Stratechery Plus

Subscribe

Learn More
Member Forum

Menu

Stratechery Plus

Subscribe

Learn More
Member Forum

Latest Podcast

Stratechery Plus Update

  • Where Warren’s Wrong

    Senator Elizabeth Warren deserves credit: I have been writing about antitrust, particularly in the context of Aggregation Theory, for years, but the most concrete proposal I have put forward is that social networks should not be allowed to acquire other social networks. Senator Warren, on the other hand, last week presented a far more wide-reaching proposal that specifically targeted Facebook, Google, and Amazon:

    Today’s big tech companies have too much power — too much power over our economy, our society, and our democracy. They’ve bulldozed competition, used our private information for profit, and tilted the playing field against everyone else. And in the process, they have hurt small businesses and stifled innovation.

    I want a government that makes sure everybody  — even the biggest and most powerful companies in America — plays by the rules. And I want to make sure that the next generation of great American tech companies can flourish. To do that, we need to stop this generation of big tech companies from throwing around their political power to shape the rules in their favor and throwing around their economic power to snuff out or buy up every potential competitor.

    That’s why my administration will make big, structural changes to the tech sector to promote more competition — including breaking up Amazon, Facebook, and Google.

    Senator Warren added Apple in an interview at SXSW with The Verge:

    There was one company that fits that description that you did not mention.

    Apple. They’re in.

    You want to break up Apple as well.

    Yep.

    You were very specific about how you’d break up Google and the rest. How would you break up Apple?

    Apple, you’ve got to break it apart from their App Store. It’s got to be one or the other. Either they run the platform or they play in the store. They don’t get to do both at the same time. So it’s the same notion.

    Unfortunately, Senator Warren’s proposal helps highlight why I have not gone further with my own: hers would create massive new problems, have significant unintended consequences, and worst of all, not even address the issues Senator Warren is concerned about (with one possible exception I will get to in a moment). Worst, it would do so by running roughshod over the idea of judicial independence, invite endless lawsuits and bureaucratic meddling around subjective definitions, and effectively punish consumers for choosing the best option for them. Mike Masnick at TechDirt gets into many of these problems, and concludes:

    This entire plan gets headlines (duh) because so many people are (perhaps reasonably!) angry at the power of big tech companies. But, very little in the actual plan makes much sense. The “platform utility” idea will lead to massive, wasteful, stupid lawsuits. The unwinding of old mergers will involve interfering with an independent agency, and seem unlikely to do much to change the main “concerns” that Senator Warren raises in the first place.

    And, again, none of this is to say we shouldn’t be concerned about big internet companies with too much power. It’s a perfectly reasonable concern, but just because you want to “do something” and “this is something,” doesn’t mean that it’s the something we should do.

    I do know what is the first thing Senator Warren should do: rectify three clear areas where I believe she is mistaken about technology. Her proposal is wrong about tech’s history, the source of the tech giants’ power, and the fundamental nature of technology itself. All three are, unsurprisingly, interrelated, and it is impossible to craft a cogent antitrust policy without getting all of them right.

    History: Microsoft and Google

    Senator Warren opens the article by crediting the Microsoft antitrust case for the emergence of Google and Facebook:

    Twenty-five years ago, Facebook, Google, and Amazon didn’t exist. Now they are among the most valuable and well-known companies in the world. It’s a great story  —  but also one that highlights why the government must break up monopolies and promote competitive markets.

    In the 1990s, Microsoft  —  the tech giant of its time  — was trying to parlay its dominance in computer operating systems into dominance in the new area of web browsing. The federal government sued Microsoft for violating anti-monopoly laws and eventually reached a settlement. The government’s antitrust case against Microsoft helped clear a path for Internet companies like Google and Facebook to emerge.

    The story demonstrates why promoting competition is so important: it allows new, groundbreaking companies to grow and thrive — which pushes everyone in the marketplace to offer better products and services. Aren’t we all glad that now we have the option of using Google instead of being stuck with Bing?

    Start with the most obvious error: Bing was not even launched until 2009, eight years after the Microsoft case was settled. MSN Search, its predecessor, did launch in 1998, but with licensed search results from Inktomi and AltaVista; Microsoft didn’t launch its own web crawler until 2005 (these details will matter in a moment).

    What is more striking is that, in retrospect, the core piece of the government’s case doesn’t make any sense: of course a browser should be bundled with an operating system; a new computer without a browser would be practically useless (for one, how do you install a browser?). Moreover, Apple, not without merit, argues that restricting rendering engines to the one that ships with the OS (all browsers on iOS have no choice but to use the built-in rendering engine) has significant security benefits; this is debatable, but ultimately, most don’t care, simply because browsers are means to information, not ends.

    This, crucially, is something Microsoft did not understand in the 1990s; Microsoft’s operating system monopoly was predicated on owning the APIs with which applications were built, creating both lock-in and an ever expanding network effect. Unsurprisingly, Microsoft viewed the web through this exact same lens; that meant that Netscape was a threat because it was “middleware”, a potential means to run applications that were not locked into Windows. This is true, by the way — web apps work across operating systems and browsers — but this fact has absolutely nothing to do with the rise of Google. After all, when Google IPO’d in 2004, Internet Explorer had 95% market share; a browser was a means, not an end.

    The reality is that Google is an operating system of sorts, but the system is not a PC but rather the entire web; what ties things together are not APIs, but links. And, crucially, the business model that makes sense is not licensing, but advertising. This is a value chain that never even occurred to Microsoft, and why would it? The entire company was predicated on controlling operating systems for physical computers, controlling the APIs on top, and earning revenue through licensing; it was fabulously profitable, and as history shows again and again, being fabulously profitable with an existing value chain is the best way to not only fail to recognize a new market opportunity (Microsoft didn’t even have a web crawler until after Google’s IPO!), but to in fact be at a massive disadvantage when you finally do so.

    Look no further than mobile: Microsoft was not encumbered by antitrust when it came to their mobile ambitions, and yet they failed even more spectacularly there than they did online. In this case the company didn’t “miss” the opportunity — Windows Mobile came out back in 2000 — it was just stuck in a PC mindset when it came to product development, attached to its Windows licensing model when it came to monetization, and institutionally incapable of producing superior end user experiences thanks to the company’s traditional focus on platforms and compatibility.

    In short, to cite Microsoft as a reason for antitrust action against Google in particular is to get history completely wrong: Google would have emerged with or without antitrust action against Microsoft; if anything the real question is whether or not Google’s emergence shows that the Microsoft lawsuit was a waste of time and money.1

    Power: Google and Aggregation Theory

    Senator Warren’s second mistake is a misstating of why large tech companies are dominant. She writes:

    America’s big tech companies have achieved their level of dominance in part based on two strategies:

    Using Mergers to Limit Competition. Facebook has purchased potential competitors Instagram and WhatsApp. Amazon has used its immense market power to force smaller competitors like Diapers.com to sell at a discounted rate. Google has snapped up the mapping company Waze and the ad company DoubleClick. Rather than blocking these transactions for their negative long-term effects on competition and innovation, government regulators have waved them through.

    Using Proprietary Marketplaces to Limit Competition. Many big tech companies own a marketplace — where buyers and sellers transact — while also participating on the marketplace. This can create a conflict of interest that undermines competition. Amazon crushes small companies by copying the goods they sell on the Amazon Marketplace and then selling its own branded version. Google allegedly snuffed out a competing small search engine by demoting its content on its search algorithm, and it has favored its own restaurant ratings over those of Yelp.

    The merger issue is a real one, but only when it comes to propagating power; Facebook was dominant before it bought Instagram and WhatsApp, Google before it bought DoubleClick or YouTube, and Amazon before it bought Diapers.com or Whole Foods (I do share Senator Warren’s concern about acquisitions; I will return to this point). Notably, Apple has not made any major acquisitions other than Beats headphones, and that too came well after the company had created the iPhone.

    Similarly, the conflict of interest Senator Warren worries about is also post-dominance; none of Google, Facebook, Amazon, nor Apple achieved their power by “using proprietary marketplaces to limit competition”. That is not to say this, like acquisitions, isn’t a worthwhile issue, but it is flat out wrong to say that these are the reasons “big tech companies achieved their level of dominance.”

    Then again, perhaps it is best for Senator Warren’s argument that her article never does explain how these companies became so big, because the reason cuts at the core of her argument: Google, Facebook, Amazon, and Apple dominate because consumers like them. Each of them leveraged technology to solve a unique user needs, acquired users, then leveraged those users to attract suppliers onto their platforms by choice, which attracted more users, creating a virtuous cycle that I have christened Aggregation Theory. Specifically:

    • Google solved search, which attracted users; Google’s supply (web pages), thanks to the fundamental nature of the web, were already effectively “on Google”, but even then web pages have worked diligently to deliver content in a way that Google expects. Why? Because users start at Google — demand is what matters.
    • Facebook digitized offline relationships, which attracted users, which were both consumers and suppliers of content; professional content creators followed, not only linking to their content on Facebook but creating content specifically tailored for Facebook’s audience, making Facebook that much more attractive for users. Again, what mattered was demand, not supply.
    • Amazon leveraged the Internet to achieve a dominant strategy of offering superior selection and the lowest price, starting with books. This gained Amazon customers, which gave the company leverage to bring on first other media like CDs and DVDs, which gained them more users, and later goods of all types; Amazon then launched the Amazon Marketplace, through which suppliers could come onto Amazon directly. Why? Because that is where demand was.
    • Apple defined the modern smartphone, gaining users who were blown away by Apple’s first-party apps; that attracted app developers, who were soon clamoring for access to iPhone users. Apple closed that loop by creating the App Store, which attracted more users, which attracted more developers, etc. Critically, though, the users came first; one of Microsoft’s many mobile mistakes was believing it could effectively “buy” a supply of apps and thus earn users, but that doesn’t work in a world where owning demand matters most.

    Aggregation Theory is the reason why all of these companies have escaped antitrust scrutiny to date in the U.S.: here antitrust law rests on the consumer welfare standard, and the entire reason why these companies succeed is because they deliver consumer benefit.

    The European Union does have a different standard, rooted in a drive to preserve competition; given that the virtuous cycle described by Aggregation Theory does tend towards winner-take-all effects, it is not a surprise that Google in particular has faced multiple antitrust actions from the European Commission. Even the EU standard, though, struggles with the real consumer benefits delivered by Aggregators.

