How to Fix Low Fill Rate in Google Ad Manager (GAM)
If your revenue drops despite steady traffic, the first metric to audit is your Fill Rate. A low fill rate indicates that your site is generating ad requests that aren’t converting into impressions. Essentially, you are leaving money on the table.
What Is Fill Rate?
Fill rate is the percentage of ad requests that result in a displayed ad.
Example: If you have 100,000 ad requests but only 70,000 impressions, your fill rate is 70%. While “healthy” rates vary by niche, most publishers aim for 80–95% on standard display inventory.
5 Common Culprits Behind Low Fill Rate
| Reason | The Impact |
| Low Geo Demand | Traffic from Tier 3 countries often lacks the advertiser depth to fill every request. |
| Ad Unit Overload | Too many ad units on one page dilute bid competition and slow down page rendering. |
| Restrictive Sizes | Using non-standard sizes (e.g., 200×200) limits the pool of available creative assets. |
| Aggressive Floors | Setting CPM floors too high rejects bids that would otherwise have filled the slot. |
| Technical Latency | If the page loads too slowly, the user may scroll past or leave before the ad renders. |
Strategic Solutions to Boost Fill
1. Optimize Pricing Rules (UPRs)
Check your Unified Pricing Rules in GAM. If your floors are set based on peak seasonal demand (like Q4), they may be too high for slower periods.
- Action: Experiment with “Floor Shifting.” Lower floors for specific low-demand geographies or devices to capture “long-tail” revenue.
2. Diversify Your Demand Stack
Don’t put all your eggs in the AdSense/AdX basket. High-growth publishers use a multi-layered approach:
- Header Bidding: Bring in Prebid.js to force AdX to compete with external SSPs.
- Open Bidding: Enable Google’s server-to-server demand.
- Backfill/House Ads: Ensure you have a “catch-all” line item (like a self-promotion or a low-priority ad network) to ensure 0% of requests go to waste.
3. Focus on “Signal” over “Volume”
Advertisers bid more—and more often—on high-viewability inventory.
- Lazy Loading: Only request the ad when the user is about to scroll to it. This reduces “unfilled” requests from the bottom of the page that never get seen.
- Ad Refresh: If using auto-refresh, ensure it only triggers when the ad is in view.
4. Standardize Ad Sizes
Enable Multi-Size Pricing. Instead of just requesting a 300×250, allow the slot to accept 300×600 or 320×50. Increasing the number of eligible creatives for a single slot naturally drives up the fill rate.
Technical Troubleshooting Checklist
If your demand looks healthy but fill is still low, check for these “silent killers”:
- CMP Issues: If your Consent Management Platform isn’t firing correctly, Google cannot serve personalized ads (or any ads) in regulated regions like the EU (GDPR).
- app-ads.txt / ads.txt: Ensure your file is reachable and up to date. Missing entries will lead to immediate bid rejection.
- Mismatched Line Items: Check if your line item targeting (Geo, Device, Key-values) is too narrow for the traffic you are receiving.
- Rendering Timeouts: If your Header Bidding timeout is too short, the auction might close before bids arrive.
Summary: When Should You Worry?
A temporary dip during the start of a quarter (January, April, July, October) is normal as advertiser budgets reset. However, a sustained drop usually points to a pricing mismatch or a technical bottleneck.
The Goal: Optimization isn’t about adding more ads; it’s about making your existing ad slots more “eligible” for the highest number of buyers possible.
