Unity bets that game value sits in monetisation data

Unity closed its $4.4 billion ironSource deal as its own ad business shrank. In mobile games, the models that price players now matter as much as the engine.
Unity's engine business, Create Solutions, grew 54% in the third quarter while Operate Solutions, its advertising and monetisation business, shrank 7%, and the company has just closed a deal to fix the weaker half. On 7 November it completed its all-stock merger with ironSource, the app monetisation company it agreed in July to buy in a deal that valued ironSource at about $4.4 billion; the quarterly results followed two days later.
The deal shows where Unity believes the money in mobile games sits. Most of it moves through the monetisation layer rather than the engine, and that layer is a machine learning business: models predict which players will stay, pay or respond to an ad, and price every impression accordingly. Unity now presents itself as the end-to-end platform for mobile app creators, the engine that builds the game plus the tools that acquire players and sell advertising inside it. The engine, priced as infrastructure earlier this year, becomes the channel that brings developers in, while the data decides the margin.
Unity's own year shows how much rides on those models. In May it said a fault in Audience Pinpointer, its machine learning tool for targeting ads, made worse by bad data from a large customer, would cost about $110 million of revenue this year. One data problem in one model moved the outlook by that much. ironSource brings more data from more devices and control of a mediation platform, where developers choose which ad networks fill their games, and the scale and quality of behavioural data are becoming the moat in game monetisation.
Studios end up depending on a few platforms that both build and monetise their games. Those that keep first-party player data and their own analytics can compare networks, switch providers and understand the value of their players without a supplier's model in between, and they should negotiate platform terms that keep that data portable from the first contract. Tool builders have the mirror opening, in the reliability of the models behind ads and live operations: validation of training data, monitoring of model drift and targeting that works with less data per user.
A merger can add revenue without repairing anything. Whether Operate Solutions returns to growth once ironSource is consolidated will show which of the two Unity bought, and whether it rebuilds its targeting models on the combined data and reports how they perform will show if the data advantage is real. Developers hold the last vote. An engine owner that also runs the ad marketplace in their games invites a response, and if a share of them moves monetisation to independent providers, the platform loses the very data it paid for.
Games are one of the few markets where machine learning already earns money at scale, in pricing players and ads rather than in making content. Fanaura counts games among its core sectors, invests with a horizon of years and looks for studios and tools that own their player data and the models that turn it into revenue. Here, whoever owns the data sets the price.
Building an AI-native company? Apply in five minutes. →
More from Fanaura
Building tools that make game monetisation models more reliable, or studios that own their player data? Send us the project.
Fanaura invests its own capital and the capital of its limited partners (LPs), usually through a separate SPV for each deal. If we see a fit, we get in touch.
Get the news by email
One short note each weekday. No marketing, and you can unsubscribe at any time.