearing next to questionable content, the instinct is to block everything — and then you lose scale, conversions, and learning signals. In 2026, Google Ads' new account-level placement exclusions changes the game: you can now apply a single exclusion set across Performance Max, Demand Gen, YouTube, and Display. That makes it possible to be both safe and scalable — if you use a prescriptive playbook.

Quick summary (what you’ll get from this playbook)

• How to create an account-level blockage strategy that preserves reach
• Exact categories, negative keyword triggers, and placement types to block — and which not to
• Step-by-step implementation for Google Ads Manager accounts and client governance
• Monitoring, testing, and escalation templates for 2026's automation-first ecosystem

Why account-level placement exclusions matter in 2026

Late 2025 and early 2026 have accelerated two trends: (1) ad automation (Performance Max, Demand Gen) is the default channel mix for many advertisers, and (2) transparency around exact placements is still limited compared to campaign-level control. Google’s January 2026 rollout of account-level placement exclusions responds to advertiser demand for stronger guardrails across automated formats, letting you prevent spend on specific sites, apps, and video placements from one central list.

"Advertisers can now apply one exclusion list at the account level. Exclusions apply across Performance Max, Demand Gen, YouTube, and Display campaigns." — Google Ads rollout, Jan 2026

This reduces admin overhead and the risk of human error when you manage dozens of campaigns and sub-accounts. But centralization also raises new questions: what to block globally vs. what to leave flexible for campaign-level targeting? Block too much and you starve automation; block too little and you invite publisher risk.

Prescriptive framework: Balance safety with scale

Apply a three-tier framework to decide what to exclude at the account level versus campaign-level or ad-group level:

1. Core Safety (Block at account-level) — Universal high-risk placements you never want across any brief or client (e.g., explicit adult, illegal content, extremist propaganda).
2. Conditional Safety (Campaign-level) — Contextual or vertical risks where allowances depend on the brand or objective (e.g., political content for a B2B SaaS brand vs. a political advocacy client).
3. Performance-Driven Exceptions (Do not block) — High-reach, high-conversion inventory that should be evaluated by data (e.g., major publishers, mainstream UGC sections on premium sites).

Why this structure works in 2026

Automation needs clean guardrails. Account-level exclusions are your broad safety net; they should remove inventory that's always unacceptable. Conditional safety needs human judgment and data; handle it at the campaign level. And keep a data-first approach for performance exceptions — use experimentation and publisher-level metrics to decide.

Detailed list: What to block at account level (and exact examples)

Use this as your default master exclusion list. Start here, then tune for clients and verticals.

Block these categories globally (account-level)

• Sexually explicit content — adult sites, escort services, porn aggregators. Example URLs: known adult domains, app categories labelled "explicit."
• Child sexual abuse and exploitation — zero tolerance; immediate exclusion and reporting.
• Illicit drugs and trafficking — marketplaces or content encouraging illegal activity (beyond basic informational health articles).
• Extremist and violent ideology — propaganda, recruitment, extremist forums.
• Suicide, self-harm promotion — anything that encourages or instructs self-harm.
• Piracy and illegal streaming — file-hosting sites, illegal live-stream portals.
• Hate speech and explicit harassment — sites that host targeted abuse or dehumanizing content.
• Malware / scam pages — domains flagged by security vendors or Google Safe Browsing.

What to block with strong negative keywords (contextual controls)

Negative keywords help prevent ads from serving alongside sensitive text content in contextual inventory. At account level, include a conservative list that maps to high-risk semantics:

• Keywords: sex, porn, explicit, child abuse, ISIS, bomb, suicide, murder, prostitution, cocaine, heroin, pornographic, escort
• Use phrase matches for compound phrases (e.g., "buy heroin", "child abuse images") and broad match negatives sparingly — they can over-filter contextual reach.

Placement types to exclude globally

• Unknown or raw inventory apps — small apps without established publishers or that are frequently flagged.
• Low-viewability ad units — domains with viewability < 30% historically in your account data.
• Domains flagged by verification tools — integrate lists from IAS, DoubleVerify, or Adalytics; exclude matches at account-level.

