Netflix is everywhere.
There are 110 million Netflix subscribers around the globe. With roughly 2.5 users per subscription, that’s an estimated 275 million viewers tuning in. The world’s leading online streaming service is a force to be reckoned with.
Netflix’s success lies in its ability to provide personalized suggestions to users based on their viewing habits – often resulting in addictive behavior. It’s all thanks to big data, AI (artificial intelligence) and machine-learning algorithms.
At Yieldr, we’ve also done our fair share of data analysis. We looked at more than 5 million airline ticket sales over the course of 6 months, from Q1 to Q2 2017, and we discovered some interesting things about traveler flexibility.
Our findings give key insights into traveler habits that can be used to build an accurate user profile. Airlines could build upon this further and combine passenger and revenue data to successfully recommend and sell tickets for underperforming flights.
Can airlines replicate the Netflix formula? Could they automate personalized flight recommendations to travelers – and perhaps get them addicted to booking more flights? Let’s check out the possibilities.
Fasten your seatbelts and get ready for our insights!
How Netflix Uses Analytics
It’s no secret that Netflix is a data-driven company. They have a deep understanding about their customers, which allows them to keep fine-tuning their insights and providing better viewing recommendations.
It starts off simple enough. After signing up, users are asked to select up to 3 shows they like, which helps Netflix get a better idea of what the user would like to watch. Viewers end up with their own profile, which gets more personalized as time goes on.
In the beginning of 2013, Netflix’s Director of Global Communications, Joris Evers, stated that “There are 33 million different versions of Netflix.” They had 33 million subscribers worldwide at that time. This highlights Netflix’s ability to cater to each individual user, for the ultimate personalized experience.
Netflix has developed an extremely intelligent set of ranking algorithms to do this. There have been many great articles about how Netflix uses data, but we’ve gone straight to the source for a simplified explanation:
We use a recommendation algorithm that takes certain factors into consideration, such as...
- The genres of TV shows and movies available.
- Your streaming history, and previous ratings you’ve made.
- The combined ratings of all Netflix members who have similar tastes to you.
We use these factors when we calculate the percent match score shown next to a title. This score is unique to you, and indicates how likely we think you are to like that title.
It goes much deeper than this, of course. For example, there’s a group of Netflix workers who are dedicated to adding meta tags to video content – not just by genre, but also seemingly innocuous details like whether a show features a “strong female lead” or if it’s a “cerebral crime show”.
When Netflix mentions “Your streaming history”, it’s not just what you watched, but how you watched it. Did you only see 10 minutes of a show, or binge through an entire season in two days? Do you watch TV series on weeknights and save films for Sunday afternoons? Do you rewatch scenes with your favorite actor? Did you watch A Christmas Prince every day for the past 18 days?
Netflix has also used data to determine a recipe for creating its own successful content. The first Netflix Originals series, House of Cards, was based on the data of Netflix users’ streaming habits. Three factors seemed to overlap with a significant amount of viewers – films directed by David Fincher, films starring Kevin Spacey, and the original UK House of Cards TV series. This prompted the creation of the US version of the series, as well as several trailers tailored to different types of viewers based on their preferences.
It’s fair to say that Netflix has nailed personalization.
Download What If Airlines Were Like Netflix? to continue reading our report, where we discuss how airlines can use machine learning and AI technology to fill empty seats and maximize revenue. We dive into traveler flexibility and follow up with a hypothetical case study.