5 Companies Making the Most of Recommendation Systems

Using recommendation systems may have been a thing of complexity and even luxury for companies in the past, but now it has rather become a necessity to most. While it has revolutionized e-commerce and video/music streaming services, many other fields also seem to be making the most of this technology. These days, recommendation systems are used for personalization in many sectors such as healthcaree-learning, nutrition, marketing, and a lot more.

Having said that, some companies have set an example of how recommendation systems can be utilized to a greater extent. Let’s take a look at 5 such companies.

1. Netflix

Netflix’s recommendation system is one of the best ones out there. While it is hard to believe, before it was big, Netflix was a DVD rental service that started in 1997. Their business started to go at a loss in the early 2000s because they were facing one major challenge: People were demanding for more and more new and superhit movies to be available for rental, which resulted in customers not being happy when they couldn’t get it.


Netflix’s old UI (view source)

As a solution, Netflix came up with an idea: a recommendation system. They started to recommend people older movies that they might like, and have not watched yet. As a result, the demand for newer, superhit movies went down by 20% as compared to their competitor at that time- Blockbuster (full case study here).

The recommendation system they initially used was called “Cinematch” and in 2006, they set up a challenge: Anyone who could improve Cimematch by 10% would be awarded 1 Millon USD. Since then, they have constantly been improving themselves and now it is one of the best recommendation systems out there.

2. Amazon

The use of recommendation systems in e-commerce is not a new concept, but Amazon has some of the best ones out there, and one of the pioneers in this field.

Amazon started item-based collaborative filtering in 1998, and there has been no going back! Amazon has a highly personalised store for each one of its users. Everyone sees different products when they enter the Amazon UI. And it surely has done wonders for the company, as now it has become the biggest e-commerce company in the world, surpassing other e-commerce giants like Alibaba. And, the highly powerful recommendation system contributes to 35% of its revenue!

Amazon’s old UI (view source)

Founded in 1994 as an online book store, Amazon has truly come a long way, and its recommendation system can be highly credited for making that journey possible.

3. Tinder

Unlike Amazon and Netflix that later incorporated recommendation systems into their existing application, Tinder was an app designed to reap the benefits of this technology. Targeting only college students at first, Tinder established itself as a dating app that selected partners for you based on your likes and dislikes.

Tinder has quite an interesting recommender system to match people. The key factors that contribute to it are the people that you swiped right on (the people you liked), your reactivation stats, your location, your profiles, and more. But this was just the tip of the iceberg for Tinder.

Now, it keeps updating itself and keeps adding more advanced recommendation strategies. For example, Tinder had an issue that 1% of the people got a ton of matches more than others, and made all others look bad in comparison. To solve this, they started showing those profiles less often and changed the trend, this was termed as “smart matching”.

Tinder is one of the biggest recommender systems in use. It is estimated that over 50 million people use Tinder worldwide in 2020.

4. YouTube

YouTube started in 2005 as a video-sharing platform and now it has turned into one of the most used applications around the world, creating a lot of new jobs and being a pioneer in content creation.

Youtube’s old UI (view source)

It is a well-known platform that streams videos, but what would it be without its recommendations? We all can relate to opening youtube for something and then four hours later you are still clicking on the recommendations it provides you. Maybe it wouldn’t have achieved the success it has today if it had chosen a different business model early on.

Every user in the app has a different list of videos sorted out just for them. The youtube recommendation system is quite complex and it uses Deep Neural Networks to provide fresh and relevant content to every individual user. The company even faced a challenge to develop its recommendation system because it had to deal with the biggest dataset of user behaviors as well as the amount of content in it.

5. Facebook

Facebook has over 2.24 billion users worldwide. Needless to say, it is an amazing feat to be able to recommend personalized content to each of those users every single day. Also, needless to say, is what a challenge that must be!

Facebook’s old UI (view source)

Facebook uses not only one recommendation algorithm, but several different ones for different sections. For example, the news feed uses one whereas the “people you may know” section uses another. Similarly, the news section, marketplace, Facebook videos, etc are different sections of Facebook, each of which will recommend different things based on your preferences.

*Caboom is a Leapfrog product.
*Also published at caboom.ai.

Drishya Bhattarai

Drishya is a Digital Marketer at Leapfrog. She is interested in SEO, digital marketing, content creation and Designing.

More in Blogs

Using FAISS Advanced Concepts to Speed Up Similarity Search in Recommender Systems  (Part II) Artificial IntelligenceInsights

Using FAISS Advanced Concepts to Speed Up Similarity Search in Recommender Systems (Part II)

Continued from the last blog. Improving on Flat Indexes Beyond the flat indexes that perform exhaustive searches, FAISS also has

Read more
Using FAISS basics to Speed up similarity search in Recommender Systems (Part I) Artificial Intelligence

Using FAISS basics to Speed up similarity search in Recommender Systems (Part I)

Measuring the similarity between vectors or arrays is a calculation that we encounter often while developing recommendation systems or other

Read more
Why is a No-Code Recommendation System worth a try? Artificial IntelligenceInsights

Why is a No-Code Recommendation System worth a try?

Lately, in the tech world, no-code systems have been getting immense attention, and for good reasons. These types of systems

Read more