Rewarded video is the only ad format in games that the user actively asks for. The player presses a button, agrees to the trade, and gets something they wanted. That single opt-in step is why the format out-earns forced units, why it does less damage to retention than interruptive advertising, and why it became the default ad layer in free-to-play. What follows is how rewarded video ads work mechanically, the four video ad types developers confuse with each other, what the format pays on browser inventory specifically, and the placements where it is the wrong tool.

What are rewarded video ads?

A rewarded video ad is an opt-in video advertisement that a user chooses to watch in exchange for a defined in-product benefit, delivered only after the video completes. The player taps a button reading something like “Watch an ad for 100 coins”, a full-screen video plays, and the game grants the reward once completion is confirmed. Nothing is forced, nothing interrupts, and the price is stated before the impression is served.

Rewarded video is a value-exchange format. The user is not an audience being sold to; they are a party to a transaction they initiated. Three properties follow from that opt-in step, and all three are what advertisers are actually paying for:

• Consent. The user requested the ad. No impression is served to someone who did not ask for one, which is why completion rates on rewarded inventory sit far above forced formats.
• Attention. The reward depends on the view finishing, so the user has a direct reason to stay in the player rather than tab away.
• Timing. The ad fires at a moment the user chose, not at a moment the ad server chose. That is why rewarded video does less retention damage than an interstitial dropped into a session mid-flow.

The commercial logic behind the format is blunt. The overwhelming majority of web users will never pay cash for a game session or an article, and a $4.99/month paywall converts a fraction of a percent of them. Rewarded video turns the other group — the people who were never going to open a wallet — into revenue at a price they will pay, which is attention. Game studios worked this out first; content publishers monetizing with rewarded video are applying the same mechanic to gated articles, downloads and premium features.

How a rewarded video ad works, step by step

A rewarded video ad runs as a six-step loop between the game, the ad provider and the advertiser. Understanding the loop matters because the reward must hang off the completion signal, not off the button press.

1. Offer. The game surfaces a clearly priced trade at a moment the player wants something — a fail state, an empty currency balance, a locked level.
2. Opt-in. The player presses the button. Until that happens, no ad request exists.
3. Consent and request. Consent status is resolved, then the ad request goes out and demand sources bid on the impression.
4. Playback. The creative renders in the ad container inside the page. Game audio ducks, game state pauses.
5. Completion signal. The provider confirms the view finished and returns a status the game can act on.
6. Reward. The game grants the currency or unlock, and returns the player to the exact state they left.

Step five is where most weak implementations fail. If a game grants the reward on any status other than a confirmed completion, it is paying out currency for impressions nobody watched, and in extreme cases that is enough to devalue a virtual economy. Reward integrity is a design requirement, not a nice-to-have — see how to prevent rewarded ad fraud and reward abuse for the attack patterns that matter on web.

Types of rewarded video ads and how they differ from other video formats

There are four types of in-game video advertising, and only one of them is genuinely rewarded video. Developers searching for “types of rewarded video ads” are usually trying to separate these four, because networks and mediation dashboards use the labels loosely.

Opt-in rewarded video

Player-triggered, tied to a stated reward, granted on completion. Engagement is high because the player treats the ad as a gameplay feature rather than an interruption. This is the format that carries the strongest CPMs and the mildest retention cost, and it is the one AppLixir serves on web inventory.

Forced (non-skippable) video

The user must watch the whole creative with no skip control. Completion is guaranteed, which lifts per-impression value, but the impression was never requested. Placed at a natural break — between levels, on a session boundary — it is survivable. Placed frequently or mid-run, it reliably costs retention.

Skippable video

The user can dismiss the ad after an initial window. User experience is better than forced video because the player keeps control, but revenue per impression is lower since only full views carry full value.

Involuntary auto-play video

Video that starts without the user doing anything — auto-play units in a sidebar, banners that expand into video. These generate impressions cheaply and annoy reliably. They are not rewarded video and should not be benchmarked against it.