    Consider the Google Shopping case: Google was found guilty of antitrust violations in a case brought by a shopping comparison site called Foundem, which complained about their site being buried when consumers were searching for items to buy. This complaint made no sense, as I explained in Ends, Means, and Antitrust:

    If I search for a specific product, why would I not want to be shown that specific product? It frankly seems bizarre to argue that I would prefer to see links to shopping comparison sites; if that is what I wanted I would search for “Shopping Comparison Sites”, a request that Google is more than happy to fulfill:

    Screen Shot 2017-06-28 at 6.40.22 PM

    The European Commission is effectively arguing that Google is wrong by virtue of fulfilling my search request explicitly; apparently they should read my mind and serve up an answer (a shopping comparison site) that is in fact different from what I am requesting (a product)?

    There is certainly an argument to be made that Google, not only in Shopping but also in verticals like local search, is choking off the websites on which Search relies by increasingly offering its own results. At the same time, there is absolutely nothing stopping customers from visiting those websites directly, or downloading their apps, bypassing Google completely. That consumers choose not to is not because Google is somehow restricting them — that is impossible! — but because they don’t want to. Is it really the purview of regulators to correct consumer choices willingly made?

    As I noted above, there are some important points made here by Senator Warren; at a fundamental level, though, any sort of antitrust proposal that does not seriously grapple with the reality that the power of these companies flows from controlling demand — that is, consumer choice, willingly made — not from controlling supply, like monopolies of old, is going to be fundamentally flawed.

    Nature: What is Tech?

    This mistake by Senator Warren only came into focus with that interview where she included Apple as a target for her proposal. Here’s more from that interview:

    Pulling that apart, the App Store is the method by which Apple keeps the iPhone secure. It’s integrated into the platform. How would you propose that Apple and Google distribute apps if they don’t run the store?

    Well, are they in competition with others who are developing the products? That’s the problem all the way through this, and it’s what you have to keep looking for. If you run a platform where others come to sell, then you don’t get to sell your own items on the platform because you have two comparative advantages. One, you’ve sucked up information about every buyer and every seller before you’ve made a decision about what you’re going to sell. And second, you have the capacity — because you run the platform — to prefer your product over anyone else’s product. It gives an enormous comparative advantage to the platform.

    This would not be the first time in US history that this kind of arrangement had to be broken up. Back when the railroads were dominant, and you had to get steel or wheat onto the railroad, there was a period of time when the railroads figured out that they could make money not only by selling tickets on the railroad, but also by buying the steel company and then cutting the price of transporting steel for their own company and raising the price of transporting steel for any competitors. And that’s how the giant grows.

    The problem is that’s not competition. That’s just using market dominance, not because they had a better product or because they were somehow more customer-friendly or in a better place. It’s just using market dominance. So my principle is exactly the same: what was applied to railroad companies more than a hundred years ago, we need to now look at those tech platforms the same way.

    This is pretty explicitly taking Senator Warren’s critique of Amazon in particular and applying it to Apple, and to be fair, it is not completely without merit: Apple has quite clearly leveraged the fact it owns the platform to compete with Spotify, for example, and has definitely suppressed competition when it comes to built-in apps like Mail and the aforementioned Safari.

    At the same time, do consumers not matter at all here? Is Senator Warren seriously proposing that smartphone be sold with no apps at all? Was Apple breaking the law when they shipped the first iPhone with only first-party apps? At what point did delivering an acceptable consumer experience out-of-the-box cross the line into abusing a dominant position? This argument may make sense in theory but it makes zero sense in reality.

    What is even more striking, though, is that the App Store does have a massive antitrust problem: it is not Apple unfairly competing with app developers, it is Apple unfairly imposing massive complexity and extracting 30% of revenue with its contractual requirement, enforced by App Review, that developers use Apple’s payment mechanism. I wrote about this extensively last year in Antitrust, the App Store, and Apple (also see this follow-up); I think there is a case Apple’s policies would be found anticompetitive under a Quick Look review, and may even be a per se antitrust tying violation.

    The important takeaway for this Article, though, is the degree to which Senator Warren missed the point: there is significant consumer benefit both to having preinstalled apps and also to Apple controlling the installation of apps. There is a big benefit to suppliers (app developers) as well: the app market on PCs died in large part due to security concerns, which Apple obviated with the App Store to the tremendous benefit of every participant in the ecosystem. Senator Warren’s proposal would make the App Store worse for everyone.

    That leads to a broader point: “tech” is not simply another category, like railroads or telecom. Tech is a means, not an end, but Senator Warren’s approach presumes the latter. That is why she proposes the same set of rules for the sale of toasters and the sale of apps, and everything in between. The truth is that Amazon is a retailer; Apple a combination of hardware maker and platform makers. Google is a search and advertising company, and Facebook a publishing and advertising company. They all have different value chains and different ways of impacting competition, both fairly and unfairly, and to fail to appreciate just how different they are is a great way to make bad laws that not only fail to fix problems but also create entirely new ones.

    That is not to say there aren’t genuine concerns about the biggest tech companies; I was absolutely genuine when I stated at the beginning that Senator Warren deserves credit for bringing these issues to the forefront. To my mind there are three major issues that deserve antitrust attention:

    Issue 1: Digital Advertising

    Senator Warren expresses concern in her article about kill zones when it comes to new startups:

    Weak antitrust enforcement has led to a dramatic reduction in competition and innovation in the tech sector. Venture capitalists are now hesitant to fund new startups to compete with these big tech companies because it’s so easy for the big companies to either snap up growing competitors or drive them out of business. The number of tech startups has slumped, there are fewer high-growth young firms typical of the tech industry, and first financing rounds for tech startups have declined 22% since 2012.

    This is decidedly not the case when it comes to enterprise-focused startups: that sector is thriving with all kinds of new businesses being created, acquired, and going public. The problem is the consumer Internet, which is to say that the problem is digital advertising. As I explained last year, both Google and Facebook are Data Factories; writing about Facebook specifically:

    Facebook quite clearly isn’t an industrial site (although it operates multiple data centers with lots of buildings and machinery), but it most certainly processes data from its raw form to something uniquely valuable both to Facebook’s products (and by extension its users and content suppliers) and also advertisers (and again, all of this analysis applies to Google as well):

    • Users are better able to connect with others, find content they are interested in, form groups and manage events, etc., thanks to Facebook’s data.
    • Content providers are able to reach far more readers than they would on their own, most of whom would not even be aware those content providers exist, much less visit of their own volition.
    • Advertisers are able to maximize the return on their advertising dollar by only showing ads to individuals they believe are predisposed to like their product, making it more viable than ever before to target niches (to the benefit of their customers as well).

    And then, in exchange for these benefits that derive from data, Facebook sucks in data from all three entities:

    • Users provide Facebook with data directly, both through information and media they upload, and also through their actions on Facebook properties.
    • Content is not simply data in its own right, but also a catalyst for generating user action data.
    • Advertisers, like content providers, not only provide data in its own right, which acts as a catalyst for generating user action data, but also upload huge amounts of data directly in order to better target prospective customers.

    The end result is that Facebook and Google are far more valuable to advertisers than anyone else: they offer the most efficient spend when it comes to a return on advertising, and thanks to their ability to reach practically everyone, combined with the infinite nature of digital content, require the lowest investment. Put plainly, the ROI on Google and Facebook digital advertising is unmatched, and the chasm is only growing.

    This is a tremendous problem for any would-be consumer Internet company, particular any product that depends on a network effect. The single most important feature when it comes to building a large user base and a leverage-able network effect is that the product be free-to-use, which means the only viable business model is advertising. As I just noted, though, the only place that advertisers want to be — for good reason! — is Google or Facebook. Ergo, consumer Internet companies are increasingly difficult to get started.

    Snap is an unfortunate example of this reality: Snapchat is a clear demonstration that it is possible to build a competing social network in a world dominated by Facebook; unfortunately, it also appears to be an example of how is is even more difficult to build a profitable advertising business.

    I don’t have a clear solution to this problem; if anything, privacy-focused regulation like GDPR are only exacerbating the issue, given that Google and Facebook acquire most user data on their platforms. Any solution that seeks to actually make a positive impact on competition, though, has to start with advertising.

    Issue 2: Acquisitions

    As I’ve hinted at a couple of times in this article, this is where I do mostly agree with Senator Warren. The truth is that Snapchat would have been a far greater threat to Facebook had the latter not been allowed to acquire Instagram. In a Daily Update last year I explored an alternate history where Instagram stayed independent:

    This is where it is critical to consider the entire ecosystem. Had Instagram continued as a standalone company I do believe it would have been successful in building out an advertising business; it just would have taken a lot more time and effort…What is more important, though, is that an independent Instagram would have been the best possible thing that could have happened to Snapchat. The fundamental problem facing Snapchat is that it wasn’t enough for the company to have higher usage or deeper engagement with teens and young adults, demographic groups advertisers are desperate to reach. As long as Instagram was using Facebook’s ad infrastructure, it would always be more cost effective to reach those groups using Facebook’s ad engine.

    This is why I have called Facebook’s acquisition of Instagram The Greatest Regulatory Failure of the Past Decade, and called for an end to social networks being allowed to buy other social networks. I do have qualms about the idea of retroactively undoing deals, but I do think Senator Warren is directionally correct in this case.

    More broadly, as I explained in The Value Chain Constraint, the price of being an Aggregator is tuning your company to the value chain within which you compete; it follows that all of these companies will face significant challenges moving into new spaces with new value chains. To that end, what makes the most sense from a management perspective is leveraging the tremendous amounts of cash thrown off by their core businesses to acquire and invest in companies competing in different value chains.

    On the flipside, to the extent regulators wish to constrain Aggregators, the single most effective lever is limiting acquisitions. There are significant problems with this, to be sure, particularly when it comes to the incentives for new company creation (most successful exits are acquisitions, not IPOs), but at least this is a remedy that is somewhat approaching the problem.

    Issue 3: Contracts

    As I have detailed, Aggregators already have massive structural advantages in their value chains; to that end, there should be significantly more attention paid to market restrictions that are enforced by contracts.

    Go back to Microsoft: in my estimation the most egregious antitrust violations committed by Microsoft were the restrictions placed on OEMs, both to ensure the installation of Internet Explorer as well as to suppress alternative operating systems. These were not violations rooted in market dominance, at least not directly, but rather contracts that OEMs could not afford to say ‘No’ to.

