Do Rewarded Ads Actually Make Money on Web Games?

An honest, web-first breakdown of what rewarded video really pays a browser game in 2026 — and why the number you were quoted is almost never the number you keep.

Rewarded Video · Monetization · Web & HTML5

If you build browser games, you have heard both stories. One says rewarded video is a money printer — $20, $30, even $40 eCPMs, opt-in rates north of 60%, revenue lifts that dwarf your in-app purchases. The other says web ad money is a rounding error, that real CPMs live on native mobile and the browser gets the scraps.

Both are repeating figures that were never measured on web inventory. The uncomfortable truth is that clean, web-specific rewarded data is genuinely scarce — the vast majority of the benchmark tables you have read were built on mobile app traffic and quietly reprinted as if they applied to an HTML5 game running on a portal. They do not.

So this post does something different. Instead of quoting you a number, it hands you a method for turning any number into a realistic web expectation — and answers the real question underneath the title: not “what is the eCPM,” but “how much of it do I actually keep?”

The short answer, before the caveats

Rewarded video is the highest-paying ad format available to a web game, and it is the only format players consistently choose rather than tolerate. On web and HTML5 inventory in 2026, gross rewarded eCPMs cluster roughly in these bands, before any revenue split:

Figures are directional 2026 web/HTML5 benchmark ranges (see the CPM & eCPM guide in this cluster), calibrated to browser inventory rather than mobile. Replace with your own AppLixir reporting before treating any figure as a target: [AppLixir data: $4]

Notice the spread. The gap between the top and bottom band is not a rounding difference — it is a factor of ten. Which means the single most important thing about your rewarded revenue is not the benchmark at all. It is where your players actually come from, and how much of that gross rate survives the trip to your account.

It also means the honest answer to the title question is conditional, not absolute. Rewarded video makes good money for a web game with a healthy tier-1 and tier-2 core, a clean consent posture, and a demand stack deep enough to fill its requests. It makes disappointing money for a game that is monetizing a mostly tier-3 audience through a shallow, tracking-heavy setup — and then blames the format when the mobile numbers fail to appear. Same format, opposite outcome. The difference is everything the benchmark leaves out.

The Web eCPM Waterfall: from quoted rate to take-home

Every headline eCPM you have ever read is a gross, best-case, tier-1 number. Four things happen to it on the way to your revenue. Name them, and a “$20 eCPM” stops being a promise and becomes a calculation.

1. The geo haircut

Benchmarks quote tier-1 rates because tier-1 pays most. Your audience is almost never all tier-1. A web game distributed on open portals often skews global, with a heavy tier-2/tier-3 tail. Blend a realistic traffic mix and the effective rate falls well below the headline before anything else touches it.

2. The consent and opt-in haircut

Rewarded is opt-in by definition — a player who does not click “watch” generates no impression. Opt-in rates are strong for the format but never 100%, and on the privacy-first web, a share of users also arrive without full advertising consent, which suppresses the rate advertisers will bid. Both effects shrink the pool of monetizable views.

3. The fill haircut

Not every ad request returns a paying ad. Fill rate depends on demand competition for your geos, your consent signal quality, and the depth of the demand stack your SDK can reach. An unfilled request pays exactly zero, no matter how high the eCPM on the requests that do fill.

4. The revenue-split haircut

This is the one mobile benchmarks never mention and the one that hits web hardest. If your game lives on a portal or uses a network that takes a cut, the eCPM is the gross rate — you keep a fraction of it. A US-heavy audience earning a top-band gross rate can net roughly half after the split. A global audience nets a blend weighted toward the low end.

A worked example

Take a mid-sized web game with a global audience and put a quoted tier-1 benchmark through the waterfall. The numbers below are illustrative, to show the shape of the erosion — not a promise.

Illustrative walkthrough only. Your real waterfall lives in your own reporting

The point is not that $4.20 is disappointing. The point is that $4.20 is real — it is money you can forecast, bank on, and grow — whereas the $20 was never yours to begin with. A studio that plans around take-home eCPM ships a business. A studio that plans around the sticker eCPM ships a shortfall.

The revenue formula that actually matters

eCPM is a rate, not a revenue figure. To know whether rewarded ads make money for your game specifically, you need the whole equation — and every input is measurable from your own game, not borrowed from a table:

Work it forward. Say 10,000 daily active users, each watching an average of 2 rewarded ads a day, at a take-home eCPM of $5. That is 20,000 impressions × $5 ÷ 1,000 = $100 a day, or roughly $3,000 a month, from a single ad line that most players actively enjoy. Double the impressions per user through better placement — a second, well-designed reward moment in the core loop — and you double the line without touching your rate or your traffic.

This is why “impressions per DAU” is the lever indie teams most often leave on the table. You do not need a higher eCPM to earn more; you need more willing views per player, delivered without nagging. Two of the three inputs in that formula are entirely within your design control.

Does it cannibalize your in-app purchases?