Format	User opts in	Reward granted	Usual placement	Retention risk
Opt-in rewarded video	Yes	Yes, on confirmed completion	Fail state, currency wall, gated unlock	Low
Forced / non-skippable	No	No	Level transition, session start	High if frequent
Skippable video	Partly — can dismiss	No	Level transition, pause screen	Medium
Involuntary auto-play	No	No	Page furniture around the game	High

The practical comparison most studios actually need is rewarded against interstitial, since those are the two units competing for the same break points in a session. That trade-off is worked through in rewarded vs interstitial ads.

What do rewarded video ads pay on web inventory?

AppLixir delivers $4+ CPM on web rewarded video, measured across 100M+ monthly impressions, against roughly $0.50–$2 CPM for typical web display. CPM is cost per thousand impressions — the price an advertiser pays for a thousand served views — so the gap compounds directly with session volume.

Two things drive the spread. First, a rewarded impression is a completed, opted-in view, which is worth more to a brand than a viewable banner. Second, the number of impressions per user is governed by engagement rather than page furniture: a player who hits three reward beats in a session generates three high-value impressions, where a display slot generates one low-value impression per page load regardless of what the user did.

Actual revenue is CPM multiplied by impressions, and impressions are a function of DAU, reward beats per session and opt-in rate — not of traffic alone. AppLixir’s minimum publisher threshold is 5,000 DAU, because below that the impression volume is too thin for programmatic demand to clear at a meaningful rate. For the modelling, including how opt-in rate and reward frequency move the total, see how much rewarded video ads pay on web games.

Do Unity Ads, AdMob and AppLovin serve rewarded video on the web?

No. As of 2026, Unity Ads, Google AdMob, AppLovin and ironSource are mobile app networks. Their rewarded video products are built around native iOS and Android SDKs and app-store identity, and they do not serve browser inventory. A browser game cannot call them, whatever the game was built in.

This catches out a specific group of developers every year: studios shipping a WebGL build of a game whose mobile version already monetizes through a mobile SDK, who assume the same account will cover the web build. It does not. The demand, the measurement and the delivery path are all app-side. A web game needs a rewarded video provider that is web-native — one where the player, the ad container and the reward callback all live in the page. The specifics of what does and does not carry across are covered in Unity rewarded ads and AdMob rewarded ads on WebGL games.

Distribution platforms are a different question entirely. Portals such as Poki and CrazyGames are complementary to a direct rewarded video provider, not alternatives to one. They bring audience and often monetize the traffic they send. AppLixir sits alongside that — on a studio’s own domain, its own portal, its own embedded builds — covering the sessions a portal deal does not. Most studios past 5,000 DAU are running both, and the two revenue lines add rather than cannibalize.

Where rewarded video works, and where it does not

Rewarded video pays when it sits on a real reward beat — a moment where the player already wants something the game is withholding. Bolted onto a button nobody has a reason to press, it earns close to nothing, no matter what the CPM says.

Placements that consistently convert:

• Continue after a fail. The player just lost a run they were invested in. Demand for the reward is at its peak and the interruption cost is zero, because the session had already stopped.
• Currency top-up at the wall. Offer the ad at the point the player is short, not in a menu they never open.
• Doubling an earned reward. End-of-level “double your coins” converts well because the player has already seen what they won.
• Timed unlocks and energy refills. An ad that collapses a wait is a clean substitute for the micro-transaction most players will never make.
• Daily bonus multipliers. Cheap to build, and they create a habitual opt-in at session start.

Placements to avoid: anything mid-run that pauses active play; rewards so generous they undercut your IAP tiers; rewards so thin the player opts in once and never again; and stacking rewarded offers back-to-back so the loop becomes an ad-watching game. Rewarded video is also a poor fit for a title with under 5,000 DAU or with sessions too short to produce a genuine reward beat — in that case the honest answer is that the monetization problem is a retention problem first.

Large studios usually run this as a measured rollout: ship the unit on one reward beat, watch day-1 and day-7 retention against a holdout, then add beats one at a time. That is also how to keep rewarded ads and IAP working together rather than competing, since the players who watch ads and the players who pay are mostly different people.

Rewarded video beyond games: content and utility publishers

Rewarded video is not only a games format. Any product with something worth gating can price that thing in attention instead of cash: watch a video ad to read a gated article, to unlock a premium calculator, to remove a download queue, to extend a free tier for a day.

