9 Ways to Maximize Fill Rates Across Web, App, and CTV Platforms
Fill rate is one of those metrics every publisher tracks but few truly understand. On the surface, it’s just the percentage of impressions filled with ads. In reality, it’s a reflection of demand confidence in your supply.
Getting to “100% fill” isn’t the goal. If you’re filling every slot with low-value or non-transparent demand, you’re inflating numbers while devaluing your inventory. The real goal is to maximize fill rate with competitive buyers, across every environment: web, app, and CTV.
Here are nine advanced ways to do it the things experienced publishers still overlook, and where the biggest leaks usually happen.
1. Stop Letting Empty Impressions Slip Through
Empty impressions aren’t harmless. They drag down viewability, frustrate users, and signal weak supply to exchanges. Yet, even large publishers let them slip through when bidders time out or SSPs pass back.
Where it goes wrong:
- GAM returns a blank because no line item is set up for passback.
- Prebid doesn’t trigger fallback creatives.
- On CTV, pod breaks aren’t filled completely, leaving dead air.
Example: One mid-sized publisher left 2–3% of impressions unfilled because their GAM line items weren’t chained to a backup. At 50M impressions/month, that’s 1.5M empty calls and thousands in lost programmatic revenue.
Fix:
- Use GAM passback line items or house ads to catch unfilled impressions.
- In Prebid, configure a default line item with safe fallback creatives.
- On CTV, structure pods with guaranteed backup campaigns.
Even if your fallback is a house ad, it’s better than serving nothing. Buyers see a clean, fully filled auction, and your site looks trustworthy.
2. Segment Demand by Platform Don’t Mix Web with CTV
A common mistake is lumping all supply together. Buyers hate it. Web, mobile app, and CTV have entirely different buyer logic, auction mechanics, and CPM expectations.
Where it breaks down:
- In-app SDK traffic routed through the same demand paths as desktop → DSPs throttle spend because they can’t parse inventory type.
- CTV bundled with display → buyers underbid because they assume lower quality.
Fix:
- Create separate line items for web, in-app, and CTV.
- Use unique placement IDs and demand paths per platform.
- Send clean signals: device type, app bundle ID, CTV pod metadata.
Result: When buyers see your supply segmented clearly, they assign budget more confidently. One large OTT publisher increased fill by 12% simply by unbundling app and CTV traffic.
3. Set Floors by Geo, Not Just Globally
Global floor pricing is one of the most overlooked killers of fill. A floor that works in the US can wipe out fill entirely in LATAM, India, or SEA.
Example:
We saw a publisher running a $2 floor globally. US traffic filled fine. But Tier-2 GEOs (where buyers typically bid $0.50–$1.20) dropped to ~60% fill. The publisher assumed “low demand,” but it was self-inflicted.
Fix:
- Use geo-specific floors in GAM or Prebid.
- Example: $2.50 for US, $1.20 for LATAM, $0.50 for SEA.
- Layer with format-specific rules your CTV floors shouldn’t match mobile banner floors.
Pro tip: Adjust floors dynamically during high-demand events (Q4 holidays, sports seasons). What clears in June won’t clear in December. Geo-based floors ensure you’re maximizing fill globally while protecting CPMs in premium markets.
4. Shorten Timeouts That Waste Fill
Timeouts are often seen as a latency issue, but they directly impact fill rate. Cut them too short and you block bidders who could’ve delivered. Set them too long and you slow down rendering, tanking viewability.
Where publishers slip:
- Running the same timeout across all platforms.
- Ignoring that video/CTV bidders respond slower than display.
- Never revisiting timeout settings after initial setup.
Example:
A video publisher dropped timeout from 2000ms to 1000ms to “speed up load.” They lost 30% of bids from a premium video SSP that responded at ~1300ms. CPMs and fill collapsed until they lengthened video timeouts again.
Fix:
Audit response-time distributions:
- 800–1000ms for display.
- 1200–1500ms for mobile.
- 2000ms for video/CTV.
Set timeouts slightly above the 80th percentile. That way you catch most valid bids without dragging out the auction.
5. Refresh Ads Without Killing UX
Refresh is one of the fastest ways to maximize fill rate, but it’s often abused. Refreshing every 20 seconds on mobile may spike impressions, but buyers see it as IVT-like behavior and throttle bids.
