How Disney+ Hotstar is using Big Data and Artificial Intelligence

by MaddOx
0 comment

Disney+ Hotstar is one of the rising streaming platforms around the globe. Although it has just begun offering services in various regions, it’s already become a fan favourite due to its collection of Disney content and a bunch of shows belonging to Fox, HBO, and Showtime. Furthermore, you can catch live match streams on it as well. 

Disney+ Hotstar is capable of giving other streaming platforms a run for their money because it has been incorporating advanced technologies like Big Data and Artificial Intelligence to enhance the platform’s capabilities. 

In case, we’ve piqued your interest, you can try Disney Plus Bundle in UK to test for yourself how well the streaming platform fares compared to the rest of its competitors.

Let’s get into it.

Utilising a cloud delivery platform for streaming

Disney+ Hotstar enlisted the help of Akamai Technologies while streaming VIVO IPL 2018. On 10th April 2018, for the game between Chennai Super Kings (CSK) and Kolkata Knight Riders (KKR), the viewership peaked at 5.5 million concurrent views. Apparently, that was quite the record for a single streaming event online by any broadcaster. 

Akamai’s Intelligent Platform has an increase of up to 75% in mobile phone streaming. The cloud delivery platform was able to deliver 95 exabytes of data, take in 2.5 exabytes of data and entertain more than 1.3 billion devices per day. 

With such an influx of data, Machine Learning (ML) engines can improve the performance, security, and overall reliability of the digital experience. Best of all, it is easily integrable with web and mobile apps to provide you with cloud security, analytics, and much more. 

Shot of Corridor in Working Data Center Full of Rack Servers and Supercomputers with Cloud Storage Advantages Icon Visualization.

Accommodating spike in traffic

According to the Disney+ Hotstar regardless of the number of matches, they’ve streamed live; the traffic patterns remain unpredictable each time. Despite that, they rely on past patterns anyway to make decisions. As per Akash Saxena, the head of technology at Hotstar, the increase in traffic is likely to overwhelm the backend system and cause disruptions in the streaming.

That’s why as long as the legacy backend upholds its end of the bargain, it’s always easy to welcome new users to the platform. One of the methods shared by Saxena was Auto Scaling for handling an influx of users. He believes that auto-scaling results in having plenty of servers at hand so that users don’t face issues while streaming. 

Every day, the team at Hotstar deals with panic protocols, caching, and exponential backoffs to ensure customers have a seamless streaming experience. Saxena went on to say their entire platform is dependent upon three pillars: 

  1. Streaming infrastructure
  2. Subscription engine
  3. Metadata engine

Low latency numbers are also a good sign for a streamlined process. 

Customisable streaming 

Another major factor that can make it or break it with a streaming service is providing a prompt and efficient customer experience. With unlimited downloads every other day, Hotstar is developing an ML algorithm to make use of user intelligence from raw data. In this way, the algorithm will suggest ways through which a user’s streaming experience can be personalised.

Personalised streaming would mean the platform will be able to suggest content or ads according to the shows you’re watching. A metric which is pivotal for such information collection is the watch time per user per month. The ML algorithm learns user preferences through their watch history and then provides similar recommendations accordingly.

Hotstar also uses the real-time data platform Knol. It is built to provide a smooth exchange of data from Hotstar. The app collects a huge amount of data (i.e. nearly 10TB) which consists of ad impressions, customer support data, behavioural clickstream, and many others. So that a user may be recommended content which is much more accurate according to their likes.

Disney+ Hotstar home page

Thinking outside the box of streaming 

The Pradesh State Election was held in two cities, Gujarat and Himachal, in India. The display campaigns for the elections were streamed by Disney+ Hotstar while collaborating with an end-to-end digital solutions agency, Performics. Meanwhile, Resultrix, a major local digital marketing agency helped in identifying audiences and delivering ads programmatically. 

Due to the analytics, it was easier to keep track of the number of people, identifying the interests of such people and leading to an eventual increase in the time spent on Disney+ Hotstar. The entire campaign was carried out with a mixture of automatic, data, analytics, predictive marketing and machine learning.

Disney+ Hotstar will be continuing to collaborate with more agencies to get a number for their audiences and how to cater to them accordingly. For instance, they signed up with Zappr Media Labs to get mobile audience analytics to provide targeted ads based on advertising offers and personalised communication. 

All in all, Disney+ Hotstar certainly doesn’t want to give up on this adventurous venture any time soon. Who knows what else they got planned for the future? They seem to be doing a pretty good job thus far. If you’re eagerly awaiting Hotstar in your region, no worries. They’re expanding at a good pace. Until then, you can simply use a VPN to get Disney+ Hotstar outside India.


For even more opinion pieces like this, click right HERE.

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.