Acast adds algorithm-driven Recommendations

Podcast and on-demand audio platform Acast announced the launch of a Recommendations feature. This tool uses machine learning to find new content tailored to a particular listener’s tastes. The beta phase for Recommendations showed a solid performance: users are 52% more likely to follow a show suggested by Acast and 49% more likely to listen to multiple episodes of those shows.

This type of machine learning recommendation algorithm has driven other discovery tools across on-demand music and audio platforms. It’s similar to the idea behind Spotify’s very successful Discover Weekly playlists. In a conversation with RAIN News, an Acast spokesperson also characterized the new feature as “Youtube for podcasting,” comparing it to how YouTube generates recommendations of related content keyed to the user’s history. Not only does Acast’s system offer personalized recommendations based on listening history, but it also offers more precise targeting for advertisers.

“Searching for new podcasts is hard and often time-consuming,” Acast CTO Johan Billgren said. “We want Acast’s users to spend their time listening to great podcasts, rather than looking for them, and that is why we are launching Recommendations.”