What not to block — preserve scale and conversions

Some instincts to block are costly. Before you add a domain or category to the account-level list, consider these rules:

• Don’t block entire mainstream publishers for isolated articles. A single controversial opinion piece doesn’t make the whole publisher a no-go. Use placement-level exclusions only after you see poor performance signals tied to the site.
• Avoid blanket topic bans that cut into your target audience. For example, "user-generated tech forums" may drive strong installs for SaaS tools — test first.
• Don’t use overly broad negative keywords (e.g., single-word negatives such as "sex") when those words appear in legitimate contexts (e.g., "sexual health clinic" for a health client). Phrase and exact negatives reduce collateral damage.
• Don’t exclude YouTube entirely unless client risk tolerance is zero. YouTube drives reach and video conversions; use channel or placement-level exclusions instead of a full block.

Step-by-step implementation: From audit to governance

Follow this operational playbook to create a repeatable, auditable account-level exclusion program.

1. Run a fast placements audit (week 0)

• Export the last 90 days of placement reports across Display, YouTube, Performance Max, and Demand Gen.
• Tag placements by risk signals: publisher reputation, content category, viewability, conversion rate, and fraud score.
• Flag immediate account-level exclusions (see list above) and a second set for campaign-level review.

2. Build the master exclusion lists (week 1)

1. Create an account-level placement exclusion list in Google Ads Manager (MCC) and populate it with: high-risk domains, app IDs, and video placements.
2. Create a separate negative keyword list for account-level use with phrase matches for sensitive terms.
3. Tag lists with IDs and maintain them in a shared repository (Sheets + version history) accessed by the team and client stakeholders.

3. Apply & verify (week 1)

• Attach the account-level list to the Google Ads account. Confirm it applies across Performance Max, Demand Gen, Display, and YouTube.
• Run a dry-run placement report after 48–72 hours to validate no ads are serving to excluded placements.

4. Governance and exceptions (ongoing)

• Define an exceptions process: campaigns can request a placement removal from the account list for a valid, documented reason and test plan.
• Set a review cadence: quarterly for general accounts, monthly for political or high-risk verticals.
• Maintain an owner (agency or brand safety lead) responsible for updates and incident response.

Testing and measurement: How to avoid overblocking

Data is the antidote to overblocking. Use this test matrix to quantify tradeoffs between safety and performance.

1. A/B test exclusion lists: Run identical campaigns with and without the account-level exclusions to measure CPA, conversion rate, and reach loss.
2. Publisher-level lift tests: For high-traffic publishers on the exclusion fence, run a 2-week micro-test and measure LTV, ROAS, and viewability.
3. Time-window checks: Automation can reallocate spend quickly. Inspect placement shifts daily for 7 days after applying new exclusions.

Accept up to a 10–20% short-term reach loss for a cleaner signal set that reduces brand risk; anything beyond that requires a targeted review to understand whether exclusions are too aggressive.

Advanced tactics for 2026: Contextual safety, cookieless signals, and AI-aided monitoring

Brand safety in 2026 relies less on brittle URL blacklists and more on layered signals.

• Contextual verification: Use semantic analysis tools to flag placements whose page content matches sensitive topics even if the domain is mainstream.
• First-party signal layering: Feed conversion and viewability data back into your exclusion decisions. If a publisher drives high conversions and low brand risk, keep it.
• AI monitoring: Deploy automated scanners that crawl landing pages and placement pages to score suitability in real time. Integrate with your exclusion list via API if possible.
• Privacy-first targeting: With cookieless identifiers and cohort signals maturing in 2026, contextual and placement-level controls are more important than ever.

Publisher risk: How to evaluate and score domains

Create a simple publisher risk score (0–100) using weighted inputs. Example weights:

• Content category risk (40%) — Does the site frequently publish high-risk topics?
• Historical performance (20%) — Viewability, CTR, conversion rate.
• Verification score (20%) — IAS/DV classification and fraud metrics.
• Brand incidents (20%) — Previous controversies that led to client complaints or media attention.

