What Are 1,000 Ad Impressions Worth? 2026 Mobile Game eCPM Benchmarks by Format, Region, and OS, and the Q4 Price Window

A mobile game's ad revenue comes down to impressions multiplied by eCPM. Impressions are something the studio controls; eCPM depends on what advertisers are willing to pay for each impression, which means it moves outside the studio. That is why the same game, with the same placements, sees revenue swing from month to month, and why you need reference numbers to tell whether a swing comes from your game or from the market.

Finding those reference numbers is harder than it sounds. Gather the public eCPM benchmarks available as of September 2026 and the same US rewarded video placement ranges from about $13 to $30 depending on the source, more than a 2x gap. The market structure keeps shifting as well. According to the benchmark Tenjin published in August 2026, 55% of game ad revenue in Q2 2026 came from Android and 45% from iOS, and the US alone generated 58% of iOS ad revenue and 32% of Android ad revenue. Which country and which OS you use as your baseline changes the conclusion you reach.

Completion, retention, and ROAS benchmarks for rewarded ads are covered in Rewarded Ad Benchmarks for 2026: Know the Numbers, Know How to Read Them, and how ad revenue feeds a daily metric is covered in Revenue = DAU × ARPDAU: The Daily Metric That Separates Real Growth From Install Spikes. This post focuses on eCPM itself. It lays out figures by format, region, and OS with their sources, explains why sources disagree, walks through the variables that move eCPM, and looks at how the Q4 price window that opens in October actually behaves.

What Is eCPM, and What Does It Mean for Publishers and Advertisers?

eCPM (effective cost per mille) is the revenue actually generated per 1,000 ad impressions. For a publisher it is the value of a placement; for an advertiser it is an input that sets the cost of each install.

The formula is simple. As Google AdMob Help defines it, eCPM is total ad earnings divided by impressions, multiplied by 1,000. Because the result is normalized to 1,000 impressions regardless of whether the buy was CPM, CPC, or CPI, it lets you compare ad networks with different pricing models on the same scale.

For a publisher, eCPM is one of the two variables that make up ad ARPDAU. Ad ARPDAU equals impressions per DAU per day, multiplied by eCPM, divided by 1,000. If players watch an average of three rewarded videos a day at a $12 eCPM, rewarded ad ARPDAU is $0.036. There are only two ways to grow that number: show more ads or raise eCPM. Showing more ads can cost you retention, so understanding eCPM first is the cheaper path.

For an advertiser, the same number is a cost. Divide what you pay per 1,000 impressions (CPM) by installs per 1,000 impressions (IPM) and you get cost per install. At a $20 CPM and an IPM of 5, CPI is $4. If creative and targeting stay the same and IPM holds, a 20% rise in CPM becomes a 20% rise in CPI. That is why the Q4 price rise matters to UA teams as much as to monetization teams.

What Is the eCPM for Each Ad Format?

Rewarded video carries the highest eCPM, followed by interstitials, native, and banners. Offerwalls are calculated differently and cannot be compared on the same line.

Almost every source agrees on the order. Rewarded video is chosen by the player in exchange for a reward, so completion is high and advertisers pay the most for it. Interstitials give the player no choice, but the publisher controls exposure, so volume is large. Banners have the lowest price but stay on screen longest. Where sources differ is the absolute value. The table below puts sources with different methods and time periods side by side, so it is safer to read it for the size of the gaps between formats than to compare cell by cell.

The Appodeal figures are read from the report's bar charts, so they are not precise to the cent. Some industry blogs quote native ranges such as $2–$6 in tier 1 markets, but they do not disclose the underlying data, so they are left out of the table. For games that rely heavily on rewarded video, native matters less for its price than for how it is integrated. RevenueFlex notes that native ads built naturally into the game screen can approach interstitial-level eCPMs while doing less damage to the experience.

Interstitial prices split sharply by creative type. In the ranges AdReact compiled in May 2026, video interstitials in the US and Canada ran $12–$25 while static and playable interstitials ran $6–$12, roughly a 2x difference. For what each impression costs you in retention, see The Real Price of an Interstitial: What One Full-Screen Ad Costs You in Retention.

Offerwalls need to be read separately. Commonly cited offerwall eCPMs run into the hundreds of dollars and look dozens of times higher than other formats, but the denominator is the number of times the offerwall screen was opened rather than one ad impression, and the numerator is what advertisers pay after players complete multi-step missions. Because the calculation is different, comparing it directly with rewarded video overstates the offerwall's value. Offerwall revenue is better read as participating users multiplied by revenue per participant, and the figures confirmed by neutral sources are collected in How Much Do Offerwalls Really Lift Retention and Revenue? 2026 Benchmarks and the Conditions That Cannibalize IAP.

