Family Encyclopedia >> Entertainment

How Netflix knows exactly what you want to watch

Netflix's rise to being the world's leading streaming media service was no fluke. It relied on a complex recipe of data and emotion manipulation that means the company knows what you want to see even before you meet.

According to the latest quarterly numbers from Netflix, the streaming media service is amassing nearly 2 million new subscribers every month .

The reason for this relentless growth (whether you think it's a good thing or a bad thing) runs deeper than having a reasonably priced and convenient service. Why You Should Pay More With Netflix Why You Should Pay More With Netflix Every time Netflix increases its capacity In pricing, millions of binge-watchers balk at the prospect of paying a couple of bucks more. But this is dumb, because Netflix is ​​an absolute steal. Read more to sell. It goes beyond your budget and mammoth marketing strategies. And it certainly goes beyond the 5 Ways to Search Netflix from Netflix, Hulu, Amazon, and More at Once 5 Ways to Search Netflix, Hulu, Amazon, and More At Once If you're still having a hard time deciding which of online movie streaming services is right for you? One of the most important factors to consider when making this decision is… Read more.

It's Netflix's secret sauce of algorithms, big data, and gut instinct that fuels this unstoppable growth. It's this secret sauce that allows Netflix to not only consistently recommend content that users will (probably) love, but also to finance the creation of that content, trusting that it will be Be a success.

Incredible amounts of Big Data

Not surprisingly, big data plays a significant role in Netflix's ability to recommend and fund the right content. What is surprising, however, is the type of data and the amount of data that Netflix tracks every time you use the service.

How Netflix knows exactly what you want to watch

According to the official Netflix Tech blog:

As part of this process, Netflix tracks your "full viewing history while [you] are a subscriber." The system "collects periodic signals throughout each view to determine whether or not a member is still watching." It also keeps track of your searches, ratings, geo-location data, device information, browsing behavior, time of day/week you're watching, when you decide to leave a show, pause and forward.

With millions of Netflix users streaming billions Of the hours of content each month, the amount of data the company collects is baffling. This data is massively important for the success of the company..

In 2014, GigaOM paraphrased Netflix's Senior Data Scientist How to Become a Data Scientist How to Become a Data Scientist Data science has gone from a term newly coined in 2007 to one of the most sought-after disciplines today . But what does a data scientist do? And how can you enter the field? Read more, Mohammad Sabah, saying:

This viewer feedback is huge, and it's imperative why the service can be so addictive. Combined with the vast array of data stored about each show, it's hard to disagree with David Carr's theory that "Netflix is ​​commissioning original content because it knows what people want before they do it." (emphasis mine).

Algorithms in constant improvement

By itself, the data is of little use. As Jason Gilbert wrote; “[Netflix's] success is based on how well it can choose the programming that its viewers like and at the same time be profitable.”

How Netflix knows exactly what you want to watch

To do this, Netflix uses algorithms. As engineering director, Xavier Amatriain told Wired:

This focus on display patterns is proving to be much more reliable than looking primarily at the rating you give a show.

As data about users. and content is fueled by this machine learning How Smart Software Is Going To Change Your Life How Smart Software Is Going To Change Your Life Skynet is coming and will be incredibly popular. New AI technologies are emerging that will enable the way we live, play, and work, Read More algorithms, viewer behaviors can be matched with shows that have certain similarities:year of production, cast, director, etc. As we can see in the number of hours of media streamed on Netflix each day, these algorithms clearly work. But they are always a work in progress..

La compañía está ejecutando constantemente un gran número de pruebas A / B (permitiendo que la experiencia del usuario y los cambios de algoritmos se implementen y se prueben en pequeños subconjuntos de usuarios) para mejorar iterativamente cada uno de esos algoritmos. Según Amatria en estas pruebas. “Probemos ideas radicales o probemos muchos enfoques al mismo tiempo.”. El objetivo principal es casi siempre mejorar. “compromiso del miembro (por ejemplo, horas de juego) y retención”.

Instinto

En otra publicación del blog de Netflix Tech, Xavier Amatriain afirma:

How Netflix knows exactly what you want to watch

La idea de espectáculos siendo fabricados y recomendados basados. only En los datos es algo perturbador. Pero la industria de la televisión siempre se ha basado en gran medida en los datos (a menudo en forma de grupos focales y números de espectadores). Sin embargo, Netflix está llevando esto bastante lejos..

Dicho esto, Joris Evers, el director de comunicaciones corporativas globales de la compañía, quería aliviar la mente de los usuarios. Le dijo al New York Times:

En otras palabras, el valor de big data y algoritmos informa la decisión de Netflix en lugar de dictarlos. Las ideas creativas para que las películas se financien, y los programas que se otorgan bajo licencia vendrán densos y rápidos. Los que sensación Bueno, se someterá a los datos. Si parece que una sección lo suficientemente grande de los usuarios de Netflix estarán interesados, y el sentimiento de los tomadores de decisiones dice que el programa será un éxito, se le da un pulgar hacia arriba y un gran cheque..

Esta receta parece funcionar

Esta combinación de datos, la mejora continua de los algoritmos y el instinto de inteligencia parecen estar funcionando para Netflix. Tanto es así, de hecho, que la compañía tiene la confianza para financiar series completas de espectáculos antes de Lanzamiento de un episodio piloto. La mayoría de las otras emisoras trabajan de manera opuesta.

How Netflix knows exactly what you want to watch

Los productores y directores pueden lanzar ideas creativas a Netflix. Si el big data y el sentimiento general se suman, y sugieren que los costos pueden recuperarse en términos de nuevos suscriptores ganados y mayor retención, Netflix puede ir todo incluido.. Castillo de naipes es un ejemplo, donde la compañía invirtió $ 100 millones en dos temporadas sin siquiera ver un episodio piloto. Y es por eso que 2016 verá a Netflix producir más contenido original que la mayoría de las otras emisoras en varios años..

Esto no sería posible si Netflix no fuera capaz de ser desconcertantemente confiable para entender y predecir lo que a usted (o al menos a la mayoría de la gente) le gustaría ver. Antes incluso de que te conozcas a ti mismo.

A usted:¿Encuentra las recomendaciones de Netflix que se adaptan a sus gustos? De lo contrario, intente estos códigos de búsqueda secreta de Netflix. 20 códigos secretos de Netflix garantizados para ayudarlo a encontrar nuevo contenido. 20 códigos secretos de Netflix garantizados para ayudarlo a encontrar contenido. Read more . ¿Y te sientes cómodo haciendo que Netflix sepa mucho sobre tu comportamiento visual, tus gustos y aversiones??