Set thresholds: score > 70 = account-level block; 40–70 = campaign-level review; < 40 = allowed but monitored.

Agency playbook: Client-facing controls and transparency

Agencies must balance client risk tolerance with performance. Here’s a practical governance policy to include in SOWs and onboarding:

• Default safety posture: Set the account-level exclusion baseline for all clients (use the global list above).
• Client approval tiers: Minor changes (campaign-level exclusions) require channel manager approval; account-level changes require legal/brand sign-off.
• Monthly disclosure: Provide clients a placements dashboard that shows excluded domains, reasons, and any exception requests.
• Escalation process: Rapid removal of placements (within 24 hours) if a brand-safety incident is reported.

Real-world example (agency case study)

Client: Mid-size FinTech with a conservative brand policy.

Problem: Performance Max campaigns delivered conversions but ads occasionally ran adjacent to volatility-driven forums and political opinion pieces. The client demanded immediate action.

Action taken:

1. Audited 90-day placement data and created a master account-level exclusion list using the categories above.
2. Ran a 14-day A/B test where 50% of spend used the new account-level exclusions and 50% used campaign-level manual exclusions only.
3. Monitored CPA, ROAS, conversion quality (lead quality scoring), and reach.

Result: Account-level exclusions reduced low-quality impressions by 28% and improved lead quality (sales-qualified lead rate) by 12%, with only a 9% reduction in reach. The client accepted the tradeoff and rolled the account-level exclusions across all regions.

Checklist & templates — quick starter kit

Account-level exclusion checklist

• Create exclusion list in Google Ads Manager account.
• Populate with core safety categories and negative keyword list.
• Apply to all eligible campaigns (Performance Max, Demand Gen, Display, YouTube).
• Run verification report after 72 hours.
• Implement monitoring: daily for week one, weekly thereafter.

Sample negative keywords (phrase match to start)

• "buy heroin"
• "child sex"
• "how to build bomb"
• "free porn"
• "escort service"

Common pitfalls and how to avoid them

• Pitfall: Applying a huge URL blacklist without testing. Fix: A/B test and measure conversion impact before adopting globally.
• Pitfall: Over-reliance on third-party lists without validation. Fix: Cross-check with your conversion and viewability data.
• Pitfall: No governance — lists change without audit trail. Fix: Use version-controlled sheets and assign an owner.

Future predictions: Brand safety to 2028

By 2028, expect three developments that will shape how you design exclusion strategies:

• Real-time semantic scoring: AI will score page suitability in real time and allow near-instant exceptions.
• Greater publisher transparency: Buyer-level APIs and publisher certifications will make real-time inventory verification standard.
• Shift to suitability over blanket safety: Brands will prefer nuanced suitability controls (contextual sentiment, audience alignment) rather than binary block lists.

Final takeaways — the rules to follow

• Use account-level exclusions for universal, high-risk inventory. That's the safety net your automation needs.
• Keep conditional or contextual blocks at campaign level to preserve scale for high-performing publishers and inventory.
• Measure, test, and iterate — data should drive permanent additions to the master list.
• Govern aggressively — define owners, approval flows, and review cadences for updates.

Google Ads' 2026 account-level exclusions are a powerful tool — but they're most effective when combined with measurement, contextual controls, and a clear governance process. Use the frameworks and lists above to centralize brand safety, preserve ad scale, and let automation do its job without exposing your brand.

Call to action

Need a ready-made, editable account-level exclusion list and A/B test setup for your next Performance Max rollout? Download our 2026 Brand Safety Exclusion Pack or book a 30-minute audit with our growth team to get a custom list tuned to your vertical and conversion goals.

Related Reading

• Martech Buying Guide for Operations Leaders: What to Buy, Pilot, and Postpone
• From kitchen to 1,500-gallon tanks: how pet-food startups can scale responsibly
• Is the Roborock Wet-Dry Vac a Good Pawnshop Flip? Repairability and Demand Explained
• Vulnerability in Songwriting: What Nat and Alex Wolff Teach Us About Emotional Honesty
• Moving Day Logistics: How to Coordinate Moves for 1,200 Real Estate Agents and Their Clients