How Much Does eCPM Vary by Region and OS?

Regional gaps in eCPM are as large as the gaps between formats: the same rewarded video can be worth 4 to 5 times more in North America than in Latin America. The gap between iOS and Android persists but is narrowing.

The most recent public table that breaks out formats at the country level is the Q2 2024 dataset from mobile measurement company Tenjin, produced with ad monetization platform CAS. It covers game apps, excluding kids apps.

South Korea stands out. It made the top three for rewarded on both iOS and Android and for interstitials on iOS, and on iOS it sat right behind the US in both formats. Korean players' ad value is at the top tier just after the US, which suggests that globally launched games can easily underestimate the ad revenue potential of Korean traffic.

At the regional level, Appodeal's report covering more than 100,000 apps and over 200 billion ad views from October to December 2024 is a useful reference. The regional averages read from its charts are below.

The OS gap looks different by region. In North America, Europe, and the Middle East, iOS is clearly higher, but in APAC, Android rewarded video came out ahead of iOS. In the Tenjin table, too, US rewarded was higher on Android ($30.25) than on iOS ($24.39). The assumption that iOS is always more expensive often fails once you split by region and format.

One of the variables behind iOS ad value is the App Tracking Transparency (ATT) opt-in rate. Without an advertising identifier, advertisers cannot target players precisely, and bids fall accordingly. According to Adjust, the global ATT opt-in rate reached 38% in Q1 2026, up from 35% a year earlier, and gaming had the highest rate of any category at 39%. RevenueFlex also observes that the eCPM gap between iOS and Android has narrowed compared with previous years.

The most recent direction comes from quarterly comparisons by ad monetization company Bidlogic. Across the US, UK, Germany, India, and Japan, iOS rose at least 8% in every format between Q1 and Q2 2026, with interstitials up 10.67% to 16.72%. Android rewarded video and interstitials mostly rose a more modest 2.55% to 13.94%, and Android rewarded video in India actually fell 5.18%.

Why Do eCPM Figures Differ Several-Fold Between Sources?

The biggest reason eCPM figures differ several-fold between sources is sample composition: whether the data covers only games, how small the apps included are, whether it reports an average or a peak, and what period it covers.

Put the major sources side by side for the same US market and the differences are clear.

US interstitials alone range from a low of $5 to $23, more than a fourfold spread. The reasons fall into four groups.

First, whether the data is games only. The Tenjin table covers game apps only, while MonetizeMore and Playwire draw on publisher data that includes non-game apps. Game traffic, especially from games with paying players, is something other game advertisers are willing to pay a premium for, so it prices higher than general app traffic.

Second, whether the long tail is included. Aggregate 100,000 apps as Appodeal does and you pull in a large number of small apps and low-price countries, which drags the average down. In the same report, the overall monthly average rewarded video eCPM was around $3 on Android and around $4 on iOS, about a quarter of the US figures.

Third, average versus peak. MonetizeMore's $8.57 is a monthly peak, not an annual average, and the "$40+ rewarded" figures some blogs quote are the top end for specific placements. Before citing a benchmark, check whether it is an average, a median, or a peak.

Fourth, timing. The same placement can move more than 30% between December and January (see the Q4 section below), so putting Q2 data and Q4 data in one table makes seasonal effects look like differences between sources.

The practical conclusion is simple. Use external benchmarks only to judge whether your game is far below the market, and set targets from your own historical data by country and format. RevenueFlex makes the same point: treat benchmarks as a check on whether you are significantly underperforming, not as targets.

What Moves eCPM?

eCPM is set by how many advertisers want an impression and how badly they want it. The variables a publisher can move directly are competition, price floors, placement quality, and ad frequency.

Google AdMob Help lists market and platform factors, ad category blocking, and price floor changes as causes of eCPM fluctuation, and recommends increasing bidding pressure in mediation and reducing category blocking. The more categories you block, the fewer advertisers can bid, and the lower the auction clears. Wanting to block ads for competing games is understandable, but it should be a decision made knowing that you are giving up price.

A price floor trades eCPM for fill rate. Raise the floor and eCPM goes up, but impressions with no bid at that price go unfilled. If the dashboard shows eCPM rising while total ad revenue falls, suspect this. The yardstick should be revenue per 1,000 ad opportunities, or ad ARPDAU, not eCPM.

