ByElise Jost, writer at
"It's a UNIX system! I know this!" Twitter @elisejost
Elise Jost

Netflix recently announced they'd be revamping their apparently painful 5-star rating system by replacing it with a thumbs-up or thumbs-down, based on whether you enjoyed what you just watched or not. Did they think it was a waste of time to wonder whether you'd give 3 or 4 stars to the last OKish movie you watched, when those precious seconds could have been dedicated to getting you addicted to a new show? Probably. Now you only have to choose between two options, and the platform's algorithm will factor in your pick for its subtle prediction of what you're going to like next.

That might seem like a simplification of the prediction system, but Netflix is actually fine-tuning its recommendations even more. Instead of suggesting a bunch of movies and shows you might like, Netflix should soon give you matching percentages for new content, like a dating website. Let's say a show has a fair chance of entertaining you — now it's got a little 80 percent tag attached, similar to declaring that there's an 80 percent chance you're going to like it.

2016's Netflix sensation 'Stranger Things' [Credit: Netflix]
2016's Netflix sensation 'Stranger Things' [Credit: Netflix]

The result of this will probably be that you'll never click on something that has less than 50 percent compatibility with the taste in shows that Netflix has determined for you, based on your viewing activities. Or maybe you'll click on the 5 percent shows just for a laugh before returning to your good ol' habits of only watching American rip-offs of that one romantic docu-drama set in Western Canada that you really loved.

There's a double risk in overly tailored recommendations, both for you and your curiosity and for the shows trying to find an audience. While some may benefit from the system if they're lucky enough to be matching with your preferences, I can imagine that the more a show defies categories, the less likely it'd be to end up in that 80+ circle for users who enjoy a very specific set of genres.

See also:

As for you, wouldn't you rather have Netflix tell you, "We know you liked this and that, but here's a hot new show that seems to have nothing to do with your taste that you could try"? In our golden age of personalized feeds and curated pages, it's becoming an old debate already, but are super-tailored recommendations really benefiting us? It's not so hard to imagine that in the near future, we won't touch something unless an algorithm has assured us it is OK to do so. (Alright, the video below might be a bit on the nose.)

Of course we don't have the time to watch every dreadful movie out there until we find The One, and reviews and ratings offer the guidance of a lighthouse in an endless sea of entertainment choices. But we should also be careful to retain the ability of being taken by surprise, whether in a good or in a bad way. If we rely excessively on computer-made opinions of what we might find pleasure in, we're taking away the joy that comes with unearthing an amazing film and the patience that movie watching requires. There's such a satisfying sense of victory to sticking with a movie or show that started out unconvincing, until it turned around half-way and became our new favorite addiction.

Skip the "You might like this" section and take a wild scroll through the bottom category on your Netflix homepage. Trust your own instinct on what looks good, and sure, maybe it'll be a complete waste of time, like when you think you can master that 2.5-hour macaron recipe and end up repainting the kitchen with cookie dough. Spoilers: Watching Netflix isn't saving lives, so why would you need to optimize every second of it?

Do Netflix's recommendations work for you? How do you usually pick a new show?

'Fight Club' [Credit: 20th Century Fox]
'Fight Club' [Credit: 20th Century Fox]


Latest from our Creators