Netflix Tests the Use of People to Suggest Offers to Users in the “Collections” Feature
How do platforms like YouTube and Netflix choose which videos to suggest? This is largely based on algorithms that take into account videos you’ve watched in the past and try to recommend other videos and movies that you think have the same context. How accurate are these recommendations? Sometimes those recommendations can be good, but is this necessarily the best way to do it?
Probably not, because Netflix seems to have decided to start testing human-proposed content with a feature called Collections | Groups. ” In a way, this will be similar to the current recommendation system where users are recommended to see new offers. However, videos are grouped into different categories like “Let Let’s Keep It Light”, “Dark & Devious TV Shows” or “Prizewinning Movie Picks”.
This looks similar to other recommendation systems used by other platforms such as Spotify, since instead of being gender-based, they are based on mood or category. Netflix has confirmed that it is already testing this feature, although the company has not yet reached the point of confirming that it will become a permanent feature.
In a statement to TechCrunch, a Netflix spokesman said: “We are always looking for new ways to connect our fans with offers we think they will love, so we are testing a new way to group Netflix offerings into groups on the Netflix app for iOS. Our tests generally vary in how long they take and in the countries in which they take place, and may or may not become permanent features in our service. ”