News

How Supercell uses machine learning to automate monetisation in Clash Royale

How Supercell uses machine learning to automate monetisation in Clash Royale

Supercell data scientist Jarno Seppänen has explained how Clash Royale uses machine learning to target card sales at individual users.

Speaking at the Games First conference in Finland earlier this year, Seppänen described how Clash Royale uses machine learning to handle aspects of the game’s monetisation. Particularly, how the technology can serve players the most valuable cards for them.

“We’re on a small team, we don’t want to grow the team. Whatever happens on a regular schedule, we try to automate that as much as possible,” said Seppänen. “Then we focus the more human efforts on high-profile events.”

Machine royale

The game determines how many of a particular card the player owns, how close they are to upgrading, and the frequency the player makes purchases - among other factors - to determine what deals to promote on the game’s store.

Many of these factors, like pricing and stack size, are determined by hard-coded design. But machine learning drives the understanding of what cards players are more willing to purchase and what offer to serve the user.

You can watch Seppänen’s full talk in a newly published video below.


Staff Writer

Natalie Clayton is an Edinburgh-based freelance writer and game developer. Besides PCGamesInsider and Pocketgamer.biz, she's written across the games media landscape and was named in the 2018 GamesIndustry.biz 100 Rising Star list.

Comments

No comments
View options
  • Order by latest to oldest
  • Order by oldest to latest
  • Show all replies
Important information

This site uses cookies to store information on your computer. By continuing to use our site, you consent to Steel Media's privacy policy.

Steel Media websites use two types of cookie: (1) those that enable the site to function and perform as required; and (2) analytical cookies which anonymously track visitors only while using the site. If you are not happy with this use of these cookies please review our Privacy Policy to learn how they can be disabled. By disabling cookies some features of the site will not work.