As TV streamers expand, an endless supply of TV titles are setting roots in streaming “forests" that are increasingly opaque and hard to decipher. Luckily, at Parrot Analytics, we use machine learning (ML) to provide visibility into the preferences and taste of billions of global consumers who, on a daily basis, express their demand for content when they search for their favorite show on Google or Wikipedia, watch a trailer on Youtube, make a comment on social media platforms, or even stream it from peer-to-peer networks among many other online activities that leave a trace that ML models ingest to actually let investors and TV executives see the forest for the trees.
Training random forest ML models with Parrot Analytics’ exclusive dataset and Netflix's "net adds” (net additional paid subscribers) in the US, provides quantitative answers to some industry questions, including:
• Are originals more important than catalog? Or are licensed libraries more important than original content? • Are titles that deliver consistent demand preferable over tentpoles? • Lastly, how many “net adds” will Netflix report in their upcoming earnings release on July 16th 2020?
