Key Takeaways
YouTube's Q4 earnings highlighted quarterly revenues greater than Netflix's, but slowing ad growth
YouTube offered creator economy, NFT and live shopping as potential models for growth
But, the underlying engine of growth may be market adoption of the AV1 codec, a core issue of last year's dispute with Roku
In other words, the AV1 codec may be both a solution for slowing ad growth and a necessary condition for these new models to succeed
YouTube's aggressiveness on AV1 suggests that those who solve for and innovate streaming business models around AV1 will adapt and survive.
Yesterday Google announced big Q4 2021 earnings for YouTube advertising - $8.63B or 10% greater than Netflix's Q4 earnings of $7.71B (though YouTube ad revenue growth has slowed over the past three quarters).
There are key, post-"streaming wars" market trends that are playing out in YouTube's quarterly earnings:
Creator Economy
Non-Fungible Tokens (NFTs), and
The Evolution of OTT Streaming technology
All three have played out in key market signals over the past few months:
Hollywood-Meets-Creator-Economy (which I wrote about two weeks ago in Why YouTube Sees Hollywood’s Future in the Creator Economy)
YouTube CEO Susan Wojcicki's "Letter from Susan: Our 2022 Priorities"; and
YouTube's eight-month-long, ugly standoff with Roku (which I wrote about in May 2021 in Roku, YouTube & Product Channel Fit, and which was resolved in early December.
They are each and all seem to be very different signals, but they are all related. The challenge is how to contextualize them.
A recent interview with EA COO Laura Miele on Variety's Strictly Business indirectly helps us to do so. Miele was asked by Variety's Cynthia Littleton about whether given "the future is very much online and streaming on demand, do you think... the hardware world of video gaming with set top boxes... do you think it's all just going to come through a stream in 10 years, 20 years time?"
She responded:
Possibly. However, again I think about this idea of "yes, and". I think the idea about streaming and cloud is about accessibility and playing anywhere. One of the things I love about our industry is that we are a blockbuster entertainment business still, and it's so fun and it's incredible, and just like the movie industry, you could download pretty much any movie to your TV or your iPad...
Let alone sitting in a movie theater experiencing the multichannels of sound and special effects and characters larger than life. And so, I would like to think in our there's potentially always going to be a dimension of our industry of this blockbuster feel and experience, and you know, hardware certainly provides that processing power locally for the experience.
But to your point, 20 years from now? You know, based on infrastructure and that there's a world where the processing and the graphics and everything are better served by streaming and cloud that will come into the living room. But, I look at our industry in innovation and tech, as I'm sure you do too as you're thinking about entertainment, as kind of in five year increments. Because you have to pace yourself about what the possibilities are. So I think it will be a combination of those for a little while.
Miele helps to frame why the Roku-YouTube stand-off was so heated: YouTube is a juggernaut with slowing ad growth, and its ability to deliver content from the cloud to 56MM Roku devices will impact its ability to find future growth models. Also, its possibilities for innovation are constrained by this dynamic.
So, YouTube was implicitly telling the market that Roku had become an obstacle to its growth because it had not adopted the AV1 codec, which allows for streaming content at 30% lower bitrates than the H. 264 codec or the H. 265/HEVC codec (both are also current market standards for streaming in higher resolution).
Most (but not all) Roku hardware did not support AV1, nor do most smart TVs, and new hardware is required to do so. YouTube needed it for 4K streaming on YouTube TV and the YouTube app.
Without a faster, mass adoption of the AV1 codec, YouTube may fear it will miss the opportunity to serve a new generation of mobile and CTV consumers.
That will have negative implications for its various creator economy initiatives, which need adoption of the AV1 codec to drive further growth because that model relies heavily on growing ad revenues because creators get 55% of those revenues. Without wider AV1 adoption, the possibilities for its YouTube business model ahead will be fewer.
Creator Economy
CB Insights has a good description of the creator economy:
The creator economy refers to the numerous businesses built by independent creators, from vloggers to influencers to writers, to monetize themselves, their skills, or their creations. It also encompasses the companies serving these creators, from content creation tools to analytics platforms.
CEO Sundar Pichai told investors on the Q4 earnings call "more creators than ever are earning money from our non-ads products like Super Chat and channel memberships".
We last had learned that over 140K channels earned money from Paid Digital Goods on YouTube in December 2020, an increase of over 140% since December 2019. Creators are "going to be an integral part" of YouTube's future.
On the earnings calls Pichai highlighted that the number of YouTube channels that made at least $10,000 revenue was up more than 40% year-over-year,
As I wrote in Why YouTube Sees Hollywood’s Future in the Creator Economy, YouTube has become perhaps the most important engine for the creator economy:
...YouTube talent is moving closer to Hollywood’s model than Hollywood is moving toward YouTube’s. That has transformed YouTube from an aspiring Netflix and Hulu competitor—the objective when it launched YouTube Red in 2015—into something more analogous to the creator economy version of a mega-agency like the Creative Artists Agency.
