Ideas
Why Streaming Algorithms Show You the Same 20 Things
By Jordan A. · · 3 min read
You've seen the same twenty thumbnails for months. You know the carousel by heart at this point. The one with the guy in the red suit. The one with the woman running. The one with the robots. Every time you open the app, there they are, waiting for you like coworkers you don't want to talk to. There's a real reason for this, and almost none of it is about your taste.
Reason one: algorithms are trained on engagement, not satisfaction
Streaming algorithms optimize for clicks and watch time, not for how good you feel about the thing you watched. If you watched two minutes of a show and left, the algorithm learns "they engaged." It doesn't learn "they hated it." The more you accidentally click, the more you get shown what you accidentally clicked, in a loop.
The result is a menu that learns your worst habits and feeds them back to you. Nobody's trying to make you miserable. It's just a side effect of optimizing for the wrong thing.
Reason two: the carousel is a business deal
Most of the real estate on your streaming home page isn't picked by an algorithm at all. It's bought. Shows the service wants to push, originals that are about to launch, licensed content that's expensive to carry and needs to justify its cost. The algorithm only gets what's left.
This isn't a conspiracy. It's how every big media platform works. But it does mean the thing you're looking at isn't really "what's best for you." It's "what's best for the service, with a small nod to what might keep you on the app."
Reason three: single-platform vision
Every streaming service can only recommend its own content. Netflix cannot and will not tell you there's a better match on Max. The recommendation engine is literally not allowed to see outside its own walls. So if your perfect next show is on a service you don't pay for, you will never hear about it from any of the services you do pay for.
Reason four: the cold start problem
Most recommendation systems are bad at new content and good at safe content. If a show just came out and five people have watched it, the algorithm doesn't know where to put it. So it doesn't put it anywhere, and it gets buried under the safe picks that are already known to hold your attention.
That's why you keep seeing the same twenty things. They're the ones the system is most confident about. "Confident" and "best" are not the same word.
Reason five: you're not actually the target
This one is the quietest and the most important. Streaming recommendation systems optimize for the median viewer, not for you. The median viewer is a composite. They have broad tastes and they don't care that much. You're probably not the median viewer, which means the median-tuned algorithm is always going to feel slightly wrong.
What you can actually do about it
You can't retrain Netflix's algorithm from the outside. But you can route around it. That's what BingeNext is for. It isn't owned by any service, so it has no reason to bury content that's on a platform that doesn't pay it. It looks across everything at once. It ranks by how close a show or movie is to what you told us you love, not by what kept you on an app the longest.
You tell it two things you love. It gives you a real list. You watch one of them. You stop scrolling forever. That's the whole pitch.
The point
The same twenty thumbnails isn't your fault. Most recommendation engines are designed to reinforce the middle of the pack, protect the business deals, and keep you inside the walls of one service. That's not a taste problem. It's a system problem. The fix is to get your recommendations from something that doesn't work for the same system.
Topics covered: algorithms, streaming, opinion
About the author
Jordan A., Founder, BingeNext
Jordan A. is the founder of BingeNext and writes every guide on this site. He built the recommendation engine after one too many nights lost to the streaming menu, and he still tests every pick the hard way: by watching it.