Case Study: Cutting Wasted Spend with Account-Level Placement Exclusions

Cutting Wasted Spend with Account-Level Placement Exclusions — a 2026 Case Study

Hook: If your Google Ads account is bleeding budget on irrelevant Display, YouTube, or Demand Gen placements while automation promises efficiency, you're not alone. In 2026, with platforms pushing more automated formats, many marketers still fight invisible wasted spend and fractured exclusion lists. This case study shows, step-by-step, how applying account-level placement exclusions centralized control, reduced wasted placements, and improved ROAS.

Executive summary (most important first)

An anonymized mid-market e-commerce advertiser implemented Google Ads’ new account-level placement exclusions (rolled out Jan 15, 2026) across Performance Max, Demand Gen, YouTube, and Display. Within 8 weeks they saw:

• Wasted placement spend down 42% (absolute dollars tied to low-quality placements)
• ROAS improved 28% on automated campaign types
• Operational time saved: weekly exclusion management went from ~6 hours to ~45 minutes

Below is the full, reproducible workflow: audits, decision rules, implementation, monitoring, and automation tips you can use today.

Why account-level exclusions matter in 2026

Google’s January 15, 2026 update added account-level placement exclusions, letting advertisers block websites, apps, and YouTube placements across all eligible campaigns from one list. This matters because:

• Automation (Performance Max, Demand Gen) increases reach and unpredictability — single campaign exclusions were inefficient.
• Brand safety and cost control require centralized guardrails that don’t break automation.
• Large accounts with many campaigns previously wasted hours replicating exclusions and opened the door to inconsistent coverage.

Case context — the anonymized advertiser

The advertiser: a mid-market retail brand (revenue: ~$60M/yr) with a diversified ad mix: Search, Shopping, Performance Max, Demand Gen, Display, and YouTube. They had a central marketing ops team and regional media buyers. The challenge: automated formats drove volume but also a long tail of low-quality placements that contributed to high CPAs and noisy reporting.

Baseline KPIs (30 days pre-intervention):

• Monthly display/YouTube spend: $120,000
• Spend on low-quality placements (identified via manual audit): $28,500 (23.8% of display/YouTube spend)
• Overall ROAS for automated channels: 2.1x
• Weekly person-hours for exclusion management: ~6 hours

Step-by-step optimization workflow

We used a phased approach: Audit → Rules → Account-level List → Test → Scale → Automate. Follow this order to avoid over-blocking and to preserve automation gains.

Phase 1 — Audit (3–7 days)

Objective: find where money was being wasted and define thresholds for exclusion.

1. Export placement-level data for 60–90 days from Google Ads and combine with GA4 / server-side conversion data. Include metrics: spend, impressions, clicks, conversions, view-through conversions, conversion value, ROAS, avg. viewability (if available), and invalid traffic indicators.
2. Segment by campaign type (Performance Max, Demand Gen, Video, Display). Automation mixes inventory differently — treat them separately.
3. Flag placements that meet exclusion criteria:
   • High spend but zero conversions over 30–60 days
   • CTR < 0.05% and viewability < 30%
   • High VTR but extremely low click-to-conversion rates (common with auto-play cheap views)
   • Known brand-safety or app inventory flagged by third-party lists
4. Calculate wasted-spend ratio: sum spend on flagged placements / channel spend.

Phase 2 — Decision rules and guardrails (1–2 days)

Create clear rules before you block anything. Rules should balance waste reduction with preserving reach for automation:

• Exclude a placement if it spent > $250 and produced 0 conversions in the last 60 days.
• Exclude if viewability < 30% and CPA > 3x target CPA.
• Flag—but do not exclude—placements with spend between $50–$250; monitor for two weeks before action.
• Always maintain a holdout sample: leave 5% of budget unexcluded for statistical validation.

Phase 3 — Build the account-level exclusion list (1–2 days)

Using the audit output, build a single, centralized list in Google Ads that contains domains, app IDs, and YouTube channel IDs you want blocked. Use clear naming and versioning: Exclusions_2026_Q1_v1.

Example list items (anonymized):

• domains: example-lowviewability-site.com
• app IDs: com.example.suspiciousapp
• YouTube channels: UCBadPlacement12345

Tip: Include reason codes in a separate CSV for auditability (e.g., no-conversions, low-viewability, suspected IVT).

Phase 4 — Controlled rollout and A/B holdout (2–4 weeks)

To prove causality, run a controlled experiment:

1. Split eligible campaigns or account traffic: 70% get the account-level exclusions, 30% remain as-is (holdout).
2. Keep other settings constant (bids, creatives, audiences) and monitor for 14–28 days depending on volume.
3. Track metrics: spend, conversions, ROAS, CPA, conversion rate, and placement spend share.

Results in this study: within 21 days the test group showed a 35% reduction in placement-level wasted spend and a 21% lift in ROAS vs the holdout (statistically significant at p < .05).

Phase 5 — Scale and refine (ongoing)

After validating, roll the account-level list across the entire account. Then:

• Run weekly placement reports and add newly-identified waste to a staging list.
• Keep a monthly review cadence and maintain a change log with reason codes.
• Re-evaluate holdout periodically to ensure exclusions aren’t removing high-performing long-tail placements.

Implementation specifics and templates

Below are pragmatic assets used in the project you can copy.

Sample exclusion CSV format (columns)

Use a simple CSV for recordkeeping and imports: placement, type, first_seen, last_seen, spend_90d, reason_code, notes

Google Ads API / Console flow

1. Admin: In Google Ads UI, go to Tools > Exclusions > Account-level placement exclusions > New list.
2. Paste your domains/channel IDs/app IDs and name the list with a date/version.
3. Apply to all eligible campaign types (Performance Max, Demand Gen, Display, Video).
4. Use Google Ads API to sync lists across manager accounts if you manage multiple MCCs—automate with a weekly job.

