Nielsen Gracenote New Insights into the Nature of Global Streaming

Nielsen Gracenote’s latest report on streaming, the 2024 State of Play Report, covering 23 countries, provides some revealing perspectives into where we are now. It focuses on the video libraries powering the industry’s leading subscription video-on-demand (SVOD) services: Amazon Prime Video, Apple TV+, Disney+, Netflix and Paramount+. Collectively these five companies have a total of 552,000 programs and movies in distribution, 88% of which are television series episodes. Looking at it from the standpoint of titles (counting all episodes of a series as 1 content chunk) the total comes to 84,200 titles, of which two-thirds are movies. (Some of the findings in this article come from the Gracenote Data Hub.)

The first pages of the report contained a specific factoid from a Nielsen survey that jumped out at me: 63% say they subscribe to SVOD streaming services to watch old TV shows and movies. In other words, the depth of catalog is a major driving factor, it’s not just the expensive heavily-promoted originals. The report makes these points:

• While new shows can, and often do, drive viewership spikes, it’s never a guarantee. The first season of Citadel, for example, cost a reported $250 million-$300 million to make, only included six episodes and never appeared on Nielsen’s weekly top 10 list of most-watched streaming programs. Big budgets aside, a six-episode season isn’t capable of engaging audiences for long periods of time. This is where trended viewing behavior highlights the foundational value of having a deep catalog of dependable favorites that keep audiences engaged outside of those intermittent windows when originals drop.
• The extensive libraries of long-standing TV franchises with 20+-episode seasons like Grey’s Anatomy, NCIS and Law & Order: Special Victims Unit provide audiences with significantly more meat on the proverbial bone than newcomers like Severance, The Bear and Reacher, which only feature eight to 10 episodes per season. In the U.S. last year, original programming accounted for just 26% of U.S. audiences’ total time with streaming video content (source: Nielsen Panel).
• We saw similar trends in 2022, which many regard as the apex of original content on streaming services. While Squid Game, Ozark and Wednesday attracted big audiences and inspired wide-reaching headlines, U.S. audiences still spent 20% more time with the top 10 most-watched licensed titles, including the ones noted above, than they did with the top 10 most-watched originals for the year (source: Nielsen Panel).
• The importance of library depth and global representation represents a much different reality from a decade ago, when streaming was just starting to appeal to cord-cutters with on-demand, binge-in-a-weekend originals. Today, sustained business viability is much more dependent on catalog diversity to maintain viewers, especially those who may subscribe to a service for a specific show with the intention of canceling shortly thereafter.
• Library composition has also become more global, especially within the Netflix and Amazon Prime Video catalogs: Only 32% of Netflix’s SVOD catalog and 45% of Amazon Prime Video’s SVOD catalog were produced in the U.S. Across all five catalogs, only 44.2% of the content was produced in the U.S.

In my conversations with broadcast networks recently I’ve heard many comments about what is working and not working on broadcast TV. Sports is the big driver of course and news is steady, but the nature of entertainment on broadcast may be moving into an era of change.

There is discussion about lower budget programming such as reality and talk shows, and even the growing belief that expensive dramas which unfold serially into a big story may no longer be the way to go on broadcast. What is perceived as working on broadcast are stories which resolve in one episode, whereas streaming is where the longer continuing stories fit best, possibly because bingeing on SVOD series allows the viewer to pick up the story whenever they feel like it, as with a book.

I hope that the networks do not stop aiming for top-quality innovative drama and comedy. The changes in the motion picture industry in recent decades have not been positive ones from the standpoint of continuing the quest for topnotch content, and for television to follow in that path based on business calculations would be disappointing to those of us who love the high points attained by the best in entertainment and the arts.

It occurred to me recently that the networks in the beginning were mostly about the creation of new programming, with their station affiliates being the distribution partners. Cable and satellite then became important parts of the distribution. Today the networks are trying their hands at direct-to-consumer distribution through streaming, but are finding the straddle across creation and distribution is becoming harder to optimize financially. 

To bolster their streaming network, they put some of their best content exclusively on their own streaming network, and license everything else they have non-exclusively to as many other platforms as possible. In January 2-8, 2023, all of the top ten SVOD streaming TV shows (by minutes viewed) were exclusive to a single platform; by January 1-7, 2024, half of the top ten were licensed to two streaming networks, and by June 3-9, 2024, 6 of the top 10 were licensed to multiple streaming networks, 2 of those 6 shows licensed to three different streaming networks. Of the five largest libraries, Apple TV+ has the highest proportion of exclusivity at 98.5%, Paramount the opposite strategy at 64.2%.

They also have the option to consider, of going for the gold in creation and being able to license broadly all of the content they create, giving up having their own streaming network. The unpredictability of the success of new content has always been so high that most practitioners would vote against giving up distribution. However, the increase in the ability to scientifically predict and maximize content success, through AI approaches such as my own RMT approach as well as others, could make a difference in the near future.

Gracenote has been developing its own psychological approaches, and now has 315 genres in its metatagging scheme, as well as 300 mood descriptors. Across all content types, the mood which has the highest incidence across these five catalogs is Tense at 37.9% of content, while Thoughtful is the most prevalent mood in TV programs. Here are the genre skews of the top five libraries:

• Among the global SVOD providers, Amazon Prime Video distributes the greatest amount of content, but at the genre level, it offers proportionally less sitcom, children’s, animated, and anime content than the others. Given its deep catalog of classic, animated programming, it’s not surprising that Disney+ has much higher proportions of children’s and animated titles. But it also has higher proportions of fantasy and adventure programming.
• Netflix offers a greater percentage of fantasy programs, while Disney+ has greater percentages of anime, children’s and animated programming than the average across the five global providers. Paramount+ has high representations of sitcom, reality and crime programs. Apple TV+ does not distribute any programs in the reality, soap or anime genres, but a significant percentage of its programming has animated, children’s and mystery genres.

Apple TV+ is also the most concentrated in U.S. productions of the five players at 86.6% U.S.-produced, while Netflix goes the other way at 31.8% U.S.-produced.

A recent Google study, for example, found that 48% of streaming subscribers have canceled a service because they couldn’t find something to watch. In the U.S., for example, a recent review of time spent with SVOD content during 2023 found that nearly 3,000 programs had zero minutes of viewing (source: Nielsen Streaming Meter).

These Nielsen and Gracenote stats prove that the program discovery function still has not been optimized for users. In response to this need, Gracenote now offers a service called Personalized Imagery which has been proven to cause an increase of 11.2% in hours watched on a specific streaming network, and a 7.7% increase in number of different programs watched on that streaming service.

In 1997, the forerunner to RMT, my company at the time Next Century Media showed that an early AI personalized program recommender using the machine learning empirically-derived metadata which today is the basis for RMT, was able to increase program adoption by 18%. These types of content development and marketing platforms are just at their dawning and will have very positive impact on the financial strength of the content business.

Posted at MediaVillage through the Thought Leadership self-publishing platform.

Click the social buttons to share this story with colleagues and friends.
The opinions expressed here are the author's views and do not necessarily represent the views of MediaVillage.org/MyersBizNet.