What Separates the Top 10% of Website Publishers from Everyone Else (Data-Backed)

Key Points

• Publishers above the median on both page depth and ad density earn 17x more per session than those below on both. This is the single most important compound effect in the dataset.
• Impressions per pageview is the strongest predictor of revenue per session across all publishers, with a correlation of 0.59.
• Fill rate, not CPM, is the most underrated lever: publishers above 90% fill earn 5.3x more per session than those below 40%.
• Amazon accounts for an average of 20.5% of total site revenue where it runs, losing that access creates a structural hole, not a marginal dip.
• The right optimization strategy depends entirely on your vertical: gaming and education win on volume, sports and news win on audience quality.

The top 10% of publishers aren't doing one thing differently. They're doing four things differently, and they're doing them simultaneously.

That's the finding that comes out of Playwire's analysis of aggregated performance data across thousands of publisher sites. Revenue per session (RPS) is the metric that matters. Not CPMs in isolation, not session duration. Correlate everything against RPS and a clear picture emerges of what separates elite publishers from the rest. It's not what most publishers obsess over.

CPM Is Not the Top Predictor of High Performing Publisher Ad Revenue

Most publishers spend time optimizing CPMs or chasing viewability. Neither is the strongest predictor of revenue per session.

Impressions per pageview is. The Pearson correlation between impressions per pageview and RPS is 0.59, the strongest single-variable relationship in the entire dataset. Impressions per session comes in at 0.55. CPM sits at 0.22. Fill rate at 0.12. Session duration, the metric everyone intuitively reaches for as a quality signal, correlates at essentially zero (−0.03).

The implication is uncomfortable for publishers who've been treating time-on-site as a proxy for revenue quality. A user who spends 10 minutes on one page is worth less than a user who spends 4 minutes across four pages. Duration without depth is just an open tab.

The step-change in RPS as impressions per pageview increases makes this concrete:

Publishers averaging 15+ impressions per pageview earn 21x more per session than those under 1 impression per page. Even moving from 2–4 to 4–8 impressions per page is a 1.6x lift. Ad layout decisions are monetization decisions. The top 10% have internalized this.

The Compound Effect That Changes Everything

Ad density per page matters. Page depth per session matters too. The real multiplier happens when both are high at the same time.

Splitting publishers on whether they fall above or below the median on each dimension reveals the true size of the gap:

High ad density alone gets you to 69. High page depth alone gets you to 56. Combine both and you're at 100 — a 17x premium over publishers sitting below median on both dimensions. This isn't a marginal gain from incremental optimization. It's a different revenue tier entirely.

Top publishers understand that content architecture is a monetization decision. The structures that move users from one page to the next — internal linking, content recommendations, lesson sequences, game levels — determine how many times your ad stack gets to work per session. Publishers who build for depth, not just engagement, earn more. Gaming and education publishers tend to crack this most effectively because their content structures do it naturally.

Essential Background Reading:

• 5 Ad Revenue Metrics Publishers Should Track that Might Surprise You: The metrics that actually predict revenue performance, and why the ones most publishers watch closely aren't on the list
• Publisher Ad Revenue Maturity Model: Where Are You on the Revenue Curve?: A framework for diagnosing where your monetization stack stands today and what the next stage of optimization looks like
• Ad Density Is the #1 Predictor of Publisher Revenue Per Session: A deep dive into why impressions per pageview outperforms every other metric in the dataset and what to do about it
• Session Duration Is a Lie. Page Depth Is the Signal.: Why time-on-site misleads publishers and what the data actually shows about session depth and revenue

Fill Rate Is Leaving Real Money on the Table

Publishers talk about CPM constantly. They should be talking about fill rate.

The summary correlation (0.12) undersells fill rate's actual impact on revenue per session. Segment by fill bracket and the picture is harder to dismiss:

Publishers above 90% fill earn 5.3x more per session than those below 40%. That's not a rounding error. That's a structural revenue gap.

The Floor Price Trap

Aggressive floor pricing is one of the most reliably self-defeating moves in programmatic monetization. Within the same demand tier, publishers with aggressive floors run at roughly half the fill rate of those with calibrated floors. Despite charging nearly 2.5x the CPM per impression, high-floor publishers generate 19% less revenue per session.

The math is blunt: a lower CPM with twice the fill wins. Every time.

Top publishers configure floors to match the demand that's actually present in their auction, not the demand they want to be there. That means dynamic floors responding to real bidder behavior, not static minimums set once and forgotten. It also means a deep enough header bidding stack to ensure genuine competition for every impression. High fill doesn't happen by accident. It's the result of multiple demand partners competing, smart unified pricing rules, and formats buyers actually want.

