84,624 Placements Later: What We Found Hiding in Google Ads
A forward-thinking digital agency brought us in to take a deeper look at their clients' Google Display campaigns. They were already managing the accounts well, hitting KPIs, and delivering results. But they wanted to know: could there be hidden waste at the placement level that standard campaign management doesn't catch?
Turns out, yes. Quite a lot of it.
The Scale of What We Looked At
Over four months, we analyzed 84,624 placements across 5 client accounts and 20+ campaigns.
Standard campaign management focuses on bids, budgets, audiences, and creative. That's the right priority. But placements (the actual websites and apps where your ads appear) rarely get audited at scale. Google manages that layer automatically, and most of the time it works fine.
Most of the time.
What Was Hiding in Plain Sight
Google's Display Network places ads across millions of websites and apps. The vast majority are legitimate. But in 84,624 placements, we found patterns that only surface when you look at the full picture:
Gaming and browser game sites. Ads for business loans and professional services showing alongside online poker and casual games. Real clicks, zero buying intent. We identified 135 domains like this across all accounts.
Bot traffic signatures. A repeating pattern: placements generating exactly RM 41 in spend with no engagement beyond the initial click. Same signature, different sites, multiple accounts. Six domains, blocked everywhere.
Wrong-intent placements. One account selling professional courses had ads running on job boards and immigration forums. The traffic was real, but the audience was looking for jobs, not courses. 13 domains removed.
Low-quality mobile apps. Kids' games, wallpaper apps, ad-farming apps. Cheap clicks, no conversions. 30 apps excluded.
None of this reflects poor campaign management. It's a structural feature of how Google's Display Network operates. The waste was baked into the default settings.
The Results
After deploying 796 placement exclusions across all accounts:
| Metric | Result |
|---|---|
| Validated annual savings | RM 38,634/year |
| CPC reduction | 22–57% depending on account |
| Placement inventory renewal | 85%+ of placements were new after cleanup |
| Conversion cost (one account) | Dropped from RM 54 to RM 37 |
| Conversions (one account) | Up 50% at lower cost |
These are measured numbers, validated over 19 days and annualized. Not projections.
The Surprise: Better Reach After Exclusions
The most interesting finding was what happened after we blocked the bad placements. Google's algorithm was forced to find new inventory. The result was more reach, not less:
| Account | Before | After | New Placements |
|---|---|---|---|
| Account A | 1,056 | 1,593 | 88.9% new |
| Account B | 217 | 563 | 92.5% new |
| Account C | 24,419 | 38,361 | 84.3% new |
| Account D | 17,370 | 25,613 | 85.8% new |
Exclusions don't limit reach. They force the algorithm to explore better inventory. Every account expanded significantly.
Why This Goes Unnoticed
Three reasons this kind of waste persists even in well-managed accounts:
Volume. 84,624 placements is not something you review manually. Most teams check the top 20 by spend and move on. The waste hides in the long tail.
Defaults. Google includes the entire Display Network by default. Most campaigns launch without placement exclusions. The waste starts on day one.
No categorization system. Without a framework for sorting placements by quality (gaming, spam, bot traffic, relevant, highly relevant), you can't see patterns across accounts.
What We Built
We didn't stop at a one-time audit. We built a repeatable system:
- Automated placement extraction and categorization
- Pattern detection that flags waste across accounts
- Bulk exclusion deployment (one script, all accounts)
- Monthly monitoring to catch new bad placements as they appear
Then we trained a second operator on the same system. They produced 27 independent reports and deployed their own exclusions. Same methodology, same results, zero supervision.
The agency now runs this system themselves. That was always the goal.
Is This Relevant to Your Campaigns?
If you're running Google Display campaigns and haven't done a placement-level audit, there's almost certainly waste you haven't seen. That's not a criticism of your team or your agency. It's how the system works by default.
A first audit can surface the biggest waste categories in a few hours. We're happy to have a straight conversation about whether it's worth looking into for your accounts.