D2C in games: the executive briefing

Where the money is moving

A handful of listed mobile publishers now take 40% or more of their revenue outside Apple’s and Google’s billing, and DoubleDown has passed half. Several large listed companies, including Take-Two, do not publish a percentage.

Every row comes from the company’s own release, report or filing. Each company measures against a different base (revenue, bookings, one segment, one month), so compare the direction of travel rather than rank the numbers. The D2C share tracker adds each quarter as it is reported.

Take-Two will not give a number. On its February 2026 call Strauss Zelnick said mobile direct-to-consumer “delivered its strongest quarter on record”, called the share “meaningful”, and said he expects “third-party take rates” to keep falling.

Estimates for the wider market point the same way. Each comes from a firm with its own method. Appfigures estimates App Store revenue from outside Apple, and Appcharge sells web shops:

• AppMagic estimates US mobile game D2C revenue grew 26% in 2025, and 38% among the top 100 grossing titles.
• Sensor Tower estimates that about 30% of the US revenue of the top 100 casino games in H1 2026 came through web shops, and that Monopoly GO’s web sales reached around a third of its revenue.
• Appfigures data reported by the Financial Times shows Apple’s US App Store commission revenue down 18% since the start of 2026.
• According to Appcharge, a web shop provider, and its survey with GDC, mobile game D2C was worth about $17bn in 2025, roughly 15% of $113.3bn in mobile in-app spending.

What changed in the rules, 2025 to 2026

Until April 2025, Apple charged 27% on US links to web checkouts and banned them in most other countries. Now several stores allow links, at different prices. The detail sits on each rules page; this is the short version.

Two things have not moved. Steam still requires in-game purchases to go through the Steam Wallet (Steam), and console makers still require their own billing or keep their terms private (consoles).

What it does to the P&L

On $100 of player spend, before your own payment costs:

The publishers that report D2C describe the effect in similar terms. Playtika’s quarterly filing puts payment processing and related costs on its direct sales at “typically 3 to 4%, compared to the 30% platform fee for third party platforms.” Its CFO said on the Q2 2026 call that cost of revenue fell 1.5% year on year, “primarily driven by lower platform fees resulting from the continued growth of our DTC business.” Take-Two’s president Karl Slatoff put it in four words: “It improves our margins.”

Stillfront’s wording is the most useful for a board, because it warns about the top line as well. Its Q2 2026 interim report says moving bookings to its own platform “is negative for net revenue but accretive to gross profit and gross margin.” Stillfront’s gross margin was 84% in the quarter, up from 82%. Two things can produce that pattern for you: a lower web price given to move players, and a provider that resells your goods so you book only its payout as revenue. A currency bonus does not change revenue per dollar of spend, because the player pays the same price and gets more items. The margin accounting page works through the cases. Your auditors decide the second point, so ask them before the board sees a revenue dip.

The platform fee is only the first line. Card processing, currency conversion, sales tax and VAT (the store used to collect and remit these as the seller), fraud screening, chargebacks and payouts all land on the web side. Published prices show the spread. Paddle, a merchant of record, lists 5% + 50¢ per transaction with tax and chargebacks handled. Stripe, a payment service provider, lists 2.9% + 30¢ for US domestic cards, plus 1.5% for international cards, 1% for currency conversion and $15 per dispute ($15 more if you contest it). Tax filing and fraud losses stay with you.

Visa’s threshold for excessive fraud and disputes fell to 1.5% of transactions on 1 April 2026 in the US, Canada, the EU and Asia Pacific. With a payment service provider, that ratio is measured on your own merchant account. The line items are on hidden fees, and the choice of model on merchant of record vs payment service provider.

The risks worth putting in front of the board

Players convert less on the web

The only public controlled test comes from RevenueCat, a subscription-billing vendor, and it is for a subscription app, Dipsea, not a game. In-app purchase converted 27.0% of users and the web link 18.1%. Game data is thinner. Tim Sweeney told The Verge that about 40% of Fortnite’s US iOS players chose Epic’s payment option in the two weeks after the game’s return, with 20% back in Epic Rewards as the incentive. That is a share of players, not of revenue. The business case rests on moving a share of your largest spenders, who buy often, rather than converting every payer.

Most players never hear about the shop

In Appcharge’s survey with GDC, studios named player awareness (50%) and player acquisition (41%) as their top barriers to D2C. 54% had no staff dedicated to D2C. Driving traffic covers what the high-share games do.

The rules can reverse

Build the plan so it still pays at 15% on US iOS links, and check whether it survives 27%.

You take on the store’s back office

With a payment service provider, sales tax registrations, VAT returns, fraud screening and chargeback disputes become your team’s work. A merchant of record takes most of that on in exchange for a higher fee. Either way, chargebacks and fraud and tax need an owner before launch.

A 90-day plan

Start with the first row this week. The spend breakdown by payer tier decides whether a web shop is worth building at all.