How Data Determines Your Next Great Watch | by Demi Wait

This is it, the season finale of that TV show you’ve been watching forever. And before you’ve even processed it, the end credits are rolling. What do you do now? 

We’ve all been there. 

So you’ll start scrolling through your streaming service, hoping that what comes up is good. Somehow, those recommendations that come up for you are always pretty solid. But how?

Unsurprisingly, there’s data behind it. These days, to keep people’s attention, businesses can’t just throw out generics – they have to make things personal to the individual now. Sure, some things are popular among many and the generic throw-out could help, but that’s not always the case. 

So, yes, you’ll likely be recommended the show that everyone’s talking about. Because, if a high percentage of users enjoy it, it’s likely that you will too. But what about the things that feel less one size fits all?

Essentially, streaming services use a ton of data to gauge exactly the sort of things you’ll love. They collect data on what you watch, if you rate it, how many times you watch it, if you pause it, where you’re watching, the time of day you’re watching, and the list goes on. And all of this is used to show you things that you are likely to enjoy.

The data is constantly used to build a profile of you: your likes, dislikes, what you could watch over and over, and what you couldn’t finish. From there, the shows and films you’re most likely to enjoy are placed in order, shown to you in a horizontal and vertical grid, the ones you’re most likely to enjoy placed earlier on in higher rows, those maybe less so later on, further down. 

Even down to the variant of the poster, the streaming services show you to keep your interest in what you’re looking at, your data is involved. If you frequently turn films off at scary scenes, for example, it’ll assume you’re not a horror fan. No scary thumbnails for you! If you rewind there to watch things again though, maybe you’re more of a fan! You’ll notice you’ll get more thumbnails like that. 

It’s not just your data that’s used for personalisation. Yours will be compared with the data of people in your country and region to determine what’s popular in your city, for example.

Whenever I finish watching something, I frequently find myself thinking, how on earth did they come up with that idea? I imagine somebody having to go to their managers to present some absurd idea and explain why it’d work, despite seeming like a risky gamble.

This isn’t quite how it works, either. Surprise surprise, data usually gives pretty decent insights into what trends could be popular based on the same user statistics that are gathered from you and used to recommend things that already exist. So while the idea may seem absurd, there’s often solid data suggesting that there’s an audience for it that the producers can advertise it to!

For example, they may include specific actors or genres or tropes that make it more likely to be successful. Notice how lots of actors are popular in similar sorts of films?

So, thanks to your data, it’s far easier to find something new to fill that gap your last show or film left! Enjoy!

Demi Wait is an Advanced Data Fellowship apprentice at Multiverse based in London, England, and is writing for The Apprentice Lens. Here's more about her:
"Hi, I'm Demi. I currently work as an Apprentice Data Scientist for Colt. Outside of work, I love writing of all sorts and am most likely found with a book in my hand. I find huge inspiration in all sorts of tech, real-world or fictional!"