Recommendation Algorithms: The Case of Netflix

One of the topics that interests me the most is the application of data analytics, particularly machine learning, to marketing. Without a doubt, one of the best applications I have seen and I am sure has a direct impact to increasing sales and benefit the long tail is the use of recommendation algorithms. As per research from McKinsey and Company (https://www.mckinsey.com/industries/retail/our-insights/how-retailers-can-keep-up-with-consumers#) 35% of Amazon’s sales is driven by recommendation algorithms and 75% of what Netflix users watch is driven from them.

Drawing from past consumption data in addition to similar consumption patterns from other users, companies such as Amazon, Netflix or Instagram have seen a huge spike in sales largely due to significant investments in recommendation algorithms. One particular case that comes to mind is the Netflix Prize, a competition conducted in 2009 by Netflix seeking to find the best collaborative filtering algorithm. The participating team that more accurately predicts user ratings for a film won a prize of $1 million.

Netflix not only uses their algorithms to recommend movie titles that you may like, but even helps them select the preview image that can be seen for every film on your feed, tailoring it to one that you may like. For instance, if you rate and watch food documentaries, in unrelated movies you might see in the preview image contents of food or actors having a meal. Netflix is a leading company in personalized marketing, and has used its recommendation algorithms to also decrease churn rates and increase viewership.

There is a growing demand and investment in data analytics for business and marketing. I believe that in the coming years these algorithms will become even more sophisticated and will have direct effects in increasing click rates and conversions without users even realizing it. I believe that as consumers, we must be aware of such capabilities that companies have and not fall into the trap of overconsumption, binge watching or impulse buying.