    This is an area where the European Commission has gotten it right with regard to Google: as a condition of access to Google apps, most critically the Play Store, OEMs were prohibited from selling any phones with Android forks. This is a restriction on competition produced not by market dominance, at least not directly, but rather contracts that OEMs could not afford to say ‘No’ to.

    This is also the issue with Apple’s App Store: the restriction on linking to a website for purchasing an ebook or subscribing to a streaming service is not rooted in any sort of technical limitation; rather, it is an arbitrary rule in the App Developer Agreement enforced by Apple’s App Review team. It has nothing to do with consumer security, and everything to do with Apple’s bottom line.

    This is an area ripe for enhanced antitrust enforcement: these large tech companies have enough advantages, most of them earned through delivering what customers want, and abetted by the fundamental nature of zero marginal costs. Seeking to augment those advantages through contracts that suppliers can’t say ‘No’ to should be viewed with extreme skepticism.

    Let me reiterate a point I have made twice now: I appreciate Senator Warren raising these issues; they are indeed critical not only for the world today, but also the world we wish to create in the future. That, though, only increases the importance of getting things right: the history, the fundamental problem, and the nature of tech. Only then can we start to grope for solutions that actually make the situation better rather than worse.

    I wrote a follow-up to this article in this Daily Update.

    1. It appears that Senator Warren’s argument is based on a New York Times Magazine piece entitled The Case Against Google; I refuted the article directly in this Daily Update. It’s worth repeating a few additional points:

      • First, Microsoft executives blaming antitrust oversight for their failure to compete with Google is self-serving of the highest order. It is a ready-made excuse for the company missing out on search — but as I noted, it doesn’t explain mobile. The reality is far more prosaic: Microsoft didn’t have the structure or culture to compete in either; to CEO Satya Nadella’s credit the company has realized this and is now focused on being an enterprise platform.
      • Second, as I noted above, Microsoft’s Google competitor was woefully inadequate; Google was so much better — and only a click away — that setting the default to MSN Search made no difference.
      • Third, there are far-fetched arguments that Microsoft could have somehow made Google inaccessible on Internet Explorer. Beyond the fact that this would only work with a transparent black list (unlike applications, Microsoft couldn’t play any tricks with APIs), such a block would be easily circumvented by…downloading another browser! Indeed, this would have been the surest route for Netscape to survive.

      In short, while there are arguments to be made about the impact of the antitrust decision on the emergence of web apps — which again, was Microsoft’s core concern — it has nothing to do with the emergence of Google, which was not only not competitive with Microsoft but not even on the company’s radar at the time of the antitrust case. ↩

  • Facebook’s Privacy Cake

    What was striking about the reaction to Mark Zuckerberg’s latest missive about the future of Facebook, A Privacy-Focused Vision for Social Networking, were the two very distinct reactions that, in my estimation, made the same mistake, but in opposite directions; one set of folks didn’t take Zuckerberg seriously at all:

    Another set took Zuckerberg entirely too seriously:

    In fact, what Zuckerberg announced is quite believable, precisely because it makes perfect sense for Facebook: this is a privacy cake that Facebook can have — and eat it too.

    The Social-Communications Map

    Zuckerberg began by describing two distinct kinds of social networks:

    Over the last 15 years, Facebook and Instagram have helped people connect with friends, communities, and interests in the digital equivalent of a town square. But people increasingly also want to connect privately in the digital equivalent of the living room. As I think about the future of the internet, I believe a privacy-focused communications platform will become even more important than today’s open platforms. Privacy gives people the freedom to be themselves and connect more naturally, which is why we build social networks.

    Today we already see that private messaging, ephemeral stories, and small groups are by far the fastest growing areas of online communication. There are a number of reasons for this. Many people prefer the intimacy of communicating one-on-one or with just a few friends. People are more cautious of having a permanent record of what they’ve shared. And we all expect to be able to do things like payments privately and securely.

    Public social networks will continue to be very important in people’s lives — for connecting with everyone you know, discovering new people, ideas and content, and giving people a voice more broadly. People find these valuable every day, and there are still a lot of useful services to build on top of them. But now, with all the ways people also want to interact privately, there’s also an opportunity to build a simpler platform that’s focused on privacy first.

    I first explored the idea of there being different types of social networks in 2013 when I created The Social/Communications Map:

    A drawing of the Social/Communications Map

    Keep in mind, this image is from 2013, but there are still some important points worth calling out:

    • First, the axes are about user perception, not technical implementations; back then tweets were very much experienced as being ephemeral; now they have more permanence, leading to people both being called-out for old tweets and/or deleting their archives.
    • That Twitter and Instagram should have perhaps been on the “permanent” side of the axis is further emphasized by the success of Instagram Stories, which is much more ephemeral than Instagram posts, along with the rise of tweet threads replacing blogging.
    • I noted in the post that LINE was representative of multiple chat networks, including WeChat and WhatsApp; obviously I should have illustrated both instead.

    The most important thing to note, though, are the relative positions of Facebook and Snapchat (it was Facebook’s attempted acquisition of Snapchat that inspired the map in the first place).

    Facebook Versus Snapchat

    Perhaps the most important moment in Facebook’s history was its shift from the private to public space on the Social/Communications Map with the introduction of the News Feed. Again, to be clear, this map is about public perception, not technical reality, and this is a perfect example: data on Facebook was public to everyone in your network from day one. What the News Feed did, though, was change Facebook data from a pull to a push model: instead of needing to seek out your friends’ profiles Facebook would push their updates to you directly.

    This inspired a strong backlash amongst users, who not only complained online but actually organized rallies in person; ultimately, though, it turned out that people loved the News Feed, and once Facebook embraced mobile, it turned out the News Feed was perfect for advertising.

    Still, that transition exposed a soft underbelly in Facebook’s product: private ephemeral communication that allowed users to be their true selves. This was the premise undergirding Snapchat, which I described in Facebook, Phones, and Phonebooks:

    It is increasingly clear that there are two types of social apps: one is the phone book, and one is the phone. The phone book is incredibly valuable: it connects you to anyone, whether they be a personal friend, an acquaintance, or a business. The social phone book, though, goes much further: it allows the creation of ad hoc groups for an event or network, it is continually updated with the status of anyone you may know or wish to know, and it even provides an unlimited supply of entertaining professionally produced content whenever you feel the slightest bit bored.

    The phone, on the other hand, is personal: it is about communication between you and someone you purposely reach out to. True, telemarketing calls can happen, but they are annoying and often dismissed. The phone is simply about the conversation that is happening right now, one that will be gone the moment you hang up.

    In the U.S. the phone book is Facebook and the phone is Snapchat; in Taiwan, where I live, the phone book is Facebook and the phone is LINE. Japan and Thailand are the same, with a dash of Twitter in the former. In China WeChat handles it all, while Kakao is the phone in South Korea. For much of the rest of the world the phone is WhatsApp, but for everywhere but China the phone book is Facebook.

    Make no mistake, the phonebook has been more valuable: it lends itself better to both data collection and advertising. Snapchat, though, threatened to break out of the phone space into the phonebook space with Stories — a product that shifted Snapchat out of the private space into the public one.

    To that end, it is instructive that it is Stories where Facebook finally mounted its Snapchat defense: I wrote in The Audacity of Copying Well:

    Instagram and Facebook are smart enough to know that Instagram Stories are not going to displace Snapchat’s place in its users lives. What Instagram Stories can do, though, is remove the motivation for the hundreds of millions of users on Instagram to even give Snapchat a shot.

    That is exactly what happened: Snap retained its place as the core of 1×1 communication for young people, but the segments more removed from Snapchat’s core use case of chat were suddenly far less likely to even give the service a try, thanks to Instagram’s intelligent leveraging of its network.

    By the same token, though, just because Facebook capped Snapchat’s growth doesn’t mean that Snapchat’s core insight about the desire for private, ephemeral communication was wrong: what Zuckerberg wrote yesterday is basically Snapchat’s reason-for-existing. In other words, while Instagram Stories built a wall around Snapchat by copying Snapchat’s secondary feature, this “Privacy-Focused Vision for Social Networking” is a clear attempt to build the core of Snapchat for everyone else.

    Zuckerberg’s Vision

    Look again at what Zuckerberg outlined:

    • Private interactions
    • Encryption
    • Reducing Permanence
    • Safety
    • Interoperability
    • Secure data storage

    The first three are all about owning the 1×1 private ephemeral space; critically, none of them have anything to do with Facebook’s core feed-based products. Facebook is going to continue to exist as it has to date, as will Instagram, including all of the data collection and ad targeting that currently exist. The “Privacy-Focused Vision for Social Networking” is in addition to Facebook’s current products, not in place of. This is the mistake made by those that took Zuckerberg too seriously.

    As for those who didn’t take Zuckerberg seriously enough, why wouldn’t Facebook want to move in this direction? There are multiple benefits:

    • First, this is a valuable space to own for all of the reasons that Snapchat succeeded in the first place. People want a place to communicate freely without fear of snooping or a historical record.
    • Second, to the extent the rise of 1×1 networking is inexorable, it is better for Facebook that it happen on their properties. Not only does Facebook preserve the ability to advertise on privacy-focused platforms — the company can leverage data from Facebook to advertise in its messaging products (although I am skeptical that messaging products are well-suited to advertising) — it also prevents would-be competitors from capturing leverageable attention.
    • Third, as we have seen over the last 24 hours, there are tremendous PR benefits from a privacy-focused service. Facebook has changed nothing about its core service or data collection policies, yet the assumption is that the company is pivoting and the only debate is whether to believe them or not.

    Perhaps most compelling, though, is the degree to which this move locks in Facebook’s competitive position. As I noted above, Snapchat already showed that Facebook is vulnerable in the realm of private ephemeral communications, but soon that will no longer be the case. Moreover, given Facebook’s focus on end-to-end encryption, the company has made it that much harder to even get off the ground: not even Snapchat is fully end-to-end encrypted (pictures are, but not text messages).

    There is an even more important benefit to Facebook voluntarily forgoing the data within messages and limiting the time it keeps surrounding metadata (make no mistake, end-to-end encryption is a real thing — Facebook will not be able to see encrypted messages); as Zuckerberg told Wired:

    Certainly, ad targeting can benefit from having access to as much content or signal as possible. You know, I’m more optimistic about this for a few reasons. One is that we aren’t really using the content of messages to target ads today anyway. So we weren’t planning on doing that. So it’s not like building a system and making it end-to-end encrypted and now we can’t see the messages is really going to hurt ads that much because of the way we were already thinking about that. Keeping metadata around for less time will have some impact, although I’m optimistic that we’ll build systems that can basically deliver most of the value with a fraction of the amount of data.

