Most of the rewarded video CPM figures circulating online describe mobile app inventory, which a browser game cannot access. This page covers what the format pays on the web specifically: the per-impression price, the formula that turns that price into monthly revenue, and the handful of inputs a studio can actually change.

How much do rewarded video ads pay on web games?

Rewarded video ads pay $4+ CPM on web game inventory through AppLixir, compared with $0.50-$2 CPM for standard web display on the same traffic. CPM means cost per mille — the price an advertiser pays per 1,000 delivered impressions — so $4+ CPM means roughly $4 per thousand completed rewarded views, before fill rate is applied.

That figure is an average across AppLixir’s web inventory, which runs 100M+ impressions per month. Individual publishers land above or below it depending on where their players are, what share of requests get filled, and how clean their consent and playback setup is. A game whose audience is concentrated in the United States, Canada, the UK, Germany or Australia will clear above a game of the same size whose players are mostly in thin-demand markets.

The second half of the answer matters more than the first: CPM is a unit price, not a revenue figure. Rewarded video is a low-volume, high-value format. You will serve a fraction of the rewarded impressions you serve banner impressions, which is why a $4+ CPM does not automatically beat a $1 CPM running continuously. The arithmetic below reconciles the two.

What is a rewarded video ad, and why does it pay more than display?

A rewarded video ad is a 15-to-30-second video a player chooses to watch in exchange for something specific inside the game: an extra life, a continue, a hint, a currency top-up, a skipped timer, a one-article paywall bypass. The defining property is opt-in. Nothing plays unless the player asks for it, and the reward is only granted on a verified completed view.

That structure is the entire reason for the price difference. A display banner may load below the fold and never enter the viewport; the advertiser pays for a chance at attention. A rewarded video is requested by the user, watched to the end, and reported as complete. Demand partners bid accordingly, and on web inventory that gap shows up as $4+ CPM for rewarded video against $0.50-$2 for display.

The trade-off is volume. Banners run for the length of the session with no player cooperation required. Rewarded video only monetizes the moments where a player wants something badly enough to spend thirty seconds on it, and there are a finite number of those moments per session. If you need the mechanics in more depth before the numbers, the developer’s guide to rewarded video ads covers how the format behaves in a browser.

Rewarded video CPM, eCPM and fill rate: the three numbers behind the payout

Three metrics determine what rewarded video pays, and they are routinely quoted interchangeably by people who should know better. CPM is the price on a filled impression. eCPM is what you actually banked per thousand requests. Fill rate is the bridge between them.

CPM

Cost per mille is the advertiser’s price for 1,000 delivered impressions on a specific piece of inventory. The winning bid on any given rewarded impression is influenced by view completion rate, player dimensions and video quality, audience quality signals, the surrounding content, the country of the viewer, and the time of year. A $4+ CPM is a blended outcome of all of those, not a rate card.

eCPM

Effective CPM is total revenue divided by total impressions, multiplied by 1,000. It absorbs unfilled requests, mixed geography and mixed device types, which makes it the only honest basis for comparing two networks. When a network quotes a CPM and a publisher reports an eCPM, those are not the same conversation. Our full breakdown of CPM and eCPM works through the calculation step by step.

Fill rate

Fill rate is the share of ad requests that return a playable ad. It is driven overwhelmingly by geography — demand for North American and Western European audiences is deep, demand in tier-three and tier-four markets is thin — and secondarily by consent and identity. A request with no valid consent signal and no resolvable identifier is hard to match to a buyer, so it either goes unfilled or clears at a discount.

This is why compliance is a revenue feature rather than a legal chore. AppLixir ships TCF 2.3 and GDPR compliance built in via Didomi, so European traffic reaches the auction with a valid consent string attached instead of being dropped or downgraded.

How much does AdMob pay for rewarded video ads, and does that apply to web?

AdMob’s published rewarded video CPMs describe mobile app inventory, and AdMob does not serve rewarded video to browsers. Neither do Unity Ads, AppLovin or ironSource. These are mobile SDK networks: they integrate into iOS and Android builds and sell inventory inside installed apps. A Unity WebGL build running in Chrome cannot request that inventory, so those figures are not a benchmark a web publisher can act on.