The most common reason developers hesitate on rewarded video is the fear that free rewards will suppress paid ones. The consistent finding across the industry runs the other way: rewarded video and in-app purchases behave as complementary streams, not competing ones. Players who engage with rewarded ads convert to purchases at a meaningfully higher rate than those who never do — the reward moment acts as an onboarding ramp that pulls non-payers deeper into the game, where a fraction of them eventually choose to spend.

In practice, adding a rewarded line tends to lift total ad revenue substantially while leaving purchase revenue roughly flat — the two channels sit side by side rather than trading against each other. For a hybrid web game, that makes rewarded video the lowest-risk revenue addition available: it monetizes the 90%+ of players who will never buy anything, without dampening the few who will. This is the core logic behind the hybrid model that anchors this cluster.

When web rewarded ads do NOT make money

An honest answer has to include the failure modes. Rewarded video underperforms on web in four recognizable situations — and every one is fixable once you can name it.

• The geo mix is almost entirely tier-3. If your traffic is overwhelmingly low-rate regions, even strong opt-in and fill cannot rescue the blended eCPM. The fix is audience and distribution strategy, not ad settings.
• Opt-in is quietly broken. A reward that is not worth 20–30 seconds of attention, a prompt buried where nobody sees it, or a value exchange players do not understand — any of these collapses your impressions-per-DAU input to near zero. See the willing-attention framing in the fundamentals cluster.
• Fill is thin because the demand stack is shallow. A single-source or weak-demand SDK on a global audience leaves a large share of requests unfilled. Depth of demand competition — the buyer-stack question — is what turns requests into revenue.
• A heavy tracking stack is dragging down consent. Ironically, bolting on invasive tracking to “improve targeting” can suppress consent rates and bids on the privacy-first web, netting you less than a clean, compliant setup would. On web, privacy discipline is a revenue strategy, not a tax.

Notice that three of the four failure modes are not about the ad rate at all — they are about design, distribution, and demand depth. Which is the whole argument of this post: web rewarded revenue is won or lost long before the eCPM is quoted.

Why the web is not the poor cousin — if you build for it

The mobile-versus-web comparison is real but often drawn upside down. Mobile posts higher sticker eCPMs, yes. But web carries structural advantages that the sticker rate hides: no app-store cut on ad revenue, instant play with no install friction, frictionless distribution across portals, and an audience increasingly primed to reward privacy-respecting experiences. A browser game built to be privacy-first and consent-clean can command bids that a tracking-heavy mobile clone leaves on the table.

There is a compounding effect here that is easy to miss. Because the web has no store tax on ad revenue, a dollar of rewarded eCPM you keep on web is worth more than a nominally larger dollar on mobile that has already been split with a platform. The sticker comparison flatters mobile; the take-home comparison narrows the gap far more than most developers assume. When you also weigh instant play, zero install friction, and cross-portal reach, the web case for rewarded video is stronger than its headline rate ever suggests.

This is where the tooling choice matters. A rewarded SDK built for the browser — privacy-first, TCF 2.3 and GDPR-aligned, and free of a heavy tracking stack — protects the two waterfall stages web developers most often lose to: consent and fill. That is the design behind AppLixir, which was built specifically for HTML5, WebGL, and browser-based games rather than retrofitted from a mobile SDK. Cleaner consent means a healthier bid; a demand stack tuned for web inventory means fewer empty requests. Both feed straight back into your take-home eCPM.

Frequently asked questions

Do rewarded ads really pay more than banners or interstitials on web?

Yes. Rewarded video is consistently the highest-paying format available to a web game, because it is opt-in and completion rates are very high — advertisers pay a premium for genuinely watched, consented impressions. Banners earn a fraction of the rate, and interstitials risk retention damage that rewarded video avoids.

Why are web eCPMs lower than the mobile numbers I keep seeing?

Because most published rewarded eCPM benchmarks are measured on mobile app inventory, not web. Web rates are lower at the sticker level and further reduced by portal revenue splits. The realistic comparison is take-home eCPM on web versus take-home on mobile, not sticker versus sticker.

What is a realistic take-home eCPM for a web game in 2026?

After geo blend, consent, fill, and revenue split, many web games land in a roughly $4–$8 take-home range on rewarded video, with tier-1-heavy audiences higher and global audiences lower. Treat this as directional and confirm against your own reporting.

Will rewarded ads reduce my in-app purchase revenue?

The evidence points the other way. Players who engage with rewarded video tend to convert to purchases at a higher rate, and adding a rewarded line typically lifts total ad revenue while leaving purchase revenue roughly flat. The two streams complement rather than cannibalize each other.

What is the single biggest lever on my rewarded revenue?

Willing impressions per daily active user. Two of the three inputs in the revenue formula — impressions per DAU and, indirectly, opt-in — are set by your game design, not by the ad network. More well-designed reward moments beat chasing a higher eCPM.

Keep reading in this cluster

• The full CPM & eCPM guide — how the rates are calculated and which formats pay most on web.
• Web vs. mobile demand structure — why browser inventory prices differently, and the Three-Layer Buyer Stack.
• The Rewarded Quality Index (RQI) — scoring whether your reward moments actually earn willing attention.
• The hybrid monetization playbook — fitting rewarded video alongside IAP without cannibalization.