    Why can Facebook deliver most of the value? Because they are still Facebook! They still have the core Facebook app, Instagram, ‘Like’-buttons scattered across the web — none of that is going away with this announcement. They can very much afford a privacy-centric messaging offering in a way that any would-be challenger could not. Privacy, it turns out, is a competitive advantage for Facebook, not the cudgel the company’s critics hoped it might be.

    Safety, Interoperability, and Strategy Credits

    The last three items in Zuckerberg’s list are interesting in their own right; to take them one-by-one:

    Safety: This is about the very real trade-offs that come with end-to-end encryption. One obvious issue is law-enforcement: Apple has already been down this road with the FBI when it comes to phone security; end-to-end encryption is both more challenging and yet simpler, simply because it is, properly implemented, truly unbreakable.

    Another issue is misinformation: for all of the issues surrounding misinformation on Facebook, at least misinformation is traceable; that is not the case if messages are encrypted, which has already been an issue with WhatsApp in India. One could certainly make the cynical argument that, in the process of cloaking itself in privacy, Facebook is washing its hands of misinformation.

    To be sure, Facebook is confident it can leverage its ability to analyze metadata to stop bad actors; that the exact same sort of audience analysis is perfectly portable to advertising is a rather happy benefit as far as Facebook is concerned.

    Interoperability: This is perhaps the feature that is easiest to be cynical about; while it can certainly be frustrating to have to balance multiple messaging apps, for much of the world consolidating Facebook-owned messaging will not fully address the problem, thanks to alternatives like Messages, LINE, Kakao, etc. Moreover, even in areas where Facebook owns both the Phone (via WhatsApp) and the phonebook (via Facebook and Instagram), exactly how much consumer demand is there for integration?

    There is, to be sure, a business argument: Facebook has already unified much of the ad infrastructure underlying its services, and unifying messaging is, to the extent Facebook wants to build a business platform on messaging, a natural next step. There is also a regulatory argument: while it is difficult to make the argument that Facebook has broken antitrust laws, the remedy, should that be accomplished, is obvious — split off Instagram and WhatsApp. That will be harder to do if they are fully integrated with Facebook, not simply on the advertising side but also the user side.

    Secure Data Storage: This is an interesting addition to this piece, as it has little to do with messaging in the communications sense, but a lot to do with messaging in the political sense. This is what Zuckerberg wrote:

    There’s an important difference between providing a service in a country and storing people’s data there. As we build our infrastructure around the world, we’ve chosen not to build data centers in countries that have a track record of violating human rights like privacy or freedom of expression. If we build data centers and store sensitive data in these countries, rather than just caching non-sensitive data, it could make it easier for those governments to take people’s information.

    Upholding this principle may mean that our services will get blocked in some countries, or that we won’t be able to enter others anytime soon. That’s a tradeoff we’re willing to make. We do not believe storing people’s data in some countries is a secure enough foundation to build such important internet infrastructure on.

    The most obvious country worth avoiding is China, which means this is clearly a Strategy Credit:

    A strategy credit is an uncomplicated decision that makes a company look good relative to other companies who face much more significant trade-offs.

    Facebook is already banned in China, so not putting data centers in China costs the company nothing (it may soon cost the company in Russia; one imagines Facebook will not mind being banned there of all places). This is in sharp contrast to Facebook’s most vociferous critic in tech, Tim Cook and Apple; the latter absolutely stores customer data in China along with encryption keys, because that is the law for companies that wish to operate in the country.

    To be clear, this is understandable, but that is what makes that Strategy Credit article rather ironic; I coined the term in response to Apple’s posturing about user data in the wake of the Snowden revelations, noting that forgoing data wasn’t really a tradeoff given Apple’s business model. Now Apple is on the other side of the coin.

    Privacy Moats

    Ultimately there are three broad takeaways from Zuckerberg’s article:

    • Stop expecting companies to act against their interests. Facebook isn’t killing their core business anymore than Apple, to take a pertinent example, is willing to go to the mat to protect user data in China.
    • Facebook doing something that benefits itself is not inherently bad for end users. It is perfectly reasonable that the company can be instituting genuinely user-friendly changes like end-to-end encryption even as it furthers its own self-interests.
    • Relatedly, and most importantly, there needs to be much more appreciation for the anti-competitive trade-offs inherent in an absolutist approach to privacy. Facebook is doing what its fiercest critics supposedly want, and enhancing its competitive position as a result.

    This was a point I made last year in Open, Closed, and Privacy:

    If an emphasis on privacy and the non-leakage of data is a priority, it follows that the platforms that already exist will be increasingly entrenched. And, if those platforms will be increasingly entrenched, then the more valuable might regulation be that ensures an equal playing field on top of those platforms. The reality is that an emphasis on privacy will only increase the walls on those gardens; it may be fruitful to rule out the possibility of unfair expansion.

    This is a debate that is woefully lacking. The reality is that the only user-friendly way to enforce privacy — which is another way of saying the only scalable way in a demand-driven world — is to severely limit inter-operability and over-burden would-be challengers. Regulators need to be far more aware of this and either choose another approach to privacy — i.e. entrust it to individuals — or regulate data-platforms, at least in terms of competition on top of their platforms, even more severely.

  • The Value Chain Constraint

    On June 16, 2017, minutes after Amazon announced it was buying Whole Foods Market Inc. for $13.7 billion, grocery store stocks fell through the floor; from MarketWatch (emphasis mine):

    Shares of grocery stores took an unexpected hit Friday as investors reeled from the news that Amazon.com Inc. was moving into their space by acquiring Whole Foods Market Inc. After Amazon announced that it was buying Whole Foods in a $13.7 all-cash deal, shares of grocery store chain Kroger Co. slid to close down 9.2%, shares of Costco Wholesale Corp closed down 7.2%, Target Corp.’s stock closed down 5.2% and shares of Wal-Mart Stores Inc. closed down 4.6%…

    The stock prices of grocers when Amazon bought Whole Foods

    Mark Hamrick, a senior economic analyst at Bankrate.com, said Amazon’s technological innovation in traditional retail is a “earthquake” for the sector, which it may have hinted at with its recent launches of brick-and-mortar Amazon bookstores. “We can only imagine the technological innovation that Amazon will bring to the purchasing experience for the consumer,” Hamrick said.

    This is why I found Walmart’s recent earnings so interesting: the company cited groceries as the biggest drivers of its ecommerce business, both last year and going forward — the company plans to expand grocery pickup to an additional 1,000 stores — because, as Walmart CEO Doug McMillon put it on the company’s earnings call:

    We strive to make every day easier for busy families as we increase convenience and save them money and time. Part of our strategy is to build on our existing strengths, such as having a broad assortment including fresh and perishable foods within 10 miles of 90% of the U.S. population.

    Amazon, meanwhile, appears to be struggling; from Bloomberg:

    The number of Amazon Prime members who shop for groceries at least once a month declined in 2018 compared with 2017, according to the results of an annual consumer survey released Wednesday by UBS analysts. The drop was surprising given the company’s Whole Foods investment and expansion of two hour delivery service Prime Now, the analysts wrote in a note to investors.

    A separate study by research firm Brick Meets Click found that households using grocery delivery and pickup services from physical retailers spend about $200 per month and place orders more frequently than Amazon grocery shoppers, who spend $74 a month.

    So where is the promised technological innovation?

    The Conservation of Groceries

    I have written several times about the Conservation of Attractive Profits, most notably with regards to Netflix, Facebook and BuzzFeed, and Zillow. To put it in generic terms, profit in a value chain flows to whatever company is able to successfully integrate different component pieces of that value chain; the other parts of the value chain then modularize and are driven into commodity competition.

    For example, this is what Walmart’s traditional value chain looked like:

    Walmart's value chain

    Walmart was able to integrate wholesale purchasing with an expansive network of stores; this provided a moat of sustainably lower prices for customer driven by purchasing power over suppliers.

    Amazon, though, thanks to technological innovation — specifically, the Internet — was able to build a different integration in the value chain:

    Amazon's value chain

    Amazon integrated wholesale purchasing and fulfillment centers with Amazon.com, relying on modularized delivery services for distribution; this provided a moat of superior selection and, at least at the beginning, lower prices, and with Prime, superior convenience, at least for non-perishable goods.

    Walmart has worked for years to respond to Amazon’s threat; the problem, though, as I explained in 2016’s Walmart and the Multichannel Trap, is that an integration built around stores was fundamentally unsuited to offering the sort of selection and convenience that Amazon does. The company needed to build up an entirely new set of capabilities and integrations, even as Amazon was leveraging theirs to integrate forward into logistics, adding on a 3rd-party marketplace to expand selection even more, and integrating backwards into their own brands. The result is that Amazon has around 50% share in e-commerce while Walmart has less than 5%.

    That, though, is precisely why groceries is worth examining: as I explained when Amazon bought Whole Foods, perishable goods are not well-suited to Amazon’s value chain. Superior selection has diminishing returns, quality varies on an item-by-item basis within a single SKU, and, most importantly, the quality of items degrades with time and transport. In other words, they are a great fit for stores, not distribution centers.

    In this view, Amazon’s purchase of Whole Foods was an attempt to acquire a first best customer for its grocery delivery operation, one that would efficiently store and sell perishable goods that weren’t suitable for Amazon’s traditional e-commerce model. And, to be clear, this strategy may yet succeed, but only to the extent Amazon builds a completely new set of capabilities and integrations that will probably end up looking a lot like Walmart, which has a massive head start it is clearly taking advantage of.

    In other words, what matters is not “technological innovation”; what matters is value chains and the point of integration on which a company’s sustainable differentiation is built; stray too far and even the most fearsome companies become also-rans.

    Google Cloud Struggles

    Consider Google, a company that, more than any other, has been predicated on “technological innovation”. This was possible because the company’s core product — Internet search — entered a value chain with no integrations whatsoever. On the supply side there were countless websites and even more individual web pages, increasing exponentially, and on the demand side were a similarly increasing number of Internet users looking for specific content.

    Crucially, all of the supply was easily accessible — just link to it — and all of the demand was capturable — they only needed to type in google.com. This meant that the best search engine — and by best, I mean the purest form of the word, i.e. best performing — could win, and so it did. Google was leaps and bounds better than the competition, thanks to its focus on understanding links — the fabric of the web — instead of simply pages, and consumers flocked to it.