YouTube now has over two million creators in its YouTube Partner Program (for monetizing content). YouTube also has created a $100MM Shorts Fund that is now available in over 100MM countries.
The 2022 letter from CEO Susan Wojcicki laid out shorts and live shopping as two key elements of the future of YouTube:
We’ve introduced a creator tagging pilot program that gives viewers the chance to browse, learn about, and shop products featured in their favorite videos. We're also in the early phases of testing how shopping can be integrated with Shorts.
Pichai made sure to highlight that YouTube is "super early" in its tests of these newer creator models, but Pichai finds "the opportunity space here pretty broad, and it’s exciting."
Non-Fungible Tokens (NFTs)
A non-fungible token is a non-interchangeable unit of data stored on a blockchain, which is a form of digital ledger.
Wojcicki also wrote in her letter that YouTube was experimenting with NFTs in the YouTube ecosystem:
We’re also looking further ahead to the future and have been following everything happening in Web3 as a source of inspiration to continue innovating on YouTube. The past year in the world of crypto, nonfungible tokens (NFTs), and even decentralized autonomous organizations (DAOs) has highlighted a previously unimaginable opportunity to grow the connection between creators and their fans. We’re always focused on expanding the YouTube ecosystem to help creators capitalize on emerging technologies, including things like NFTs, while continuing to strengthen and enhance the experiences creators and fans have on YouTube.
I wrote about NFTs in the creator economy last Friday in Where Web3 & Legacy Media Overlap, and how the Bored Ape Yacht Club (BAYC) model hits four out of five attributes of the PARQOR Hypothesis:
If one agrees with the PARQOR Hypothesis and the argument laid out in the IAC Letter to Shareholders that inspired it - as I still do, and strongly - then BAYC suggests that the optimal media business model no longer needs scale as a necessary condition. In other words, making Exclusive membership the second attribute, while holding all other four attributes of the PARQOR Hypothesis static, may be the optimal Web3 model for media.
That is a significant takeaway at a time when the creator economy is beginning to eat into Hollywood's model, as I argued in an opinion piece for The Information last week, Why YouTube Sees Hollywood’s Future in the Creator Economy.
The challenges are Aspirational brand - are there any creators who offer that? - offline sales, and the ability to produce IP that attracts a user base. Ironically, all three remain strengths of legacy media over the creator economy.
The implication is that YouTube may be unusually well-positioned to drive Web3 innovation because the value proposition of NFTs and DAOs like BAYC is basically an iterative improvement to - if not a more exclusive version of - YouTube memberships built upon blockchain technology.
It would not be unreasonable to imagine that YouTube is trying to figure out how to help creators to build their own versions of BAYC that would work across mobile, desktop and CTV.
The Evolution of OTT Streaming Technology
A key point about these two growth models is that, as per Pichai, the creator economy is "an integral part" of YouTube's future. But, other aspects like live shopping and NFTs are still possibilities and not actual growth drivers yet.
The question, then, is why YouTube needs mass adoption of the AV1 codec for these models and more, and why now.
To date, YouTube has been the most aggressive service pushing for the standard industry-wide, including on Android TV devices. There are between 70MM and 80MM Android TVs worldwide.
We have not heard much of anything about AV1 being pushed by other streaming services. Obviously, none have the same market heft as YouTube. But, Netflix does, and it has been less bellicose on this issue in the OTT space despite recent data suggesting subscriber trends are flatlining.
To date, it has primarily relied on Android for rolling out AV1, and there are over 3B active Android devices worldwide. As Netflix wrote on its TechBlog back in 2020:
we see a good fit for AV1’s compression efficiency in the mobile space where cellular networks can be unreliable, and our members have limited data plans.
But, it also added:
As codec performance improves over time, we plan to expand our AV1 usage to more use cases and are now also working with device and chipset partners to extend this into hardware.
A recent Netflix TechBlog post from November 2021 outlined its approach to AV1 in Smart TVs. Key advantages from AV1 it had learned from testing included:
Higher visual quality across the full spectrum of streaming sessions
More streaming at the highest resolution,
Fewer noticeable drops in quality during playback, and
Reduced start play delay.
The Netflix post focuses on Smart TVs, but the Roku-YouTube standoff also highlighted how OTT devices are also not up-to-date. [1]
Roku is a necessary channel for distributing YouTube apps given its scale: 56MM households, and ~40MM estimated in the US. But, as Protocol's Janko Roettgers wrote back then in What Roku and Google are fighting about: Video codecs, voice search and millions of eyeballs, Roku is also implicitly an expensive channel for YouTube [2]:
Roku is particularly unhappy that Google has been leveraging the negotiations to force an upgrade to the AV1 codec, an open-source video codec that promises better-looking 4K videos at lower bitrates (basically, more efficient streaming). Roku is telling the marketplace that it is not incentivized to pass on to consumers the additional costs that the AV1 codec would require, but it would be more incentivized to do so if Google would agree to change the terms of its separate YouTube deal with Roku.
