How Is Netflix Just So Damn Good?

This topic was decided upon stemming from the thought, “why is so much good content being produced on Netflix?”. Some of my favourite shows in recent years have been Netflix original productions (House of Cards, Orange Is The New Black), or Amazon original productions like Transparent. Shows like these have and continue to dominate prestigious Hollywood awards seasons, winning Emmy’s and Golden Globes year after year. The amount of VoD services that have original productions nominated and critically acclaimed grows every year. So clearly, my previously mentioned thought, has some validity to it, despite the fact that the term ‘good’ relative to content can be incredibly subjective.

I have realised that I’ve failed to mention that I intend on presenting my final project in the form of a research report. Therefore, I have also realised that I need to get crackin on a literature review. Luckily for me, I have come across an academic thesis written by Henry Zhu Tang in 2014,The Collaborative Filtering Effect of Netflix Ratings for Indie Films versus Blockbusters and Heavy Users versus Casual Users. This source is incredibly valuable to me as it incorporates many of the themes I discussed (and intend to expand on) in my previous blog post. Tang writes about the way Netflix uses recommendation algorithms to assist it’s users in finding content they presumably would be interested in and how this correlates to the type of content Netflix chooses to buy and also fund production of. Before reading this, I wasn’t even entirely aware of this connection. Everyone knows about the recommendation algorithms, love them or hate them, if you use the service, you are subjected to them. Personally, I don’t know where I stand on the privacy issue of Netflix knowing intricate details about my personality based on my TV and movie taste, but I do like a good recommendation. I hadn’t thought deep enough about the connection to how they utilise the recommendation algorithm for the type and quality content they offer. As it turns out, Netflix started out in 1997 as a service dedicated to providing more alternative content:

“In 1997, Reed Hastings and Marc Randolph founded Netflix, an online DVD-by-mail retailer that usurped the traditional brick-and-mortar model. At once, a wider library of titles had become available to consumers than ever before. Netflix introduced a proprietary recommendation system, powered by a collaborative filtering algorithm, to select movies to watch for its customers, a feature it continues to use for its global video streaming service today. This collaborative filtering algorithm would further highlight indie or niche films that could not be found (or were prohibitively difficult to find) in stores.”

Many of the ideas Tang writes about are connected to 4 of my 5 main talking points so far:

  1. Content with better diversity.

      2. Creators having more freedom around the production of content.

      3. VoD services content favouring audience viewing habits.

      4. Netflix buying up the rights to more low budget, yet ‘prestigious’ films at Sundance.

Due to how supportive this thesis is to my talking points for my report, I will likely go ahead and rely heavily on it throughout.

4 thoughts on “How Is Netflix Just So Damn Good?

  1. As someone who always used to google ‘shows like blah blah’ I quite enjoy the suggestion feature from Netflix. It also saves me a bit of time looking through the catalogue and figuring out what other show I want to waste my time on. I found for a while, my suggestions were a bit random and that’s because it was suggesting shows based on location and not personal preference. The newer design allows for each account to be personalised and proactive with content sharing. This article from The Verge (http://www.theverge.com/2016/2/17/11030200/netflix-new-recommendation-system-global-regional) was quite interesting if you want to read up further on the algorithm and it’s production.

    Like

  2. Being a frequent user of Netflix, I feel very strongly about the fore mentioned “recommendation system”. Whilst spending hours on Netflix , trying to choose what to watch, I always amuse myself by seeing what bizarre recommendation Netflix has constructed for me, based on what I like. For example, having rated Interstellar 5 stars, Netflix recommended me Friends with Benefits, a film which was pretty much mad for ‘Netflix and Chill’. So it’s fascinating to me to understand how exactly the algorithm actually works. Great work btw!

    Like

  3. Do you find the Netflix recommendations accurately predict what you might want to watch? I hugely enjoy teen dramas or college movies, particularly anything to do with cheer leading, dancing or football – though I don’t subject my friends to my (maybe poor) taste. But that’s for mind-numbing, no-thought-required viewing when I just want to blob on the couch. More and more I’m finding it difficult to find other types of content on Netflix, and when I look at friends’ accounts, I find titles I’ve never seen in the browse section on my account. I’ve been tempted many times to delete my account and start over, or create a separate user account for when I want to watch something with substance.
    Do you feel that recommendation engines lock us into pigeon-holes? And how does that apply to the way services like Google collect user information for Ad Analytics? Or Facebook’s targeted advertising?
    User pattern analysis is an interesting field with a long way to go.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s