    This set off the positive cycle I have described in Aggregation Theory: owning demand gave Google increasing power over supply, which came onto Google’s platform on the search engine’s terms, first by optimizing their web pages and later by delivery content directly to Google’s answer boxes, AMP program, etc., all of which increased demand, resulting in a virtuous cycle.

    At the same time Google was building out two critical pieces of the value chain in integration with Search: the first was infrastructure — supporting that much demand required huge investments in servers, fiber optic cables, etc. — and the second was advertising. Ultimately the company’s model looked like this:

    Google's value chain

    Note how Google is so dramatically optimized on all three sides of this integration: users, suppliers, and advertisers interact with Google through their own volition, thanks to the infrastructure Google has built to facilitate that interaction, with almost no person-to-person contact with anyone from Google. It is a model that works very, very well — for search and digital advertising, anyways.

    Things have not gone so well for Google Cloud. At first glance, selling infrastructure seems like an obvious opportunity for Google, and much ink has been spilled about how the company — any day now! — will threaten Amazon or Microsoft. After all, Google was building out worldwide infrastructure before anyone else, and the company remains at the forefront of technological innovation.

    The problem, though, is that the company’s value chain is completely wrong. The world of enterprise software is not a self-serve world (and to the extent it is, AWS dominates the space); what is necessary is an intermediary layer to interact with relatively centralized buyers with completely different expectations from consumers when it comes to product roadmap visibility, customer support, and pricing.

    It has taken Google many years to learn this lesson: Google Cloud remains a distant third to AWS and Microsoft with a strategy that simply wasn’t working. I wrote in a November Daily Update upon the occasion of Google Cloud changing CEOs:

    A strategy predicated on being “better” on specific product attributes, though, may fit the culture of Google, but it doesn’t necessarily lead to a winning enterprise strategy. To that end, Google Cloud faces three major problems:

    • First, Google has not made an effective case about how specifically machine learning can benefit business that is appreciably different than traditional business analytics. That is not to say it can’t, just that the company hasn’t really made the case.
    • Second, Google isn’t competing with Lycos and Yahoo: AWS and Microsoft have machine learning offerings of their own, and Microsoft in particular is much more accomplished at productizing offerings in a way that are understandable and approachable to CIOs.
    • Third, and most importantly, the technical attributes of a product are only one piece of what matters to success in the enterprise. Just as important are customization, support, and the ability to sell. Google is widely regarded as being the worst in all three areas.

    In short, what Google Cloud needs is not a CEO that fits the culture, because the culture of Google is about making the best product technologically and waiting for customers to line-up. That may have worked for Search and for VMWare, but it’s not going to work for Google Cloud. Instead the company needs to actually get out there and actually sell, develop the capability and willingness to tailor their offering to customers’ needs, be willing to build features simply because they move the needle with CIOs, and actually offer real support.

    In short, Google Cloud is competing in a different value chain than is Google search, and it needs to build new integrations accordingly. To that end, note the strategy chosen by Thomas Kurian, Google Cloud’s new CEO; from the Wall Street Journal:

    The new leader of Google’s cloud-computing business plans to dramatically expand its sales team, addressing one of the biggest challenges he faces as rivals Amazon.com Inc. and Microsoft Corp. race ahead in the market…While Google has long offered cloud technology, it has seen Amazon and Microsoft surge ahead to become the leaders in providing computing power and storage services for rent over the web. Those companies have robust sales and service staffs that large corporate customers demand to support their technology needs, an area where Google has trailed, analysts have said.

    In other words, Google Cloud needs to look a lot more like Microsoft.

    Microsoft’s Enterprise Value Chain

    Microsoft, unlike Google, has always been first-and-foremost an enterprise company. That means its integration was between its operating system and the associated APIs on which enterprise apps were built:

    Microsoft's value chain

    Note, though, that unlike Google’s value chain, Microsoft is much further from the end-user: devices were built and sold by OEMs, sometimes to end users, but especially to enterprise IT departments by dedicated sales forces. Similarly, Microsoft developers were by-and-large enterprise software developers, working not for end users but for management.

    This had obvious downsides in the consumer market: products in the Microsoft value chain were typically feature rich and user experience poor, exactly what you would expect from a world run by top-down purchase order, not individual consumer choice. To the extent Microsoft did succeed in the consumer space, the reason was a spillover from their dominance in enterprise; by the time pure consumer markets like the web or mobile came along, Microsoft was woefully unprepared to compete. They were basically the opposite of Google.

    That, though, is also why Microsoft is succeeding with Azure even as Google struggles with Google Cloud: the company is used to value chains that include sales forces and top-down decision-making, and has the right business model and integrations to take advantage.

    The Netflix Exception

    Perhaps the most famous example of a prominent company “pivoting” and succeeding is Netflix, but that is very much the proverbial exception that proves the rule. Netflix built its initial customer base and IPO’d through a business model predicated on renting DVDs via mail. The value chain looked like this:

    Netflix's value chain

    What was critical to making this value chain work was the first-sale doctrine: when a DVD was sold the rights of the copyright holder were exhausted; that means that Netflix could buy all of the DVDs it wanted and rent them to customers without copyright owners restricting them in any way. Critically, this meant that Netflix could integrate the customer relationship with content ownership.

    Notice that that is the exact same integration that Netflix enjoys today: more and more of the company’s content catalog — particularly the portions that attract new customers — is original content owned by Netflix. In other words, the point of integration — the customer relationship and content ownership — is the same as in the DVD days.

    To be sure, it took time for Netflix to transition to this model, and the company was absolutely helped along by hapless studio executives more interested in bumping up their annual profit than in considering their long-term position in the content value chain. There are any number of points in the early days of streaming when Netflix — because it was, if only temporarily, in a vulnerable non-integrated position in its value-chain — could have been stopped. I suspect, though, those days have past, which is why Netflix Flexes.

    More generally, from a value chain perspective, Netflix’s transformation was less of a pivot than it might have first appeared: sure, the technology of DVDs by mail and streaming video are fundamentally different, but the value chain is the same. That is a far more viable transition than trying to leverage broadly similar technology into completely new markets and value chains.

    The Solipsism Trap

    It is understandable why the Internet giants in particular move into seemingly adjacent territories: the growth imperative is strong, both for financial and strategic reasons, and the technology seems easy enough, particularly given the resources these companies bring to bear. And yet, the truth is that those massive resources do not stem, at least in the long run, from technical excellence, but rather integration in specific value chains that produces positive feedback loops and outsize profits.

    It follows, then, that without that integration, the positive feedback loops quickly disappear, along with the profits, which is the exact pattern we see again and again. Microsoft spent billions on phones and consumer Internet services, Amazon spent billions on Whole Foods, Google has spent billions on not just Google Cloud but a whole host of initiatives that have nothing to do with Search, Facebook has spent billions on Watch and VR, and now Apple is getting in the game with billions spent on Video, and the expected outcome of all these should be that they will fail.

    To be sure, failure takes time: these companies do have nearly unlimited resources thanks to their core business models, and the reckless optimism bred by structural success. And, I suppose, sometimes they can actually push products across the line to profitability, kind of. Bing, for example is profitable — if you exclude traffic acquisition costs, which makes my point.

    The reality is that technology has an amplification effect on business models: it has raised the Internet giants to unprecedented heights, and their positions in their relevant markets — or, more accurately, value chains — are nearly impregnable. At the same time, I suspect their ability to extend out horizontally into entirely different ways of doing business — new value chains — even if those businesses rely on similar technology, are more limited than they appear.

    What does work are (1) forward and backwards integrations into the value chain and (2) acquisitions. This makes sense: further integrations simply absorb more of the value chain, while acquisitions acquire not simply technology but businesses that are built from the ground-up for different value chains. And, by extension, if society at large wants to limit just how large these companies can be, limiting these two strategies is the obvious place to start.

    I wrote a follow-up to this article in this Daily Update.

  • The Cost of Apple News

    Apple is, according to the Wall Street Journal, driving a hard bargain with publishers ahead of the launch of its rumored News subscription service:

    Apple Inc.’s plan to create a subscription service for news is running into resistance from major publishers over the tech giant’s proposed financial terms, according to people familiar with the situation, complicating an initiative that is part of the company’s efforts to offset slowing iPhone sales. In its pitch to some news organizations, the Cupertino, Calif., company has said it would keep about half of the subscription revenue from the service, the people said. The service, described by industry executives as a “Netflix for news,” would allow users to read an unlimited amount of content from participating publishers for a monthly fee. It is expected to launch later this year as a paid tier of the Apple News app, the people said.

    The rest of the revenue would go into a pool that would be divided among publishers according to the amount of time users spend engaged with their articles, the people said. Representatives from Apple have told publishers that the subscription service could be priced at about $10 a month, similar to Apple’s streaming music service, but the final price could change, some of the people said…

    Another concern for some publishers is that they likely wouldn’t get access to subscriber data, including credit-card information and email addresses, the people said. Credit-card information and email addresses are crucial for news organizations that seek to build their own customer databases and market their products to readers.

    Probably the most obvious way to understand this story is that it, along with the report that Apple would have a launch event on March 25, appear to be attempts to negotiate through the media. I’m reminded of the January 2010 report in the Wall Street Journal that Apple’s impending tablet would cost $1,000; when Steve Jobs announced the iPad three weeks later, the $499 starting price seemed like a bargain. Perhaps leaking a 50/50 revenue share, along with an impending deadline for negotiations, is a way to make a 60/40 or 70/30 revenue share seem like a reasonable compromise?

    The Growth of Apple News

    Let’s back up for a moment: Apple News has grown to be a major force in publishing, at least in terms of traffic. According to a New York Times story that Apple cooperated with, the service “is read regularly by roughly 90 million people.” That has translated into traffic for news publishers that, according to Slate, often outpaces Facebook post-last January’s algorithm change.

    The problem, as Digiday explained, is that traffic — which is almost completely realized within the Apple News app, not on publisher’s web pages — comes with minuscule amounts of revenue. Yes, Apple News allows for advertising, but that advertising is either sold (poorly) by Apple or sold directly by the publisher with no allowance for either programmatic ads nor data about users.

    So why do publishers bother?

    Apple News and Aggregation

    There are a number of factors that should ring familiar to anyone familiar with the travails of publishers on the Internet.