Google is saying it is not incentivized to revisit those terms.
CTV is YouTube's fastest-growing screen, according to Chief Business Officer Philipp Schindler on the Q4 earnings call. But, as both the Roku-YouTube and Netflix disputes highlight, CTV is also the slowest to upgrade to AV1.
The implication is that without wider adoption of the AV1 codec across mobile and CTV devices, YouTube may not be able to support its growth.
Part of it is cost: AV1 would drive the marginal costs of distributing content down because the higher compression efficiency from a lower bitrate stream results in streaming relying on less bandwidth.
Also, AV1 is open source while Panasonic has over 1,000 patents related to H.264, and Samsung has over 4,000 patents related to H.265. So there are additional licensing fees attached to those codecs.
With YouTube being the only player publicly pushing for greater AV1 adoption, we don't know where various mobile and TV manufacturers stand in their hardware development cycles for AV1 adoption.
AV1 & 1B Streaming Subscribers
Both the Laura Miele quote and the AV1 codec shed a different light on why WarnerMedia CEO Jason Kilar told Bloomberg's Lucas Shaw he thinks there is a "ceiling" of 1B users for an SVOD service like Netflix or HBO Max:
“If there is a ceiling, and everything has a ceiling … it will be closer to 1 billion vs 200 million and change. I’m not commenting on who gets to that number but I don’t think folks truly appreciate yet just how much value is present in a great streaming service.”
Kilar believes only three major streaming services can compete in the current market: Netflix, Disney and HBO Max. That list expands to include Apple and Amazon because entertainment is not a primary business for them. But Paramount+? Peacock? “There will need to be consolidation, there have to be changes,” he said.
Meaning, I think Kilar is making an implicit point about where the technology for streaming will be in five to ten years: the AV1 codec is driving a technological shift that opens more use cases and more possibilities in streaming that did not exist before.[3]
But he is also implying that we are very early in this shift, which is why I think it is a nod to the adoption of the AV1 codec being a key step. I also wonder whether he is implying that the streaming business models which will reach 1B subscribers are also going to be very different from the models currently being pursued in the "streaming wars". If he is, it's a bearish take on legacy media's ability to adapt.
Conclusion
Notably, Google did not discuss AV1 codec adoption in its Q4 earnings call. But it has revealed its hand in having been unusually open and aggressive about its needs for Smart TVs, OTT devices and mobile phone hardware to be able to support AV1. In some ways, it tells us that creator economy models, NFTs, Shorts and live shopping may not be the key drivers of YouTube's future growth.
Instead, the implication is that but for AV1 adoption growing across mobile and TV devices, the continued growth of YouTube and all other streamers is less guaranteed.
Creator economy models and NFTs offer the foundations for a very different relationship between audience and content. But YouTube seems to be telling the marketplace that the future of these models will be predicated on AV1 and lower bitrates driving more accessibility - and therefore reaching and engaging more audiences - than their previous business models.
With its dominance in Connected TV, YouTube appears to be on the forefront of rethinking how "everything [can be] better served by streaming and cloud that will come into the living room", as per Laura Miele, above.
This all would imply the AV1 codec will be driving a quiet revolution for the next five years of streaming business models. YouTube's aggressiveness on pushing adoption suggests that those who solve for and innovate streaming business models around AV1 will adapt and survive, and those who do not will not survive.
Footnotes
[1] Roku was pushing back, in part, because the economics of adopting the AV1 codec would also making many devices obsolete due to outdated hardware to support the chip, and more expensive economics to produce Roku devices with the new chip. But YouTube needs more favorable economics in streaming, especially as it grows on Connected TVs (120MM connected TV viewers in December 2020).
Notably, we do not know if YouTube is profitable.
[2] I spoke to Roettgers for this piece and he highlighted an interesting implicit competitive dynamic at play with Chinese OEM TCL, too.
Both Roku and Google have deals with TCL to produce Smart TVs. However, Google is paying a percentage of ad revenues to keep its costs down, while Roku says it offers cheaper materials for TCL TVs. So, if YouTube's demands for the AV1 codec would require a new chip that would drive up TCL's unit costs of producing Roku Smart TVs, TCL may be better incentivized by Google to adopt its Chromecast with Google TV OS over Roku's.
So, this is not only about the AV1 codec.
[3] Kilar's math may have another source: in 2020, the World Bank found that there were 1.202B fixed broadband connections worldwide. Excluding China (483.55B), that totals 718MM, which falls right in the range that Kilar projected.
Assuming 5% growth year-over-year for the next five years, as per Miele's five-year window, there will be 1.534B fixed broadband connections, of which 617.15B will be China, leaving 917B fixed broadband connections. This may be the number he has in mind.