This is a blog post from
Streaming services are cataloguing the entire world’s audiovisual content onto their platforms. If you had told someone ten years ago that most of the world’s movies and TV shows would be available on demand in their pocket they would have given you a patronising look of disbelief. If at present of the streaming industry is pathbreaking, the future promises to build even further on it.
Streaming is funded by subscriptions and guided by big data analytics. Knowledge of how consumers behave on platforms such as Netflix and Prime Video lets the services gauge what else they might be willing to pay for. It is hard to say which way streaming is likely to go over the next decade. However, it is possible to make an educated guess based on the frameworks of information economics.
Three main areas are likely to be affected by the continued usage of big data to improve streaming, user experience, security, and pricing.
When the news broke that Netflix customises individual thumbnails for each user, it was another endorsement of how the platform was using big data to keep customers hooked. Netflix doesn’t just use a film or show’s original art; it employs an algorithm to source high-quality images from the content. Then it does more testing to determine what individual subscribers are most likely to click on. Based on that, each user’s Netflix homepage looks different, even if they have similar tastes. The idea is to have users spend as much time on Netflix as possible, and personalised thumbnails are a small cog in the working of this big machine. Data on who binge-watches which shows and how long each visit on the website lasts is also crucial when it comes to deciding what to invest in. This is not a new phenomenon. The TV show House of Cards is a case study to understand this.
Big data tells Netflix (and Prime Video and Hulu and Hotstar) what users want even before they themselves know it. The data-based knowledge that David Fincher’s movies were in high demand — this insight was based on the number of times people played and paused them and how long they watched for — was a powerful resource for the TV decision-makers. Combining Fincher with a star-studded cast was not a shot in the dark. Netflix bet $100 million on two seasons (26 episodes) of the show at first, without watching a single episode. They even went as far as to make different trailers and filtered their distribution according to user preferences. This just shows that the information about our tastes and tendencies, as exhibited by big data, is empirically reliable.
Going forward, big data analytics will continue to tell companies what the users want. This will have a significant impact on how funds are distributed across genres. For instance, the success of Narcos and Stranger Things will drive investments in more original content in their particular genres. This also means evolving content markets all over the world to keep users hooked and get new users to subscribe (think about the success of Sacred Games in India).
No free-loading
The increasing use of big data analytics will also mean tighter security for accounts. So, no chance of four people pooling their money together to get one streaming account. Also, no mooching off your friend’s account. The free-rider problem means streaming giants lose money on every individual that watches content without paying for it. Because Netflix has data on usage patterns — laptop model, user location over time — it can identify when someone other than the paying customer is watching. So, it is no wonder that the company is now planning on using AI to keep off account-moochers. Though such algorithms have not taken mass effect, there is reason to believe that this might change soon.
Lastly, big data will be transformative when it comes to pricing streaming services. As companies compete for a higher share of users’ e-wallets, data on how much the consumers are willing to pay will be transformative in determining how the service is priced. The marginal cost of adding an additional user to a streaming service in negligible, which means that the price they can be charged is relatively flexible as compared to traditional industries such as cars. This is exactly what Netflix has been trying to leverage in India. In July, the company unveiled a mobile-only plan for the price-sensitive market. It is a novel move that might help Netflix compete with Prime Video, Jio TV, and Hotstar in India, all of which are cheaper options. The same could hold true for markets where the consumer is willing to pay more for premium services. Data will decide.
User experience, security, and pricing are three key areas where big data analytics could be transformational for the streaming industry. This is by no means an exhaustive list. It would have taken a mental leap ten years ago to conceive of the current streaming scenario, and we might find the same a decade from now. New applications of insights from big data will continue to come to light. And the interesting thing is that, for us who are now aware of the speed at which data engineering and digital ecosystems can evolve, none of these developments are situated too far off in the future to be imagined. In big data analytics, the enabler of the present is also the driver of the future.
