How recommendations help you stay relevant in content overload

How Recommendations Help You Stay Relevant in the Era of Content Overload

One thing we can guarantee about the future: we’re never going to run out of content.

Take TV for example. Where once we had a handful of channels broadcasting one program at a time, we now have multiple streaming platforms, countless cable channels, on demand, and DVRs.

Or music: You’re not limited by your carefully curated CD collection anymore. You can choose from almost any song ever recorded on Spotify.

For the content consumer, it’s an embarrassment of riches. For businesses that rely on advertising or subscription revenue, it’s a challenge.

Attention spans are shrinking. With endless options, consumers will move on in matters of seconds if what they see or hear doesn’t capture their interest.

To stay relevant in the media industry — bringing targeted audiences, charging top dollar for your ads and maintaining a healthy growing subscriber base — your content needs to be relevant.

And, of course, every consumer’s tastes are different. The key to relevance is personalization recommendations.

For example, after revamping its mobile website to deliver a personalized, Facebook-like experience, USA Today saw a 75-percent increase in time spent per article.

Recommendation Models Used By Successful Advertising and Subscription Businesses

As content executive Paul Lentz points out, publishers have been using data to target specific content at specific audiences since the print era.

In today’s digital era, a few successful media companies have developed recommendation techniques to engage and retain users with almost supernatural precision.

  • After experimenting with content-based and collaborative filtering, the New York Times settled on a best-of-both-worlds approach that models the content and adjusts it according to viewing signals from readers, models reader preferences, and uses the resulting data to make recommendations.
  • Netflix’s recommendation engine divides users up into “a couple thousand” taste groups. Netflix claims the engine is worth $1 billion a year and is responsible for more than 80 percent of the shows users choose.
  • Spotify’s Discover Weekly playlists have become a favorite feature among users for introducing them to new songs and reminding them of old favorites. The “magic” of the algorithm, the man behind the playlist says, comes from comparing your listening habits to those with similar taste and “filling in the blanks.”

What does each of these approaches have in common? Each media company leveraged a massive database of user data to make comparisons among users, identify trends in their preferences, and anticipate their behavior.

You can do the same with Blueshift’s AI-powered marketing platform. Blueshift can help you

Learn how to configure recommendations in a single click or bring your own algorithms to BlueShift Personalization Studio.