Frequency affects price too. Show the same player more ads and the advertiser demand for that player gets used up, so the marginal price falls. Bidlogic attributed eCPM declines in some markets in Q4 2025 to a reduced revenue share for top-tier networks alongside a rise in ad impressions. When supply increases, price falls.

Finally, there is the value of the player. Players with purchase history or clear genre preferences are players advertisers can expect post-install revenue from, so bids are higher. In that sense ad revenue and in-app purchases are less competitors than two ways of capturing the same player value, and how to balance the two is covered in Balancing IAP and Rewarded Ads Monetization: How Two Revenue Models Work Together.

How Much Does eCPM Actually Rise in Q4, and What Should Studios Prepare?

Advertiser budgets concentrate in Q4, and eCPM tends to run 20% to 40% above Q1. But the rise is uneven across regions and formats, and the more dependable pattern is the January drop.

The general seasonal pattern looks like this. AdReact puts Q4 eCPMs typically 20% to 40% above Q1 baselines. Felix Braberg, who writes a mobile ad revenue newsletter, put the peak-to-trough swing at 30% to 40% on US iOS and 20% to 35% on Android and most tier 2 countries in an August 2026 post, with the example of a US iOS rewarded placement clearing $18 in December and $11 to $12.50 by mid-January. In January, spending stops until advertisers' new annual budgets are approved, and August is a low point as European holidays pull demand out.

Prices on the buying side follow the same curve. In Superads' US data covering more than $3 billion in Facebook ad spend, the median CPM over the latest 13 months was $23.42, peaking at $27.47 in November 2025. The largest single monthly move was the $4.62 jump from October to November, and the low was $20.24 in July 2026. These are not game-app figures specifically, but they show how commerce advertisers heading into Black Friday compete for the same inventory and push up impression prices for game UA as well.

Look at quarterly data, however, and not every placement rises.

The Bidlogic data shows the Q4 rise concentrating in specific formats in tier 1 countries, while other markets see double-digit declines in the same quarter. In the Appodeal report the text and charts do not match; the more natural reading is that in an overall average mixing long-tail apps and low-price countries, the Q4 effect barely shows. In other words, "eCPM rises in Q4" holds for certain placements in markets like the US and UK, but it is not a rule that applies automatically to every game.

Based on this data, preparation splits between the monetization side and the UA side.

On the monetization side, first, do not read a Q4 revenue increase as the result of better ad design. Compare with the same month last year, not the previous quarter, to filter out seasonality. Second, if you plan major changes to your mediation setup or price floors, finish them within October. Change them in November or December and the effect of the change blends with the seasonal effect, leaving you unable to identify the cause. Third, do not project January revenue from Q4 eCPM. Assuming a drop of around 30%, as in the example above, is the safer default. Fourth, be careful about raising ad frequency just because prices are up. Q4 prices come back down in January, but players lost to frequency do not come back.

On the UA side, there is room to shift budget around the November CPM peak: finish creative testing in early to mid October, set clear bid caps in November, and load volume into late December and January when commerce demand pulls back. Channels bought on CPM absorb seasonal swings in full, while channels bought per install or per in-game action pass less of that swing into the unit price.

Buying Outcomes Instead of Impressions: The Playio Perspective

In periods when impression prices swing as much as they do in Q4, channels that price by installs or in-game actions rather than impressions help stabilize cost forecasts.

With impressions bought on CPM, cost per install rises in step with CPM even if IPM holds. In a buying model that prices installs or specific in-game actions, the unit the advertiser pays for is the outcome, so seasonal swings in the auction market are not passed directly into the advertiser's cost. On the ad revenue side, as covered above, traffic made up of players with purchase history and clear preferences is traffic advertisers value more.

Playio runs an Android-based gamer community of five million players, analyzing genre preferences, play history, and in-game behavior data to connect games with players whose tastes match. Pricing is CPI by default, and CPE based on reaching a playtime threshold or completing a specific in-game action is also supported. Even in November, when impression prices are highest, what the advertiser is buying is installs and post-install play rather than impressions, which makes it a useful counterweight to the volatility of CPM channels when planning Q4 budgets.

You can find more details here. (https://playioadsen.oopy.io/bizdeck)

Key Takeaways

As of September 2026, mobile game eCPM runs highest for rewarded video, followed by interstitials, native, and banners, while offerwalls are calculated differently and cannot be compared on the same line. Regional gaps are as large as format gaps, with the same rewarded video worth 4 to 5 times more in North America than in Latin America, and South Korea ranks in the top three for rewarded on both OSes and second only to the US for iOS interstitials. The iOS-Android gap persists but is narrowing, and in some regions and formats Android comes out ahead. Public benchmarks differ by as much as 4x between sources because of whether they cover games only, whether they include the long tail, whether they report averages or peaks, and when they were measured, so use external figures only to check whether you are far behind, and set targets from your own data by country and format. In Q4, some placements in tier 1 markets rise on the order of 20% to 40%, but the rise does not arrive evenly across markets and formats, and the January drop is the more dependable pattern. Finish structural changes within October, compare against the same month last year, and budget January conservatively: that is the baseline preparation for this Q4.