Pseudocode for a weekly automation job (high-level) (high-level):

Query placement performance > apply decision rules > append new placements to staging CSV > human review > if approved, push to account-level exclusion list via API.

Key metrics to monitor (what to watch every week)

• Placement spend share: percent of channel spend coming from excluded placements (aim to reduce).
• Wasted-spend dollars: absolute $ tied to flagged placements.
• ROAS by campaign type: compare automated formats vs holdout baseline.
• Conversion rate and CPA on automated channels after exclusions.
• Traffic quality signals: viewability, session duration, bounce rate (GA4 or server-side).
• False positive risk: percentage of newly blocked placements later producing conversions (should be low; if not, loosen rules).

Outcomes from the anonymized case

After full roll-out and 8 weeks of monitoring, the advertiser saw the following:

• Wasted placement spend reduced from $28,500 to $16,500 (42% reduction)
• ROAS on automated channels rose from 2.1x to 2.7x (+28%)
• CPA improved 22% on Display/Video
• Operational time for exclusion management dropped from ~6 hours/week to ~45 minutes/week
• Brand complaints related to off-target creatives fell to near zero

Key takeaway: Centralized, account-level exclusions cut waste and improved returns faster than campaign-by-campaign micro-management.

Risks, trade-offs, and how to avoid over-blocking

Exclusions are powerful but blunt. Over-blocking can reduce reach and hurt machine learning signals. Avoid these mistakes:

• Blocking too aggressively: Don’t exclude placements with low short-term conversions but high assisted conversions without testing.
• No holdout: Always maintain a control group so you can attribute impact.
• Poor documentation: Keep change logs and reason codes so you can roll back if needed.
• Ignoring audience overlap: Sometimes a placement drives branding metrics not immediate conversions — measure view-through and assisted conversions.

Automation & operational scale (how we cut the 6 hours/week to 45 minutes)

Automation was pivotal. Here’s what was automated:

• Weekly placement harvest job: SQL job (BigQuery) joins Google Ads placement report with GA4/server-side conversions and flags candidates using decision rules.
• Staging queue with human review: A Slack notification with top candidate placements and a one-click approve/deny button (via a small webhook) reduced review time.
• API push: Approved placements are appended to the account-level exclusion list via Google Ads API with versioned naming.
• Audit dashboard: Looker/Sheets showing recent exclusions, reversal history, and performance deltas for transparency.

2026 trends and what they mean for placement controls

Looking ahead in 2026, advertisers should expect:

• More platform-level guardrails: Google’s account-level exclusions are the start — expect similar centralized controls on other large platforms and programmatic partners.
• Deeper integration with privacy-safe measurement: as cookieless measurement evolves, placement signals will shift — rely more on server-side & modeled conversions.
• AI-assisted exclusions: Automated suggestions will flag suspicious placements, but human-in-the-loop governance will remain essential to avoid false positives.

Actionable checklist — implement within 30 days

1. Export placement-level data for 60–90 days; identify high-spend zero-conversion placements.
2. Create decision rules (spend thresholds, viewability rules, etc.).
3. Build an account-level exclusion list and a staged CSV with reason codes.
4. Run a 2–4 week holdout A/B test (70/30) to prove causality.
5. Automate weekly harvesting & human review; push approved blocks via API.
6. Monitor ROAS, wasted spend, and false positives; iterate monthly.

Proven templates (copy/paste starting points)

Decision rule (copy):

Exclude if spend > $250 AND conversions = 0 in last 60 days OR viewability < 30% AND CPA > 3x target.

Staging CSV header (copy): placement,type,first_seen,last_seen,spend_90d,conversions_90d,reason_code,reviewed_by,approved_on

Final thoughts — why this matters now

In 2026, with automation expanding and platforms offering account-level controls, advertisers who centralize exclusions gain both efficiency and performance. The anonymized case above shows clear causality: centralized exclusions reduced wasted placements and materially improved ROAS without undermining the reach advantages of automation. The key is disciplined auditing, controlled testing, and automating the repeatable parts while keeping humans in the loop for judgment calls.

Call to action

If you manage automated campaigns and still wrestle with wasted placement spend, start with the checklist above. For a hands-on jumpstart, download our free Account-Level Exclusions Audit Kit (2026) or book a 30-minute optimization review to map the first 30 days. Centralize exclusions, protect your automation, and convert wasted impressions into higher ROAS.

Related Reading

• Field Review: Five Cloud Data Warehouses Under Pressure — Price, Performance, and Lock-In (2026)
• Engineering Operations: Cost-Aware Querying for Startups — Benchmarks, Tooling, and Alerts
• Practical Playbook: Responsible Web Data Bridges in 2026 — Lightweight APIs, Consent, and Provenance
• Edge-First Model Serving & Local Retraining: Practical Strategies for On‑Device Agents (2026 Playbook)
• Smart Fermentation & Low‑Glycemic Meal Prep: Advanced Strategies for People with Diabetes (2026)
• Trade-Offs of Rechargeable Hot-Water Devices: Battery Waste, Heating Efficiency and Safety
• Platform Exodus Playbook: When to Move Your Fan Community From Big Tech to Friendlier Alternatives
• DIY: Set Up a Safe, Timed Boost for Bathroom Fans Using Smart Plugs and Humidity Sensors
• How Agricultural Export Reports Affect Freight Corridors — And Where to List Truck Parking