    To start, Apple News readers visit Apple News and, for the most part, read what Apple News presents to them; the front-page may be human-selected, as Apple sought to make clear in that New York Times article, but just as is the case with algorithmic selection (which is what determines what users see for the rest of Apple News, it just happens to be called “Suggested by Siri”), no one publication is favored:

    Apple News Today view

    On one hand, this is obviously not good for publishers: there is limited wherewithal to build brand affinity, there is no customer data shared (for purposes of follow-up, much less ads), and as noted above, there really isn’t much money to be made.

    On the other hand, what are publishers really giving up? Readers going to the Apple News app have already made the decision to not visit a particular publisher’s website directly, and, given that digital content has zero marginal cost, why not support Apple News on the off chance some article hits it big?

    It should be noted that publisher pages within Apple News complicate this narrative a bit: on one hand, they are a place to build brand affinity; on the other hand, they are more likely to cannibalize direct visits to the publisher’s website. But how many Apple News users are likely to switch to a browser for a particular publisher should they leave Apple News?

    What is happening is Aggregation: Apple News attracts the users, which means publishers are coming onto Apple’s platform on Apple’s terms, which makes Apple News more attractive to users, making publishers ever more reticent to leave even though they aren’t getting much out of the deal.

    Apple News and Publisher Subscriptions

    For suppliers, the antidote for Aggregation is to go direct to consumers; the key is to embrace the same forces that drive Aggregation. First, the addressable market should be the entire world, not just a limited geographic area. Second, the same sort of automated payment tools available to advertisers on Aggregators can be leveraged for consumers; indeed, the tools for consumers, particularly given the lower dollar amounts and decreased need for paperwork, can be as simple as Apple Pay, and they can scale indefinitely. Third, a freemium approach to content means that social networks can be used for user-generated marketing.

    Apple News as currently construed is actually somewhat helpful in this regard: publishers can push subscription-only content (as well as free content) into Apple News, and give users the option to subscribe using the App Store. For example, the Wall Street Journal elected to make the piece that triggered this Article free:

    Free Wall Street Journal article in Apple News

    However, the next story over, about Google Cloud, requires a subscription:

    Subscription story in Apple News

    It’s not perfect: clicking on that subscription link means the publisher has to pay Apple 30% the first year and 15% after that, and they don’t get any customer data (unless the customer creates an account in order to use their subscription on other platforms). Still, to my mind it is somewhat less egregious than Apple’s restrictions on in-app purchase; Apple News is driving the customer to a publisher’s content and charging accordingly (as opposed to taking a tax simply because there is no alternative to the App Store), but at the end of the day the publisher is still establishing a direct paying relationship with a subscriber.

    The Spotify of News

    What Apple is reportedly building now, though, is decidedly different. The so-called “Netflix of News” — although, given that Apple will pay out on a marginal basis as opposed to buying content, a better descriptor would be the Spotify of News — would entail customers paying one monthly fee to Apple which Apple distributes to publishers based on what subscribers read.

    Publishers should be very clear about the implications of this model: it is not a direct-to-consumer model. Rather, it is an Aggregation model that happens to monetize via subscriptions instead of ads. That means it has all of the same problems for publishers that are posed by Aggregators:

    • Publishers do not form a direct connection with users; that connection is with Apple News
    • Publishers get no meaningful data (including no email addresses); there is no means to increase engagement or monetization down the road
    • Publishers must compete with every other publisher for attention

    That last point is the most important, and should weigh heavily on publishers that have committed to the subscription model. What makes subscriptions work is an alignment between editorial and business model: the former is incentivized by quality and differentiation because the payoff is a customer with a high lifetime value; the New York Times put this succinctly in their 2020 Report:

    We are, in the simplest terms, a subscription-first business. Our focus on subscribers sets us apart in crucial ways from many other media organizations. We are not trying to maximize clicks and sell low-margin advertising against them. We are not trying to win a pageviews arms race. We believe that the more sound business strategy for The Times is to provide journalism so strong that several million people around the world are willing to pay for it. Of course, this strategy is also deeply in tune with our longtime values. Our incentives point us toward journalistic excellence.

    The proposed Apple News model, on the other hand, which pays out according to reader engagement, pushes in the opposite direction — the Facebook direction. The motivation is “to maximize clicks” and “win a pageviews arms race”, with some “time-spent” variables mixed in; sure, the driver isn’t low-margin advertising, but shifting the means of monetization doesn’t change the ends as far as incentives go.

    The Cost of Apple News

    It is absolutely worth noting what a great deal for consumers an Apple News subscription bundle would be: I totally get the idea of subscription fatigue, and having one place to get all of the best journalism would be amazing. That, though, doesn’t mean that Apple News wouldn’t be an Aggregator: that confirms it! Aggregators win because consumers prefer them, leaving publishers no choice but to go where the consumers are.

    To that end, I am sure that a significant number of publications will sign up for Apple’s offering; clearly the company is confident enough to leak a date. And, frankly, many publications should: most publishers are already locked into the volume game when it comes to their editorial direction, and Apple News subscription payouts will be additive to the bottom line.

    Publishers that have truly committed to subscriptions, though, should say no: not only will it be difficult to make up revenue that will be cannibalized lower per-customer payouts from Apple News, but more importantly a reversion to a model predicated on page views will hurt their business in the long run. This is especially the case if Apple News becomes a major revenue driver; yes, digital content can be distributed with zero marginal cost, but the incentive cost should not be discounted — it works directly against the quality imperative that is the critical factor in making the Aggregator-avoiding direct-to-consumer business model work.

    I wrote a follow-up to this article in this Daily Update.

  • Spotify’s Podcast Aggregation Play

    Credit Spotify CEO Daniel Ek with honesty; in a blog post announcing a major move into the podcast space, Ek wrote:

    More than 10 years ago we founded Spotify to give consumers something they couldn’t get — music any time, anywhere, and at the right price. Along the way, we broke the grip piracy had on our industry and restored the growth of global music through paid on-demand streaming. I’m proud of what we’ve accomplished, but what I didn’t know when we launched to consumers in 2008 was that audio — not just music — would be the future of Spotify.

    There is a lot packed into this paragraph. First and foremost, Ek is absolutely justified in taking a victory lap: as I noted last month music revenue is growing sharply; over the next few years the industry’s total revenue will likely exceed the peak of the CD era, something that seemed unthinkable a decade ago.

    U.S. music industry sales over time

    What happened is that Spotify dragged the record labels into a completely new business model that relied on Internet assumptions, instead of fighting them: if duplicating and distributing digital media is free (on a marginal basis), don’t try to make it scarce, but instead make it abundant and charge for the convenience of accessing just about all of it.

    The problem for Spotify is that the company’s financial returns are not nearly representative of its impact on the music industry. The company did make its first operating profit of last quarter on revenues of €1.5 billion, but the biggest driver was the fact its operating costs were down 17% due to lower “accrued social costs” in Sweden that resulted from Spotify’s stock price going down. To be fair, Spotify said it would have made a small profit even without that adjustment, but the long-term outlook is tough when the company’s gross profit is, as it was last quarter, only 27%.1

    The issue, as I laid out last year in Lessons From Spotify, is that Spotify’s primary cost driver is not, like most tech companies, fixed investments in R&D or Sales & Marketing, but rather marginal payouts to record labels. Basically, the more revenue Spotify makes the more its costs increase, which can be overcome at large enough revenue numbers — see last quarter — but limits the company’s long-term upside.

    At least, that is, as long as Spotify was a music company; thus the new declaration from Ek that Spotify is now about audio — the honesty was his admission that he didn’t originally see this coming.

    Podcasts Versus Music

    The shift in purpose from “music” to “audio” is, for now anyways, about podcasts. And, at least from a user perspective, it is a natural extension: playing music and playing podcasts entail downloading or streaming some sort of digital file, decoding it on a device, and playing it back through some sort of speaker. That one involved melodies and harmonies and the other primarily the spoken word (although there are plenty of music podcasts) is, from a technical perspective, a distinction without meaning.

    From a value chain perspective, though, music and podcasts could not be more different:

    The music value chain versus the podcast value chain

    • Music is primarily controlled by three large labels; podcasts are controlled by individual podcasters
    • Music can only be played legally through licensed services or via licensed downloads; podcasts can be played by anyone
    • Music generated $8.7 billion in revenue in the U.S. in 2017; podcasts generated only around $300 million in the U.S.

    This last point is directly related to the first two: the money that can be made from a value chain is directly related to the degree of friction and centralization in that value chain. Consider Spotify’s two primary business models:

    • Subscriptions capture money directly from consumers who, as noted above, are paying for the convenience of accessing all of the music they want (i.e. overcoming friction) in one centralized place.
    • Advertisements capture money from advertisers who wish to reach listeners; effectively selling advertising, though, means having one place for advertisers to go to reach a large number of listeners.

    This importance of centralization to an advertising business model is best seen by the fact that Spotify drove €542 million ($616 million) in advertising revenue last year, far outpacing all of podcasting, even though half of the company’s users didn’t hear any ads at all. Moreover, the total amount of advertising revenue driven by music is even greater when you add in YouTube.

    Podcasts and the Web

    The current state of podcast advertising is a situation not so different from the early web: how many people remember this?

    The old "punch the monkey" display ad

    These ads were elaborate affiliate marketing schemes; you really could get a free iPod if you signed up for several credit cards, a Netflix account, subscription video courses, you get the idea. What all of these marketers had in common was an anticipation that new customers would have large lifetime values, justifying large payouts to whatever dodgy companies managed to sign them up.

    The parallels to podcasting should be obvious: why is Squarespace on seemingly every podcast? Because customers paying monthly for a website have huge lifetime values. Sure, they may only set up the website once, but they are likely to maintain it for a very long time, particularly if they grabbed a “free” domain along the way. This makes the hassle of coordinating ad reads and sponsorship codes across a plethora of podcasts worth the trouble; it’s the same story with other prominent podcast sponsors like ZipRecruiter or SimpliSafe.

    The problem is that the affiliated marketing for large lifetime-value purchases segment is not a particularly large one; that meant that the amount of consumer attention paid to the Internet far exceeded the amount of advertising spend. From Mary Meeker’s 2005 Internet Trends report:

    A slide from Mary Meeker's 2005 Internet trends report showing how the Internet was under-monetized

    It seems very likely that were a similar slide to be made about podcasting it would look very similar: according to Edison Research 73 million people in the U.S. listen to podcasts monthly, and 48 million weekly; the average listener listens to seven podcasts a week. That seems like it should be worth a lot more than $300 million or so!