For inquiries about Playio's advertising solutions, reach out at: [email protected]

Sources

• Tenjin, Ad Monetization in Mobile Games – Benchmark Report 2026, updated 2026-08-13 (Q2 2026 ad revenue Android 55% / iOS 45%; US share 58% of iOS and 32% of Android): https://tenjin.com/blog/ad-mon-gaming-2026/

• Tenjin × CAS, Ad Monetization in Mobile Games in Q2'24, summary by GameDev Reports (top three countries by format and OS): https://gamedevreports.substack.com/p/tenjin-ad-monetization-in-mobile-00b

• Appodeal, The Latest eCPM Report 2025 (October–December 2024; 100,000+ apps, 70+ ad networks, 200B+ ad views. Regional, country, and monthly figures are read from the report's charts; the text's "December highest" statement does not match its monthly charts): https://appodeal.com/wp-content/uploads/2025/03/Appodeal-The-Latest-eCPM-Report-2025.pdf

• Playwire, AdMob eCPM Benchmarks: What Publishers Should Expect, 2025-09-17 (tier 1 and global ranges by format; methodology undisclosed): https://www.playwire.com/blog/admob-ecpm-benchmarks-what-publishers-should-expect

• MonetizeMore, 2026 eCPM Insights you Missed Out On!, 2026-06-12 (monthly 2024 data from 500+ publishers; US interstitial peak $8.57 in July; US app open peak $10.51 in March): https://www.monetizemore.com/blog/ecpm-insights/

• RevenueFlex, App Ad Revenue Benchmarks 2026, 2026-03-22 (native can approach interstitial levels; iOS-Android gap narrowing; benchmarks are not targets): https://revenueflex.com/blog/app-ad-revenue-benchmarks-2026/

• AdReact, Interstitial Ad Best Practices for Mobile Games, 2026-05-08 (US and Canada video interstitial $12–$25, static $6–$12; Q4 20–40% above Q1): https://adreact.com/blog/interstitial-ad-best-practices-mobile-games/

• Bidlogic (optAd360), What happened to mobile app eCPMs in Q4 2025?, January 2026 (Q3 to Q4 2025 changes by format and OS across five countries; declines attributed to reduced top-tier network share and more impressions): https://bidlogic.io/2026/01/30/what-happened-to-mobile-app-ecpms-in-q4-2025/

• Bidlogic (optAd360), Q2 2026 eCPM growth: Interstitial, Rewarded Video and Banner trends, 2026-07-31 (Q1 to Q2 2026: iOS all formats 8%+, iOS interstitials 10.67–16.72%, Android rewarded and interstitial 2.55–13.94%, India Android rewarded -5.18%): https://bidlogic.io/2026/07/31/q2-2026-ecpm-growth-interstitial-rewarded-video-and-banner-trends/

• Felix Braberg, Your eCPMs Didn't Break; They Went on Holiday, 2026-08-26 (peak-to-trough 30–40% US iOS, 20–35% Android and tier 2; US iOS rewarded $18 in December to $11–$12.50 by mid-January; practitioner observation, dataset undisclosed): https://felixbraberg.substack.com/p/your-ecpms-didnt-break-they-went

• Superads, Facebook Ads CPM Benchmarks in United States ($3B+ ad spend; 13-month median $23.42; November 2025 peak $27.47; October to November +$4.62; July 2026 low $20.24): https://www.superads.ai/facebook-ads-costs/cpm-cost-per-mille/united-states

• Google AdMob Help, Understand eCPM fluctuation (market and platform, category blocking, price floors, bidding pressure): https://support.google.com/admob/answer/15337570?hl=en

• Mobile Marketing Reads, summary of Adjust Mobile App Trends 2026 (Q1 2026 ATT opt-in 38%, up from 35%; gaming 39%): https://mobilemarketingreads.com/adjust-report-app-installs-up-10-in-2025-as-casual-gaming-and-att-opt-ins-climb/

• Ad ARPDAU and CPI examples (3 rewarded views × $12; $20 CPM ÷ IPM of 5): illustrative calculations