Best practice:
- Display: 60–90s refresh.
- In-app: refresh tied to engagement events (scroll, active session).
- CTV: refresh only at pod breaks never mid-content.
Example:
A publisher running 30s refresh on desktop saw a short-term CPM bump, followed by a steep fill drop as DSPs flagged inventory for “artificial inflation.” Switching to 90s refresh restored fill and stabilized CPMs.
Lesson: Refresh is powerful, but buyers value engagement-based impressions. Abuse it and fill collapse.
6. Add Direct Campaigns as a Safety Net
When auctions run thin, open exchange won’t always fill everything. That’s where direct campaigns act as insurance.
Use cases:
- Evergreen brand campaigns filling unsold slots.
- Programmatic guaranteed deals with floor commitments.
- House ads promoting owned products or content.
Example:
One CTV publisher ran PG deals with two agencies that only filled 15% of inventory but guaranteed delivery during low-demand hours. That safety net kept fill above 90% even when open exchange bids thinned out.
Takeaway: Direct demand isn’t just about premium CPMs it’s about keeping fill consistent when the open market fluctuates.
7. Block Invalid Traffic Before It Blocks You
Invalid traffic doesn’t just cost clawbacks it cuts your fill rate. Buyers don’t throttle only the flagged traffic; they often scale back spend across your whole domain.
Example:
A mobile web publisher using cheap arbitrage traffic saw a 20% fill drop after one SSP flagged them for IVT. Even their clean organic traffic stopped filling because DSPs distrusted the entire domain.
Fix:
- Run IVT filters (Pixalate, HUMAN, or SSP-provided).
- Monitor discrepancies: big SSP vs GAM mismatches often signal IVT.
- Cut bad sources before buyers cut you.
Clean traffic = higher fill. Buyers want to spend where impressions are safe.
8. Keep Your Transparency Files Clean
Buyers don’t guess. If they can’t verify your supply path, they skip bidding. Broken transparency files = broken fill.
What to check:
- ads.txt must be up to date with every SSP.
- sellers.json must include correct business info.
- s-chain must pass cleanly downstream.
Example:
One publisher forgot to add a new SSP to ads.txt. DSPs excluded 100% of that supply. Fill fell 10% overnight and didn’t recover until the ads.txt was fixed.
Fix:
Audit monthly. Confirm that every partner is properly listed. Even one broken entry can mean thousands of lost impressions. Transparency isn’t compliance theater. It’s how you stay eligible for demand.
9. Treat CTV Like Its Own Market, Not an Extension of Display
The fastest-growing channel is also the easiest to mismanage. Too many publishers approach CTV like display, and their fill rates suffer.
Where it fails:
- Using display floor rules on long-form video.
- Unstructured pods → buyers won’t fill.
- Refreshing mid-stream like web banners.
Fix:
- Structure pods with max ad length, frequency, and clear breaks.
- Set floors based on CTV benchmarks (GRP/CPM), not display CPMs.
- Work only with SSPs specialized in CTV demand.
Example:
A FAST channel running display-style refresh logic mid-content saw massive buyer drop-off. After restructuring pods and floors, fill jumped from 62% to 94%.
Lesson: Treat CTV as its own ecosystem. Display rules don’t apply.
Maximizing fill isn’t about hitting 100% it’s about making sure every impression is filled with demand that buyers trust and value. Publishers who segment their demand, set smarter floors, fine-tune timeouts, and keep transparency airtight consistently see stronger CPMs and more resilient revenue across web, app, and CTV.
At MagicBid, this is the standard we build for publishers every day keeping setups lean, compliant, and demand-ready across every platform support@magicbid.ai
If you’re not making the most of your ad space, you’re leaving money on the table.
MagicBid helps web, app, and CTV publishers maximize revenue with smarter ad placement and optimization tools.
- Web Monetization: Get better ad visibility, higher engagement, and more revenue from every impression.
- In-App Monetization: Connect with premium advertisers to effortlessly boost fill rates and eCPMs.
- CTV Monetization: Deliver high-quality, tailored ad experiences that keep viewers engaged and advertisers paying more.
With MagicBid’s advanced ad tech and expert support, you can turn your traffic into higher earnings without the guesswork.
Connect with us now to get a free ad revenue evaluation.
