A Free Games Feed Costs You Two-Thirds of Net and Your App Rights

HTML5 game distribution networks hand you the catalogue free and keep two-thirds of net ad revenue, and their publisher terms rule out apps and subscriptions. That trade is reasonable for a side project and quietly fatal for a product.

Most people shopping for games start with the wrong question. They ask how many titles they can get and what the titles cost. The number that decides the business is who the counterparty is — because the counterparty type, not the price, determines what you are allowed to do with the games, who owns the ad stack, and whether you end up with an asset or a plugin.

There are four routes into a catalogue. They are not variations on one deal. They are four different businesses that happen to share the word "licensing," and the published terms of each say so plainly if you read them.

📦 Four Counterparties Wearing One Word

The distribution network. You embed a feed. The games cost nothing up front, the network's ad code ships inside every title, and you take a percentage of what that ad code earns. GameDistribution, GamePix and GameMonetize all work this way. GameDistribution's own about page puts its catalogue at more than 19,000 games across upwards of 4,800 publishing portals — a genuinely enormous supply, available in an afternoon.

The catalogue licensor. You pay for a licence, you receive the game files, and you run your own monetization. The revenue split conversation disappears entirely because there is no revenue to split — you paid instead. This is where Forestry Games sits, and where most white-label and telecom deals sit.

The studio, direct. One game or a handful, negotiated title by title, sometimes with source code. Slow, expensive per unit, and the only route that gets you something nobody else has.

The code marketplace. CodeCanyon, Codester and their equivalents. Cheapest headline number in the industry and, as covered below, the licence attached to it is narrower than almost every buyer assumes.

One thing that is not a supply route

Poki and CrazyGames get named in these conversations constantly, and they do not belong in the list. They are destinations. You send a game to them; you do not get a catalogue from them. Poki's own blog reports 600+ developers, over 100 million monthly players and a billion plays a month as of June 2025. That is a place to publish, not a place to buy. If someone is pitching you "Poki-style content," ask which of the four routes above they actually mean.

💸 What the Networks Publish, Side by Side

The useful thing about the network route is that the terms are public. You do not need a sales call to price it.

• GameDistribution — its developer agreement states the developer is entitled to 33% of Net Revenue, paid within 60 days of monthly reporting once the balance reaches EUR 100. The publisher agreement sets the publisher's share at the same 33%. Three-way split, network keeps the middle third.
• GamePix — 45% revenue share, stated on both its developer page and its publisher page. Same headline figure on both sides.
• GameMonetize — 45% to developers on its developer page, with payment inside 25 days and a claimed reach of 7,500+ publishers.
• CrazyGames — publishes no split at all. Its FAQ confirms monthly payouts above a EUR 100 threshold, full retained ownership and no exclusivity requirement, and points to the contract for the actual percentage. Figures circulating in jam terms and blog posts are not the same thing as published standard rates.

Poki is widely reported in trade coverage as running a 50/50 split, which would be the most generous number in the category. Poki does not state it in its own developer materials, so treat it as reported rather than documented.

Read those numbers as a range: the supply side of web games clears somewhere between a third and half of net, and the buyer side of a network clears about the same. Anyone quoting you a number materially outside that band is either describing a different deal structure or selling something.

🧮 A Third of Net Is Not a Third of Anything You Can Audit

The percentage is the part everyone negotiates. The definition underneath it is the part that decides the money.

GameDistribution defines Net Revenue as gross ad revenue less in-game ads and hosting costs, less any applicable invalid-traffic and fraud deductions. In-app purchase revenue is gross less payment provider costs and VAT. All of those deductions are computed by the counterparty, from data you cannot see, before your third is calculated.

This is not an accusation. It is a normal structure and the deductions are real costs. But it means two things operators routinely get wrong:

• Your 33% is 33% of a residue, not of the money advertisers paid. The gap between gross and net is entirely determined by someone else's cost base and someone else's fraud filters.
• You cannot benchmark it. There is no version of your traffic running through a different ad stack to compare against, so you have no way of knowing whether your inventory is being monetized well or badly.

Compare that with running your own stack on licensed games. You see gross, you see the mediation waterfall, you see fill by geography, and you can fire an underperforming network on thirty days' notice. That visibility is the actual product difference, and it is worth more than the fifteen percentage points between a 33% deal and a 45% one.

📵 The Clauses That Decide Whether You Have a Product

Here is the part that costs people a roadmap. GameDistribution's publisher agreement states that publishers cannot show the games as part of a subscription or pay-to-own model, or in native apps, without explicit written approval. Games must be displayed on publisher properties — desktop and mobile websites the publisher owns. On termination, the tags come out and the games go with them.

Now line that up against the things a games business normally wants to do next:

• Ship an Android app. Ruled out by default. If your growth plan involves an APK, a telecom app store or an OEM slot, the free feed does not travel with you.
• Charge for access. Ruled out by default. Any subscription tier, any ad-free upgrade, any operator bundle billed to a phone account is a different licence conversation.
• Put games inside someone else's property. A publisher-properties clause means the media client, the bank, the airline and the hotel group you were going to serve are all outside scope.
• Own what you built. When the agreement ends, the catalogue leaves. Everything you spent on SEO, retention and brand pointed at pages that are now empty.