Geography Sets Your Fill Ceiling

Fill rate isn't entirely in a publisher's control. A significant portion is dictated by traffic geography. Audiences outside North America and Western Europe attract meaningfully less advertiser demand, which caps fill regardless of how well the monetization stack is configured.

The practical implication: benchmarking your fill rate against publishers with fundamentally different audience geographies produces meaningless comparisons. The goal is to maximize fill within your geographic cohort through smarter demand diversification, geo-specific floor pricing, and formats that attract broader global demand.

What Viewability Optimization Gets Wrong

Viewability matters, but it has a ceiling effect most publishers don't account for. The 80–90% viewability bracket outperforms the 90%+ bracket on median RPS. Past 80%, buyer differentiation shifts to fill rate, inventory volume, and audience quality. Chasing 95% viewability at the expense of fill is not a winning trade.

Related Content:

• How to Build a Website Content Architecture That Earns More Ad Revenue: The structural content decisions that drive page depth, session volume, and the compound revenue effect
• How to Increase Impressions Per Pageview Without Hurting User Experience: Practical approaches to building ad density while keeping UX intact and fill rates healthy
• How to Right-Size Your Price Floors Without Leaving Money on the Table: How to calibrate floors to actual demand instead of aspirational CPMs, with the math that explains why it matters
• The Publisher's Guide to Viewability Optimization: Chasing Perfect isn't the Answer: Where the viewability ceiling sits and what to optimize once you've crossed the threshold that actually moves revenue
• Fill Rate Is the Most Underrated Revenue Lever, Though Complex to Change: Why fill rate impact is systematically underestimated and the specific levers that actually move it

The Amazon Gap Is Structural, Not Marginal

Here's a number that should get every publisher's attention: Amazon accounts for an average of 20.5% of total site revenue where it runs actively, with a median of 17.6%.

That's roughly one dollar in every five to six earned. Amazon isn't a supplemental bidder. It's a structural revenue pillar for publishers who have it. Publishers who've lost access to Amazon as a bidder aren't dealing with a 5% revenue shortfall. They're operating with a 20% hole in their demand stack that CPM optimization alone cannot fill.

Demand concentration risk is one of the most underappreciated vulnerabilities in publisher monetization. When revenue is structurally dependent on any single bidder, even a high-performing one, you're exposed in ways that floor pricing and viewability tuning can't protect against. Top publishers treat demand breadth as a non-negotiable: multiple strong bidders competing for every impression, with no single source accounting for more than a manageable share of total revenue.

Your Vertical Determines Your Primary Lever

One of the most practically important findings in the dataset is also one of the most frequently ignored: the right optimization strategy depends entirely on what kind of content you publish.

The data is clear across six major verticals:

Gaming, entertainment, and education are inventory volume businesses. More impressions per session is what moves RPS. Fill rate is essentially uncorrelated with gaming revenue (r = −0.01). Adding demand partners matters far less than adding ad slots. Education shows the strongest single-variable relationship in the entire dataset: impressions per session at r = 0.93, driven by lesson loops that naturally push students through page after page.

Sports, news, and technology are audience quality businesses. The CPM premium in sports runs 64% above gaming, but impressions per session average less than half. Revenue is overwhelmingly driven by who's reading, not how many ad slots are on the page. A sports publisher obsessing over ad density is solving the wrong problem. So is a gaming publisher chasing CPM improvements.

Applying a generic playbook across verticals doesn't just underperform. It actively misdirects effort. Knowing your primary lever is the prerequisite for using it correctly.

Next Steps:

• Publisher Revenue Optimization by Vertical: How to Maximize Ad Revenue in YOUR Vertical: A full breakdown of the primary revenue lever in each content vertical and how to apply it to your specific situation
• How to Diagnose and Fix Your Fill Rate Problem: A step-by-step diagnostic for identifying where fill is leaking and which fixes will move the number fastest
• How to Recover Publisher Revenue After Losing Amazon as a Bidder: Specific strategies for rebuilding the demand stack when a structural revenue pillar disappears
• What Separates the Top 10% of Website Publishers from Everyone Else (Data-Backed): The full data-backed profile of elite publisher performance and the variables that separate them from the median

What High Performing Publishers Do Differently

Across all of these findings, a consistent profile emerges. High-performing publishers aren't doing one thing well. They're doing several things well in combination.