The mechanic works for the same reason it works in games. A subscription prompt asks a first-time visitor for a payment method and a recurring commitment; a rewarded offer asks for one completed view. Readers who would bounce off a paywall convert on the trade, and the publisher earns on inventory that would otherwise have produced a bounce. The requirements are identical to the game case: a clearly stated price, a reward granted only on confirmed completion, and the user returned to exactly where they were.

What separates a web rewarded video setup that earns from one that does not

Two publishers with the same DAU can see materially different rewarded revenue, and the difference usually sits in four places that have nothing to do with the creative.

Competition on every impression

A single buyer setting the price programmatically pushes clearing prices down. Header bidding lets multiple exchanges bid on the same rewarded impression simultaneously, so the price is set by competition rather than by one buyer’s floor. That auction pressure is the mechanical reason web rewarded CPMs hold above display rather than drifting toward it.

Playback that does not fail

Video creatives arrive in inconsistent shapes — VAST and VPAID variants, differing wrappers, badly built ads. Every creative that errors out is a filled impression that pays nothing and a player who opted in and got no reward, which is worse than not offering the ad. A player-side engine that handles malformed creatives and plays them through to completion converts fill into paid impressions instead of errors.

Reward data that cannot be forged

Reward callbacks are an obvious target. If a player can replay or fabricate a completion signal from the console, they can mint currency at the developer’s expense, and a virtual economy that gets hacked at scale stops being worth anything. Protecting the callback path with encryption and validation keeps granted rewards tied to real completed views.

Consent handled before the impression

For European traffic, consent is a revenue question as well as a legal one: buyers pay less, or not at all, for impressions without a valid consent signal. AppLixir has TCF 2.4 and GDPR support built in via Didomi, so consent is collected and passed before the impression is served, rather than being bolted on by the publisher.

Adding rewarded video to a web game

AppLixir integrates through an HTML5/JavaScript SDK that runs in the browser. The game creates a container element, calls the player when the user opts in, and grants the reward only on a completed status.

initializeAndOpenPlayer({
  apiKey: "YOUR_API_KEY",
  injectionElementId: "applixir-ad-container",
  adStatusCallbackFn: (status) => {
    if (status.type === "complete") grantReward(); // only a completed ad earns the reward
  },
});

Everything else — setup, configuration and the full status handling — is in the AppLixir documentation. For a worked example inside a game loop, including where to put the opt-in button, see the guide to adding rewarded video ads to an HTML5 game.

FAQ

What is rewarded video?

Rewarded video is an advertising format in which a user voluntarily watches a video ad in exchange for a stated benefit, such as in-game currency, an extra life or access to gated content. The reward is granted only when the ad server confirms the view was completed. It is the only common ad format the user initiates themselves, which is why it carries higher completion rates and higher CPMs than formats the user did not ask for.

What are the types of rewarded video ads?

In-game video splits into four types: opt-in rewarded, forced non-skippable, skippable, and involuntary auto-play. Only the opt-in type is true rewarded video, where the player triggers the ad and receives a defined reward on completion. The other three are served without a request from the user and carry no reward, which is why they are weaker on both revenue per impression and retention.

How much do rewarded video ads pay on web games?

AppLixir delivers $4+ CPM on web rewarded video across 100M+ monthly impressions, compared with roughly $0.50–$2 CPM for typical web display. Total revenue depends on impressions as well as CPM, so DAU, the number of reward beats per session and the opt-in rate all move the figure. AppLixir’s minimum publisher threshold is 5,000 DAU.

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

No. Unity Ads, AdMob, AppLovin and ironSource are mobile app networks and do not serve browser inventory, so a WebGL or HTML5 build running on the open web cannot call them for rewarded video. Browser titles need a web-native rewarded video provider such as AppLixir. Distribution platforms like Poki and CrazyGames remain complementary — they bring audience, and a direct rewarded integration covers the sessions on your own domain.

Are rewarded video ads GDPR compliant for players in Europe?

They can be, provided valid consent is collected before the ad request is made. AppLixir includes TCF 2.4 and GDPR support via Didomi, so the consent signal is resolved and passed with the impression rather than being left for the publisher to assemble. This matters commercially as well as legally, since impressions without a valid consent string attract less demand.