Spotify (goal: more listening time)
Spotify learns what you like—artists, songs, moods—and builds playlists (Discover Weekly, Daily Mix) that feel made for you. When the music fits, you listen longer. Longer listening keeps you coming back and makes you more likely to pay for Premium. Simple ROI: if better recommendations add just 5 extra minutes a day for 100,000 people, that’s millions of extra minutes a month. Assume a conversion rate of 1 extra percent for the 446 million users, then that is 4.46M x $11.99 = more than 53M additional revenue per month.
LinkedIn (goal: more visits per day)
LinkedIn watches who you know and what you interact with, then fills your feed, notifications, and even send emails with posts, people, and jobs you’ll likely care about. The result: you open the app more often. ROI idea: if personalization nudges +1 more sessions per day for 2 percent of the 1.1B members, that’s about 22M extra sessions a day. A small slice of those sessions turn into job views, messages, or applications—high-value actions that justify the investment and let people be more willing to pay for premium with the network effect.
TikTok (goal: better ad performance)
TikTok tracks tiny signals—what you watch to the end, skip, rewatch, or share—to predict the next video on “For You.” Better matches mean people watch longer, and ads get shown to the right viewers. ROI idea: a +0.1% lift in ad click-through across 100 million impressions is 100,000 extra clicks. Even a small increase in the price advertisers pay per thousand views can turn that into clear revenue.
Personalization serves a different “money metric” for each app—time for Spotify, frequency for LinkedIn, ad results for TikTok. The payoff doesn’t need to be huge; small, steady lifts at scale easily cover the cost of smarter recommendations.
