Peloton’s Personalized Worjout Recommendations: How Do They Work?

Peloton is making it easier for members to find the perfect workout class with smarter technology. With over 53,000 classes, 120+ types, and 50+ instructors, the variety is amazing—but it can also feel overwhelming. That’s where Peloton’s Personalization team steps in, using advanced tech to recommend classes tailored to each user.

The Vision for AI-Powered Personalization Programming

Peloton is making it easier for members to find the perfect workout class with smarter technology. With over 53,000 classes, 120+ types, and 50+ instructors, the variety is amazing, but it can also feel overwhelming. That’s where efforts in Personalization come into play, using advanced tech to recommend classes tailored to each user.

Recently, Peloton detailed these technical developments in a blog post shared on LinkedIn, and we were intrigued! While we’re not AI experts, we are learning along with everyone these days how it can help us work smarter not harder, and that goes for working out too! 

Why Personalization is In Focus Now

New CEO Peter Stern has been talking this year about leaning in to AI use in many ways with Peloton. In both his Q3 and Q4 Earnings Calls in 2025, he highlighted ways that the company is leveraging technology for both member-facing experiences and internal productivity tasks. In our June article, “Peloton Innovation: Personalization and Expanding Reach,” Clip Out writer Nikki noted:

One of the most exciting topics Stern discussed was the integration of AI to provide even more personalized fitness experiences. Peloton recognizes that every member is unique, with varying fitness levels, goals, and preferences. With AI-driven solutions, the platform ultimately will be able to adapt to these differences, ensuring a tailored approach for every workout.

Stern highlighted existing features like Power Zones for cycling and Pace Targets for running, which already customize workouts to suit individual effort levels. But the real game-changer, he says, is the Personalized Plans feature. Over 600,000 members have used it to create custom programs designed to achieve their goals. The personalization results are clear too—with new treadmill owners reducing their mile times by 30 seconds on average within the first four weeks.

In mid-August, Peloton announced more specifics about future plans, summarized once again by Nikki in this article, Peloton’s Fall Refresh which announced lots of exciting changes including “a new Bike, Tread and Row, plus AI-powered features” and noted that “CEO Peter Stern emphasized a strategic approach combining hardware, software, and content.”

How Peloton’s Recommendations Got Smarter Using AI

Starting with Workout History

Peloton began by looking at members’ workout habits over time. For example, if you regularly alternate between 30-minute cycling classes with Alex Toussaint and 10-minute yoga sessions with Aditi Shah, the system learns this pattern and suggests similar combinations.

The Big Upgrade: Transformers

To make recommendations even better, Peloton adopted a cutting-edge AI model called a “transformer.” But what exactly is a transformer, and why is it so special?

Imagine a transformer as a super-organized librarian. Instead of just remembering the last book you borrowed, this librarian keeps track of all the books you’ve ever checked out, the order you read them in, and even the genres you prefer at different times of the year. It uses this information to recommend your next book with incredible accuracy. Similarly, Peloton’s transformer model looks at your workout history as a sequence, understanding patterns like your favorite instructors, class types, and even the times you prefer to exercise.

For example, if you usually do a 20-minute strength class with Adrian Williams after a cycling session, the transformer recognizes this pattern and suggests a similar combo. It’s like having a personal trainer who knows your preferences and adjusts your plan accordingly.

What makes transformers even more powerful is their ability to focus on the most important details. Just like a GPS focuses on your current location and destination to give you the best route, the transformer zeroes in on your recent workouts and preferences to make spot-on recommendations.

What’s New?

• Learning from What You Skip
  If you ignore a class suggestion, the system now learns from that. For example, if you keep skipping 60-minute classes, it will prioritize shorter options like 20- or 30-minute sessions on your home screen.
• Faster, Cheaper Updates
  Peloton’s new system updates daily instead of using a year’s worth of data. This means if you suddenly start exploring new fitness disciplines, like barre or Pilates, the system will quickly adapt and suggest more of those classes.

What’s Next?

Peloton is exploring real-time recommendations. Imagine your home screen updating instantly based on your mood or schedule. For example, if you log in after a stressful day, it might suggest a calming meditation class. Or, if you’re feeling pumped, it could recommend a high-energy HIIT session.

Where Do Personalized Workouts Go From Here?

Will these suggestions help you make decisions about your workouts as new equipment comes online? Are you using and enjoying the Personalization Plans on your screens? And how about those emails… here at The Clip Out, some of us don’t see the connection between those class recommendations and what we actually do… yet. Is your experience similar? What’s on your wishlist for AI integration in your Pelo-verse? Let us know! 

Source: Peloton Blog

---

Tune in to The Clip Out every Friday to hear Tom and Crystal’s take on this and other hot Pelotopics. We’re available on Apple Podcasts, Spotify, Google Podcasts, iHeart, TuneIn. Be sure and follow us so you never miss an episode. You can also find the show online on Facebook.com/TheClipOut. While you’re there, like the page and join the group. Lastly, find us on our YouTube channel, YouTube.com/TheClipOut, where you can watch all of our shows.

See something in the Peloton Universe that you think we should know? Visit theclipout.com and click on Submit a Tip!