The practical consequence is that a lot of web studios anchor their expectations to numbers from a different market. Mobile rewarded CPMs are often quoted higher than web rewarded CPMs, which leads publishers to conclude that web monetization is broken when the real issue is that they are comparing against inventory they cannot reach. The relevant comparison for a browser game is rewarded video against the other formats that actually serve in a browser — display, interstitial, offerwall — and on that comparison rewarded wins clearly on price per impression.

Google AdSense does serve web, but it serves display and in-page video, not opt-in rewarded placements inside a game loop. It is a sensible thing to run on the page around your game; it does not monetize the moment a player dies and wants a continue. For a side-by-side of which networks serve which environment, see Unity rewarded ads and AdMob on WebGL.

AppLixir was built for the browser rather than ported into it: a JavaScript rewarded video ads SDK for HTML5 canvas games, Unity WebGL builds and ordinary web pages, with demand sourced specifically for web inventory.

How to calculate your own rewarded ads revenue

Rewarded ads revenue follows one formula: daily revenue = (DAU × rewarded views per active user × fill rate) ÷ 1,000 × CPM. Multiply by 30 for a monthly figure. Of those four inputs, the CPM is set by the market, the fill rate is mostly set by your audience’s geography and consent quality, and views per active user is a design decision you own outright.

Here is the arithmetic with assumptions stated plainly so you can substitute your own. Take a game at 20,000 DAU — four times AppLixir’s 5,000 DAU minimum. Assume 25% of daily active users take a reward once. That is 5,000 rewarded views a day, roughly 150,000 a month, and at $4 CPM about $600 a month.

Now change one input. Design two natural reward moments instead of one and lift participation to 40%. That is 16,000 views a day, roughly 480,000 a month, and about $1,920 at the same CPM. The traffic did not change. The number of reward moments and the participation rate did.

The 25% and 40% figures above are illustrative inputs, not published benchmarks — treat them as placeholders until you have measured your own. [DATA NEEDED: median rewarded participation rate across AppLixir web publishers, with time window.] Measure participation from your first week live, because it is the single number you will be optimising for the next twelve months.

Rewarded video CPM compared with other web ad formats

Rewarded video carries the highest CPM of the common web game formats and the lowest volume ceiling. The table below sets the trade-offs side by side.

Format	Typical web CPM	User action	Volume ceiling	Session risk
Rewarded video	$4+	Opts in, expects a reward	Low — capped by reward moments	Low; the player asked for it
Display banner	$0.50-$2	None; passive	High — runs for the whole session	Low, but subject to banner blindness
Interstitial video	Above banner, below rewarded	Forced view, skippable or timed	Medium — frequency-capped	Higher; interrupts play
Offerwall	Varies widely by offer	Completes a multi-step task	Very low participation	Low, but pulls players off-site

The useful comparison is not rewarded against everything else — it is rewarded plus everything else. Display monetizes the passive surface area of the page. Rewarded video monetizes the specific friction points where a player will trade attention for progress. The two draw on different demand and running both is standard practice. Rewarded video versus interstitial ads covers where the middle format fits.

Distribution platforms such as Poki and CrazyGames sit in a different category entirely. They bring players to your game and handle monetization on their own domains. AppLixir runs on the traffic you own — your site, your embeds, your direct players — and sits alongside portal distribution rather than replacing it.

Geography and seasonality: why identical DAU pays differently

Geography moves revenue per user more than any other single variable in rewarded video. It sets two things simultaneously: what an advertiser will bid, and how many campaigns are available to serve your players in any given hour. Thin demand produces low fill, and unfilled requests drag eCPM down even when the CPM on filled impressions looks healthy.

Two web games with identical DAU can differ several-fold in monthly revenue purely on traffic mix. An IO game with a large Southeast Asian and Latin American player base will not earn what a puzzle game with a US and UK base earns, and placement tuning does not close that gap entirely. The response is to stop treating DAU as one number: break it out by country tier and look at revenue per thousand sessions in each tier separately.

Seasonality is the second pattern worth planning around. Q4 is the peak — holiday budgets, Black Friday and Cyber Monday concentrate advertiser spend into a few weeks, raising both clearing prices and fill on traffic you already have. Q1 is the trough, as budgets reset and January and February campaigns run thin. The pattern has held across multiple cycles. If your first month live is January, do not extrapolate it; if it is November, do not extrapolate that either. Plan against a rolling twelve-month average and ship any product work that adds reward inventory before Q4 rather than during it.