    Ad Centralization

    Again, what happened to the web is likely instructive: in 2003 Google launch AdSense, an advertising network for websites. Now advertisers could buy ads in one centralized place, and those ads could be better targeted by one company that spread its cookies across the entire Internet (and, of course, combined them with data from search, email, etc.).

    By 2010, five years after the above slide, Meeker had an update:

    A slide from Mary Meeker's 2010 report showing how web monetization had improved

    Internet attention still outpaced monetization, but the gap was significantly closer: yes, the ad formats were still mostly the same, but increased centralization brought far more advertisers on board.

    To be sure there have been attempts to centralize podcast advertising as well: a company called Midroll, which was acquired by E.W. Scripps in 2015, is the largest player in the space. Midroll sits between advertisers and mostly larger podcasts like the Bill Simmons Podcast or WTF with Marc Maron, and handles the nitty gritty of coordinating ad reads and distributing discount codes and specialized URLs in exchange for about a third of revenue.

    Three years ago Midroll also acquired a podcast player called Stitcher; as I explained at the time there was a lot of value to be gained from controlling both the listening experience and ad sales, particularly in terms of data: with better data Midroll could more easily sell podcast advertising inventory to companies with business models that did not rely on generating outsized lifetime values.

    The problem for Midroll, though, is that Stitcher never gained a meaningful share of the market, which meant Midroll never achieved the sort of data necessary to expand the podcast advertising market. Sure, some brand advertisers are dipping their toes in the market, but the leading advertisers are the same sort of companies they have always been, and while users no longer need to punch any monkeys, they do still need to punch in those discount codes and specialized URLs.

    Meanwhile Apple, which does have the users thanks to the dominance of the iOS Podcast app,2 has shown little inclination of being that centralized player. I wrote about the company’s opportunities in the space two years ago, but despite the shift in strategy to services nothing has changed.

    The Value of Gimlet Media

    All of this is critical context for understanding Spotify’s strong interest in the podcasting space. Spotify needs (1) a way to differentiate its service from Apple Music in particular, and (2) content that it does not have to pay for on a marginal cost basis.

    Gimlet Media fits the bill in both cases. While the company’s current roster of podcasts will remain freely available, future podcasts will almost certainly be exclusive to Spotify. More importantly, it seems likely that Spotify bought the company not simply for its library but also its management: expect a big jump in output with additional investment.

    It’s worth considering why it is exclusivity in podcasts will likely play out differently in podcasts than in music. CEO Daniel Ek said on the company’s earnings call yesterday:

    I think these are two very, very different businesses. We’ve spoken in the past about the music industry and not being a space, where exclusivity makes sense for a number of different reasons, but including one of them, that music, radio can put any piece of music up. So exclusivities won’t have the same effect, as you won’t be able to keep it exclusive.

    And the second thing obviously is the artists and the label have the incentive to push the content out in many places as possible, because so much of an artist revenue derives from touring. I think in audio and certainly in podcast, the dynamics is very, very different, and what we’re doing here and what we’re excited about is really building the market, it’s at a very, very different stage of maturity. So we’re investing in that and we think we can be one of the tent-pole players in that space.

    Basically, the wall that Spotify can put up around podcasts is much stronger than the one it can put up around music, and podcasters have fewer alternatives. Or, to put it another way, podcasts are a market where Spotify — to the extent they are willing to pay — actually has power over supply.

    Meanwhile, for Spotify podcasts are fixed costs: that means that driving more listening flows directly to the company’s bottom line in a way that increased music listening does not. This is a very big deal — it is entirely possible that if Spotify succeeds in the space that podcasts will drive a relatively small percentage of revenue and a much larger percentage of profit.

    Spotify’s Aggregation Play

    At the same time, the Gimlet Media acquisition on its own does not seem like a sustainable strategy: paying three-quarters of the amount generated in annual revenue by an entire industry for 25 or so podcasts does not scale. That, though, is where the Anchor acquisition comes in: Anchor is a service for easily making and distributing podcasts, with a nascent advertising service for monetization.

    To put it another way, Anchor is a means of generating supply, and it is supply that has always stood in the way of Spotify’s ambitions to be an Aggregator. Aggregators bring suppliers onto the platform on their terms; Spotify, on the other hand, has had to scratch and claw to get labels to give them the music they needed to be viable. And again, the acquisition of Gimlet Media, while better from a long-term leverage perspective, is not a big improvement: Spotify almost certainly overpaid if the only goal was to obtain supply.

    What I think Spotify senses, though, is that while podcasts, at least in theory, solve many of their business model problems, Spotify is also uniquely positioned to solve the problems of many podcasters/suppliers. To wit:

    • Increasing advertising revenue for the entire industry requires a centralized player that can leverage a large userbase. Spotify is still a distant second to Apple in podcasts, but they are growing fast. Just as importantly, Spotify already has a strongly growing advertising business — again, larger than the entire podcast market — that it can extend to podcasts.
    • The open nature of podcasts means it is very difficult to monetize users directly; Spotify, though, has already built an entire infrastructure around monetizing users directly. Podcasts exclusive to Spotify can likely make meaningful money from Spotify subscribers that still gives Spotify far higher margin than music.

    This explains Spotify’s multi-prong approach:

    • Anchor provides a way to capture new podcasters, leading them either to Spotify advertising or, in the case of rising stars, to Spotify exclusives. Critically, because Spotify has access to all of the data, they can likely bring those suppliers on board at a far lower rate than they have to pay for established creators like Gimlet Media.
    • Spotify Advertising, as I just suggested, makes a strong play to be the dominant provider for the entire podcasting industry. Spotify Advertising is already operating at a far larger scale than Midroll, the incumbent player, and Spotify has access to the data of the second largest podcast player in the market.
    • Gimlet Media becomes an umbrella brand for a growing stable of Spotify exclusive podcasts. Critically, as I noted above, the majority of these podcasts come to Spotify not because Spotify pays them millions of dollars but simply because Spotify is better at monetizing than anyone else.

    This will be the determinant as to whether or not Spotify’s podcast gambit succeeds: being an Aggregator doesn’t simply mean acquiring a large pool of captive customers, it means controlling the value chain such that suppliers come on to your platform on your terms because you monetize them better than anyone else, even as you capture the excess value.

    To that last point, it’s worth highlighting this comment from Gimlet Media co-founder Matt Lieber to Peter Kafka on the Recode Media podcast:

    We did tell [Gimlet Media employees] that based on what we were talking about this would be a great thing for the company because really what everyone here is motivated by is making amazing shows for listeners who crave more, and…being acquired by Spotify puts us with the world’s largest audio platform that’s reaching more than 200 million people globally, so it’s a way for our storytelling and our work to have a lot more impact. So generally people are really excited about it.

    This is the Aggregator’s advantage: particularly when it comes to media, whether it be print, video, or audio, suppliers are often motivated to simply reach the most people and make a living doing so. It is a fundamentally short-term outlook that is entirely understandable and defensible. That, though, leaves the Aggregator with an arbitrage advantage: by controlling access to customers and, by extension, the most attractive means of monetization, Aggregators can offer the best relative deal to suppliers that is still, in absolute terms, a great deal for the Aggregator.

    To that end, it is worth considering if this is good for the podcasting industry generally. After all, to return to the web analogy, the price of the Internet finally monetizing effectively was the shift of content to centralized platforms like Facebook. Is the web better today than it was when we were punching monkeys?

    I do think the answer is yes, but I don’t mind if you disagree: granted, most supply has moved to Facebook and other social networks; it is no longer possible to build a viable web business with display ads. At the same time, the web is still as open as can be, which means there is room for new business models like subscriptions, a model that has only gotten started and is already producing far better content than the old mass market media model ever did (I’m obviously biased in this regard!).

    I can see a similar future for podcasts: Spotify, if they are successful, may end up being the biggest player, but that doesn’t mean new and different business models that directly link suppliers and consumers won’t emerge. It will, in other words, look like everything else touched by the Internet: very large winners on one end, and small niche winners on the other.

    1. This number was slightly inflated due to a one time accounting change ↩

    2. iTunes is very important to podcasting, but it is only a directory of podcasts that are hosted elsewhere; that means it is not a means to collect user data ↩

  • The BuzzFeed Lesson

    If you remove the societal impact, just for a moment, the story of publishers’ demise — first newspapers, and now digital-only companies like BuzzFeed and Huffington Post, which both announced significant layoffs last week — is rather banal: infinite competition combined with an inferior product resulted in failed business models.

    Infinite competition is the result of the Internet: any piece of content is only a tap away, a far cry from a world where geographic areas were dominated by a small number of newspapers. The inferior product is advertising: when newspapers were the only option, advertising inventory was scarce; now advertisers — which only paid for newspaper space as a matter of convenience, not principle — can reach the exact customers they want exactly where they spend most of their time and attention, namely Facebook and Google. And thus the failed business model: is it any surprise that commoditized content and non-competitive ad inventory did not work?

    The BuzzFeed Disappointment

    Still, the BuzzFeed layoffs in particular are disappointing, precisely because of the societal importance of journalism. Back in 2015 I wrote that BuzzFeed [Was] the Most Important News Organization in the World:

    Perhaps the single most powerful implication of an organization operating with Internet assumptions is that iteration – and its associated learning – is doable in a way that just wasn’t possible with print. BuzzFeed as an organization has been figuring out what works online for over eight years now, and while “The Dress” may have been unusual in its scale, its existence was no accident. What’s especially exciting about BuzzFeed, though, is how it uses that knowledge to make money…

    More importantly, with this model BuzzFeed has returned to the journalistic ideal that many — including myself — thought was lost with the demise of newspapers’ old geographic monopolies: true journalistic independence. Just as journalists of old didn’t need to worry about making money, just writing stories that they thought important, BuzzFeed’s writers simply need to write stories that people find important enough to share; the learning that results is how they make money. The incentives are perfectly aligned…The world needs great journalism, but great journalism needs a great business model. That’s exactly what BuzzFeed seems to have, and it’s for that reason the company is the most important news organization in the world.

    So what went wrong?

    BuzzFeed’s Pivot

    It was only two weeks after that post that CEO Jonah Peretti announced a pivot; from an interview with Peter Kafka of Recode:

    JP: As [full-stack media companies] started to become received wisdom, it started to stop being true, that it was the best way to build a company, and that happened largely because there was this jump to mobile and to mobile apps, and probably the majority of content consumption is happening inside of mobile apps. You think “Facebook traffic”, but in a way that’s people opening Facebook, seeing a BuzzFeed story, clicking a BuzzFeed story…That has started to create an environment where media is much more distributed…

    PK: So you built this system that was optimized for generating traffic and making money from stuff that happened on BuzzFeed.com and now you’re realizing that’s not what you want to do.

