Can algorithms change the way we watch TV?

Netflix is among one of many streaming services seeking to influence our viewing habits through technology

Getting over genre bias in no easy feat, but new technology is ensuring that – for television at least – our predilections are starting to recede.

On-demand viewing services are accelerating at breakneck speed across the world and, though the Middle East still has a hefty contingent who are still using free-to-air TV services – the gap is narrowing. The region saw Netflix being introduced in January of last year, adding to a local rosta of on-demand services that includes Starz Play, icflix, and OSN Go, to name just a few.

The benefit of using a streaming service has been touted as absolute freedom. Viewers could watch whatever they wanted to watch, whenever they wanted to watch it, without having to endure the regular monotony of advertising. However, an unexpected benefit has started to appear: that of genre blindness.

Netflix’s newest show The Defenders centres around the lives of Marvel Comics characters Daredevil, Jessica Jones, Luke Cage and Iron Fist, who form a team of superheroes.

Working to upend a natural bias against comic book-based content, Netflix uncovered interesting patterns that act as what they call a “bingeing gateway” to the Marvel Universe .

For example, viewers that favoured content featuring anti-heroes and moral ambiguity were led from favourite shows like House of Cards or Breaking Bad to the character of Daredevil – whereas series with smart humor like Master of None were intended to draw viewers to Jessica Jones (as did stories with strong female leads, such as Orange is the New Black).

Below, Todd Yellin, Vice President of Product Innovation at Netflix, describes how algorithms can both predict and widen our viewing tastes. 

todd yellin
Todd Yellin, Vice President of Product Innovation at Netflix

How do Netflix algorithms generate suggested content for viewers?

Personalisation is about creating the right connection between a viewer and their content. To do that, we have to understand everything there is to know about the content. Genres are just one way to categorise a show, but they aren’t complete, which is why we also tag series and movies with thousands of additional qualifiers - from mood (goofy) to aesthetic (visually striking) to pace (slow pace) and beyond.

This “tagging process” is just one stage of the Netflix recommendation process. We also have to know about the viewer. For that we use advanced algorithms to examine the viewing habits of millions of members around the world along with their specific taste preferences and viewing histories. The end result is how we present the catalog uniquely to each member, bubbling to the top of the experience titles that are both relevant and diverse.

Taking the Defenders as an example – how will the algorithms help the success of this show?

One of the great things about personalisation is that we are able to present members with content they may not have thought to search for on their own. We know viewers tastes are varied, which is why we strive to provide diverse recommendations. For Defenders the series will not only be a likely suggestion for those who’ve watched one of our Marvel shows, but also those who enjoy dark crime dramas, ensemble casts, suspenseful action, and so on.

How is data on behaviour is taken – and does this data inform Netflix’s commissioning decisions?

One thing to note, the data we get from viewing habits and behaviours is never made available to external parties - we don’t have advertising on the platform so we don’t need to. Instead we use it to improve the viewing experience for our members, from personalisation to improving the user interface. Whilst data supports us in how we place commissioning bets, it’s never the sole factor in the decision making process. We have a talented team, led by Ted Sarandos, who lead our content commissioning charge.

As it is now in 190 countries, to what extent does Netflix customise how viewers see content from country to country? Similarly, do large regional markets with different viewing tastes from North America influence your commissioning decisions?

We personalise at an individual/profile level, not at a country level. There is no “average” Netflix member: every single one is unique and tastes aren’t confined to geography. There are many diverse tastes on the platform and we need to offer a completely different experience to everyone to satisfy their needs. This is why we continue to redefine the art of personalisation.

We also know that great storytelling also transcends geography. Look at Narcos, a Dutch production company:  a show about a Colombian drug lord, played by a Brazilian actor, with over 70 per cent of the dialogue in Spanish was enjoyed by viewers across the globe! Through the power of internet TV, international audiences are being given access to content from around the world.