When rewarded video is not worth the integration time

Rewarded video is the right format for most web games with a progression loop, and the wrong one in four identifiable situations. Saying so up front saves an afternoon.

• Below 5,000 DAU. AppLixir’s publisher threshold is 5,000 daily active users. Below that, the absolute revenue is small enough that retention and acquisition work returns more than monetization work. Get the audience first.
• No natural friction moment. If your game has no failure state, no scarce resource, no timer and no locked content, there is nothing to reward. Bolting an arbitrary “watch an ad for coins” button onto a game with no coin economy produces low participation and annoyed players.
• A fragile paid economy. If a meaningful share of revenue comes from in-app purchases of exactly the thing the reward grants, you will cannibalize it. The fix is calibration, not abstention — reward a smaller quantity, or a different currency — but it needs deliberate design.
• Sessions under about a minute with no return visits. Very short, one-and-done sessions rarely reach a second reward moment, and the revenue per user stays low regardless of CPM.

Everywhere else — and that covers most IO games, HTML5 puzzle and arcade titles, Unity WebGL builds and content sites with gated material — rewarded video adds revenue on traffic that display alone underprices.

The levers that actually move rewarded ads revenue

Once a rewarded video integration is live, the CPM is largely the market’s business. These inputs are yours, listed roughly in order of impact.

• Reward moments, not ad slots. Place the offer exactly where the player already feels friction: death, out of hints, a locked cosmetic, a slow timer. A reward the player wants converts; one they do not want is a wasted request and a small goodwill cost.
• Frequency caps. Over-placement shortens sessions, and shorter sessions mean fewer impressions in total. Larger studios typically cap rewarded offers per session and per hour, then test upward rather than starting uncapped.
• Reward value calibration. Too small and participation collapses. Too large and you devalue your own economy and undercut in-app purchases.
• Consent quality. A correctly implemented CMP keeps European traffic addressable. A broken or missing one silently removes a share of your fill without producing an error anybody notices.
• Abuse controls. Reward farming, repeat views from a single session and automated traffic degrade the audience quality signal buyers price against. Preventing rewarded ad fraud and reward abuse covers the server-side checks worth having.
• Player size and load behaviour. Video dimensions and clean playback affect completion rate, and completion rate feeds back into what buyers bid next time.

Rewarded video also pairs with in-app purchases rather than competing with them. Players who will never pay still have attention worth something, and players who do pay often use rewarded ads as a low-commitment on-ramp into the paid economy. The eCPM optimization guide for web games goes deeper on the tuning side.

FAQ

What is a good rewarded video CPM for a web game?

On web inventory, $4+ CPM is the figure AppLixir delivers across its 100M+ monthly impressions, and that is a reasonable bar for a game with a mixed international audience. Games weighted toward the United States, Canada, the UK, Germany and Australia should expect to clear above that blended average; games weighted toward thin-demand markets will sit below it. Judge yourself on eCPM rather than CPM, since eCPM includes the requests that went unfilled.

Do Unity Ads or AdMob work for WebGL and browser games?

No. Unity Ads, AdMob, AppLovin and ironSource are mobile SDK networks that serve inventory inside installed iOS and Android apps. A Unity WebGL build running in a browser cannot request their rewarded video inventory, which is why their published CPMs are not a usable benchmark for web publishers. AppLixir serves rewarded video to browsers through a JavaScript SDK built for HTML5 and WebGL.

Are rewarded video ads GDPR compliant for players in Europe?

Rewarded video can be fully GDPR compliant provided a valid consent signal reaches the ad auction. AppLixir includes TCF 2.3 and GDPR compliance built in via Didomi, so European traffic arrives with a consent string attached rather than being dropped or downgraded by demand partners. For enterprise publishers this is a revenue question as much as a legal one, because unconsented and unidentifiable requests are the ones most likely to go unfilled.

How do large studios add rewarded ads without hurting retention?

They treat rewarded video as a player-facing feature rather than an ad slot: the offer appears only at genuine friction points, it is always optional, and it is frequency-capped per session from day one. Retention cohorts are monitored against a holdout group that sees no rewarded offers, so any session-length impact shows up before it is rolled out widely. Because rewarded video is opt-in, well-placed implementations usually show neutral-to-positive retention — players who take a continue stay in the session longer than players who do not.