None of this is hidden. It is in the terms, published, in plain English. It is simply not what people hear when a network says "free games."

🛒 The Marketplace Route Has Its Own Licence Trap

Code marketplaces look like the cheap way to own something outright. Read the licence.

CodeCanyon's Regular License permits you to use the item to create one single end product, for yourself or for one client. The end product must be distributed free to end users — charging for access requires the Extended License. You cannot sell the end product except once, to that one client. You cannot redistribute the item as stock or with source files.

Apply that to a portal. A hundred games bought under Regular Licences, assembled into one site, is arguably one end product — and the moment you add a paid tier, run it for a second client, or spin up a second branded portal on the same catalogue, you are outside the licence on every title simultaneously. The per-game price was never the issue. The licence count was.

Marketplaces are a reasonable way to buy a game. They are a poor way to buy a catalogue, and a bad way to buy the basis of an agency business.

✅ When the Free Feed Is Genuinely the Right Answer

I am not arguing nobody should use networks. They are the correct choice more often than licensing vendors like to admit:

• You have no traffic yet. Paying for a catalogue before you know whether anyone shows up is a way to convert uncertainty into a sunk cost. A free feed lets you find out cheaply.
• Games are a side feature. A news site adding a games tab does not want an ad stack, a mediation contract and a payments dependency. It wants a tag.
• You are testing a category. Feeds are the fastest way to learn which genres your specific audience actually plays before you commit to buying any of them.
• Your traffic is small and geographically thin. Below a certain volume, your own ad stack will not get decent fill anyway, and a network's aggregated demand genuinely beats what you could negotiate alone.

That last one deserves more weight than it usually gets. Direct demand relationships reward scale. If you are serving a modest audience concentrated outside Tier 1, a network's pooled inventory may earn you more per play at 33% than your own stack earns you at 100%.

🔀 What Actually Triggers the Switch

The move from network to owned licences is not a traffic threshold. It is a product event. You switch when one of these becomes true:

1. You need a surface the terms exclude. An app, a smart TV build, a carrier portal, a kiosk, an embed inside a client's site. This is the most common trigger and it arrives suddenly.
2. You need to charge. The moment a subscription or an operator bundle enters the plan, the free feed is structurally incompatible.
3. You are selling to a third party. Agencies, integrators and white-label operators are serving other people's properties. Publisher-properties clauses are written specifically to prevent that.
4. Your ad revenue is large enough that the split is real money. Not "large" in the abstract — large relative to the fixed cost of licensing and running your own stack, which is mostly headcount, not software.
5. You need the catalogue to survive the relationship. Anyone building enterprise value rather than monthly cash needs games that do not disappear when a contract lapses.

Most operators hit trigger one or two long before trigger four, which is why "we will switch when the revenue justifies it" is usually the wrong plan. The switch gets forced by a roadmap decision, not by a spreadsheet.

🚫 Five Ways Operators Pick the Wrong Counterparty

1. Comparing 33% against 45% and calling it analysis. The percentages differ by twelve points. The Net Revenue definitions, surface restrictions and termination consequences differ by an entire business model. Compare the clauses, then the numbers.
2. Building the brand on rented inventory. Two years of SEO, a mailing list and a retention programme, all pointed at games that leave when the tag comes out. If the catalogue is not yours, the audience equity attached to it is not durable.
3. Assuming a network deal covers the app. It does not, by default, on the largest network with published terms. Teams discover this in the sprint where they were going to ship the APK.
4. Buying marketplace source and treating it as a catalogue licence. One end product, one client, free to end users. Multi-portal and multi-client plans break the licence on every title at once.
5. Paying for a licence before proving traffic. The inverse mistake, and just as expensive. Owning 500 games nobody plays is worse than sharing revenue on 500 games somebody does.

🎮 Where Forestry Games Fits

Forestry Games is the second route — a licensor. It has operated since 2017 with a catalogue of 1,049 titles spanning HTML5 games and Android APK titles, develops HTML5 games in-house, and works with branded IP including partnerships with Disney, Nickelodeon, Cartoon Network and Warner Bros. Licences are bought rather than revenue-shared, which means the monetization stack, the surfaces and the pricing model stay yours: apps, subscriptions, operator portals and client deployments are licence questions to be scoped, not clauses that rule them out.

If you are working out which route fits, the licence and pricing page sets out what a licence grants, white-label games covers the multi-property case that publisher-properties clauses block, and the full catalogue is browsable without a sales conversation.

🧭 Before the Next Catalogue Decision

Do this in an hour, before you talk to anyone selling you games.

Write down every surface your games will appear on in the next twenty-four months — website, app, TV, kiosk, operator portal, client property. Write down whether anyone will ever pay for access. Then open the publisher agreement of whichever network you are considering and check each item against it. Most of the decision resolves right there, on clauses that are public and take ten minutes to read.

If everything on your list is a website you own, monetized by ads, for the foreseeable future — take the free feed and put the saved capital into traffic. If anything on that list is an app, a paid tier, or somebody else's property, you need owned licences, and finding that out now costs nothing. Finding it out in the sprint before launch costs a quarter.