• Ad density built for their vertical: Gaming and education publishers serve 14+ impressions per session on average. Sports and news publishers protect audience premium instead.
• Content architecture that drives depth: Multi-page consumption patterns, internal linking, engagement loops. Every page loaded is another shot at revenue.
• Fill rate above 75%, ideally above 90%: Achieved through header bidding depth, well-calibrated floors, and high-demand formats, not aggressive CPM minimums.
• Diversified demand with no single-bidder dependency: Amazon is a structural revenue pillar where it runs. Losing any single major bidder shouldn't collapse your revenue stack.
• Vertical-matched optimization strategy: Volume levers for volume verticals. Quality levers for quality verticals. No universal playbook applied blindly.

Publishers sitting in the bottom half of the RPS distribution aren't suffering from bad content or weak audiences. They're optimizing the wrong things. Watching CPM when they should be watching impressions per pageview. Chasing viewability percentages past the point of diminishing returns; the 90%+ viewability bracket actually underperforms the 80–90% bracket on median RPS. Holding floors that kill fill without raising total revenue.

The gap between median and elite publisher performance is large. It's also closable. Not through one change, but through getting the right combination of variables pointed in the right direction at the same time.

See It In Action:

• Gaming Publisher Revenue Guide: Why Ad Density Is Everything: How the volume-first strategy plays out in practice for gaming publishers — with the data showing exactly why density is the primary lever
• Education Publisher Ad Revenue Monetization: The Lesson Loop Advantage: How education publishers achieve the highest impressions-per-session correlation in the dataset and what other verticals can learn from it
• Sports Publisher Ad Revenue Optimization: Why the Sports Playbook Is Different: Why the same density tactics that work for gaming actively underperform in sports — and what to do instead
• News Publisher Ad Revenue Monetization Strategy: When More Ads Won't Save You: The audience quality story behind news monetization and why adding ad units rarely moves the revenue needle in this vertical
• Entertainment Publisher Ad Revenue: Why Volume and Geography Define Your Revenue Story: How entertainment publishers navigate the intersection of volume-first optimization and geographic demand constraints

Frequently Asked Questions

Does session duration affect ad revenue?

Session duration alone has virtually no relationship with revenue per session. The Pearson correlation between session duration and RPS is −0.03 across publisher data. What actually predicts revenue is pageviews per session (r = 0.27), a user loading multiple pages generates fresh ad inventory each time. A visitor who spends 10 minutes on a single page is worth less than one who spends 4 minutes across four pages.

What fill rate should publishers target?

Publishers above 90% fill earn 5.3x more per session than those below 40% fill. The 75–90% bracket is also strong, delivering 3.5x the RPS of the sub-40% group. One caveat: fill rate is partially determined by audience geography. Publishers with predominantly US and Western European traffic will structurally achieve higher fill than those with heavy international audiences, regardless of stack configuration. Benchmark within your geographic cohort.

How many ad units per page is optimal for publisher revenue?

More impressions per pageview consistently predicts higher revenue per session, up to 21x more at 15+ impressions per page versus under 1. The relationship isn't unlimited, though: too many ad units per page fragments request volume and reduces fill efficiency (requests per pageview correlates at −0.18 with fill rate). The sweet spot is density without fragmentation, enough competing demand to fill inventory, not so many units that each slot becomes harder to fill.

What is the floor price trap in programmatic advertising?

The floor price trap occurs when publishers set aggressive price floors that reduce fill far more than they increase CPM revenue. Within the same demand tier, publishers with high floors run at roughly half the fill rate of those with calibrated floors. Despite charging nearly 2.5x the CPM per impression, high-floor publishers generate 19% less revenue per session. Right-sizing floors to match actual demand clears more inventory and generates more RPS.

Why does ad density matter more than CPM for publisher revenue?

Impressions per pageview correlates with revenue per session at 0.59 — the strongest predictor in Playwire's publisher dataset. CPM correlates at only 0.22. More ad impressions per page means more inventory per visitor, and more inventory means more total revenue regardless of the per-impression rate. A publisher with a modest CPM but dense ad layout will consistently outperform one with a high CPM and sparse layout.

How Playwire Pulls These Levers

The findings in this dataset reflect what's observable across a network of 1,200+ publisher partners and 100 billion+ annual ad impressions. Playwire's RAMP platform is built to execute on exactly the levers that move RPS: ad layout configuration that drives impressions per pageview, yield optimization that calibrates floors to actual demand, header bidding infrastructure that maintains fill depth, and access to Amazon and other high-value demand sources that most publishers can't assemble independently.

Publishers on Playwire consistently see meaningful revenue increases within the first 90 days. That's not because the platform is doing something exotic. Getting density, depth, fill, and demand breadth all working together simultaneously is genuinely difficult to execute without the right infrastructure and the right team behind it.

The data tells you what the top 10% do differently. Playwire's job is to get you there. If you're ready to find out where your stack sits against these benchmarks, start at playwire.com.