    JP: What we realized is that that was just one piece of our business…What I’ve been doing is meeting with every team in BuzzFeed with this little chart that is our model for making content that people love — News, Buzz, Life, Video, Lists, Quizzes, all different types of content, and have great tools for making content that people love — and then we send that content to various places. We send it to our own websites and to our own apps, which are owned-and-operated properties and remain important to us, where we have a certain ability to get data and learn from what we’re doing, but we also send it natively to other platforms like YouTube, or Facebook.

    2015 was the year that Facebook unveiled Instant Articles: publishers could put their content directly on Facebook, and Facebook, at least in theory, would help them monetize it. That seemed like a great deal! Facebook, for reasons I laid out in Popping the Publishing Bubble, was much better at advertising than any publishing company could hope to be:

    In the pre-Internet era publishers had it easy: on one hand, they employed journalists whose goal it was to reach as many readers as possible. On the other, they were largely paid by advertisers, whose goal was to reach as many potential customers as possible. The alignment — reach as many X as possible — was obvious, and profitable for the publishers in particular.

    A drawing of Pre-Internet Publishing

    […]

    The shift from paper to digital meant publications could now reach every person on earth (not just their geographic area), and starting a new publication was vastly easier and cheaper than before…The increase in competition destroyed the monopoly, but it was the divorce of “readers” from “potential customers” that prevented even the largest publishers from profiting much from the massive amounts of new traffic they were receiving. After all, advertisers don’t really care about readers; they care about identifying, reaching, and converting potential customers. And, by extension, this meant that differentiating ad inventory depended less on volume and much more on the degree to which a particular ad offered superior targeting, a superior format, or superior tracking.

    A drawing of The Post-Internet Bifurcation of Incentives of Publishers and Advertisers

    […]

    The above graph shows the inefficiency of this arrangement: publishers and ad networks are locked in a dysfunctional relationship that doesn’t serve readers or advertisers, and it’s only a matter of time until advertisers — which again, care only about reaching potential customers, wherever they may be — desert the whole mess entirely for new, more efficient and effective advertising options that put them directly in front of the people they care about. That, first and foremost, is Facebook…

    A drawing of Facebook As a More Efficient Advertising Option

    With Instant Articles it appeared that the social network would share the spoils: Facebook collects the advertising money, and publishers that embrace the platform share in the reward.

    The core problem for BuzzFeed is that never really happened: Instant Articles relied on the Facebook Audience Network, not Facebook’s core News Feed ad product, and nearly all of Facebook’s energy went into the latter. Companies that embraced Instant Articles — and, in the case of BuzzFeed, built their business models around them — were left earning pennies, mostly on programmatic advertising.

    Complete Commoditization

    For the record, I was completely wrong about the degree to which Facebook would help publishers monetize Instant Articles: it seemed to me that it was in Facebook’s interest to create sustainable models for quality content that lived directly on its platform. Sure, the company would be giving up a slice of its revenue, but the impact on the overall user experience generally and establishing Facebook as the center of not just the consumption of content but the monetization of content specifically would be powerful moats.

    The truth, though, is that the short-term incentives to maximize revenue, primarily through News Feed ads that Facebook kept for itself, were irresistible, and besides, the company had other fish to fry: Snapchat was looming as a threat through 2015, and by 2016 the company was starting to warn that ad loads were saturating. Quarterly growth was very much the priority, and once Snapchat was neutralized, was a content-based moat really necessary?

    I suspect, thought, that there is a more fundamental reason why BuzzFeed’s strategy was untenable. I wrote about the Conservation of Attractive Profits in the context of Netflix back in 2015:

    The Law of Conservation of Attractive Profits,1 [was] first explained by Clayton Christensen in his 2003 book The Innovator’s Solution:

    Formally, the law of conservation of attractive profits states that in the value chain there is a requisite juxtaposition of modular and interdependent architectures, and of reciprocal processes of commoditization and de-commoditization, commoditization, that exists in order to optimize the performance of what is not good enough. The law states that when modularity and commoditization cause attractive profits to disappear at one stage in the value chain, the opportunity to earn attractive profits with proprietary products will usually emerge at an adjacent stage.

    That’s a bit of a mouthful, but the example that follows in the book shows how powerful this observation is:

    If you think about it in a hardware context, because historically the microprocessor had not been good enough, then its architecture inside was proprietary and optimized and that meant that the computer’s architecture had to be modular and conformable to allow the microprocessor to be optimized. But in a little hand held device like the RIM BlackBerry, it’s the device itself that’s not good enough, and you therefore cannot have a one-size-fits-all Intel processor inside of a BlackBerry, but instead, the processor itself has to be modular and conformable so that it has on it only the functionality that the BlackBerry needs and none of the functionality that it doesn’t need. So again, one side or the other needs to be modular and conformable to optimize what’s not good enough.

    Did you catch that? That was Christensen, a full four years before the iPhone, explaining why it was that Intel was doomed in mobile even as ARM would become ascendent. When the basis of competition changed away from pure processor performance to a low-power system the chip architecture needed to switch from being integrated (Intel) to being modular (ARM), the latter enabling an integrated BlackBerry then, and an integrated iPhone four years later.2

    The PC is a modular system whose integrated parts earn all the profit. Blackberry (and later iPhones) on the other hand was an integrated system that used modular pieces.

    More broadly, breaking up a formerly integrated system — commoditizing and modularizing it — destroys incumbent value while simultaneously allowing a new entrant to integrate a different part of the value chain and thus capture new value.

    This is the theoretical explanation of what happened to publishers: newspapers previously integrated editorial and advertising:

    A drawing of The Old Media Model

    Then Facebook came along and integrated users and advertising:

    A drawing of The New Publication Media Model

    The result was the commoditization of content that I described above, which is exactly what you would predict given the integration elsewhere in the value chain. What I think is important, though, and under-appreciated by me (which is why I got Instant Articles wrong) is that the scale of integration — and correspondingly, the scale of commoditization — matters as well.

    In the case of Facebook the integration is absolute: the social network has two billion users, which gives the company not only a network effect, but also a gargantuan amount of user-generated content to populate the News Feed where the ads targeted with an even larger set of user data can be placed. It follows, then, that content suppliers are absolutely commoditized: Facebook doesn’t need to do anything to keep them on the platform, because where else will they go? Might as well keep the money for itself.

    Aggregation and Commoditization

    You see a similar dynamic with other large aggregators: Google’s Answer Box trades away the long-term viability of sites generating the content that makes Google useful in exchange for a short-term benefit that, yes, accrues to users, but accrues even more to Google, keeping those users on Google properties. And why not? It is not as if the web is running out of content — indeed, most website owners are paying Google supply sourcing agents SEO specialists to figure out how to get their content into those Answer Boxes in pursuit of whatever crumbs of traffic result.

    Amazon is following the same playbook: the company is ramping up its private label business, producing products that compete directly with companies that both sell to Amazon and are on the platform as 3rd-party merchants. After all, Amazon has integrated users and logistics: if suppliers pull their goods they will not pull customers away from Amazon; they’ll simply lose sales.

    It’s the same thing with Apple and the App Store: the most valuable customers in most markets are on the iPhone, which is why Apple can get away with charging 30% on digital goods that have nothing to do with the iPhone. Customers are not abandoning iOS just so they can have a better experience buying digital books, and Apple’s management certainly can’t afford a hit in Service revenue, particularly right now.

    That’s the thing, though: all of the big aggregators have been pursuing similar policies for years. To point to short-term pressure, whether that be falling China iPhone sales or Facebook ad load saturation is to miss the broader point: the more dominant an aggregator the more powerless the supply, and none of these companies are in the charity business.

    Avoiding Aggregators

    While I know a lot of journalists disagree, I don’t think Facebook or Google did anything untoward: what happened to publishers was that the Internet made their business models — both print advertising and digital advertising — fundamentally unviable. That Facebook and Google picked up the resultant revenue was effect, not cause. To that end, to the extent there is concern about how dominant these companies are, even the most extreme remedies (like breakups) would not change the fact that publishers face infinite competition and have uncompetitive advertising offerings.

    What is clear, though, is that the only way to build a thriving business in a space dominated by an Aggregator is to go around them, not to work with them. In the case of publishers, that means subscriptions, or finding ways to monetize, like the Ringer, beyond text.3 For web properties it means building destination sites that are not completely reliant on Google. For manufacturers it means building relationships with retailers other than Amazon and building brands that compel customers to go elsewhere. And for digital content providers…well, this is why I view Apple’s policies as the most egregious of all.

    As for BuzzFeed, it is not as if the company is dead: there is talk of mergers (which makes sense to reduce costs), and multi-pronged monetization strategies that emulate the success of the Tasty cooking videos: the company not only earns video advertising, but creates branded videos, has a line of branded cooking ware, and yes, takes programmatic advertising dollars on the companies owned-and-operated sites. Advertising can augment a publisher, but it’s hard to believe it can support one, even one expressly built for the Internet. That is now the realm of Aggregators.

    I wrote a follow-up to this article in this Daily Update.

    1. Later renamed the Law of Conservation of Modularity ↩

    2. As I’ve noted, the iPhone is in fact modular at the component level; the integration is between the completed phone and the software. Not appreciating that the point of integration (or modularity) can be anywhere in the value chain is, I believe, at the root of a lot of mistaken analysis about the iPhone in particular, including Christensen’s  ↩

    3. The Ringer is following the exact strategy I laid out in Grantland and the (Surprising) Future of Publishing  ↩

  • Netflix Flexes

    Bird Box, the Netflix original film, started streaming on December 21 while I was on vacation.1 That perhaps explains why the majority of my exposure to the Netflix Original came via NBA Twitter — and most of that exposure had absolutely nothing to do with the film, at least not directly.

    For example, the Memphis Grizzlies had a Bird Box-inspired contest for courtside seats:

    The Minnesota Timberwolves promoted an upcoming game with the Los Angeles (née Minneapolis) Lakers:

    I personally quite enjoyed this tweet from the Atlanta Hawks, which came in the middle of a 144-112 shellacking by my Milwaukee Bucks:

    Speaking of the Bucks, I couldn’t resist getting in on the meme either:

    The most meaningful Bird Box tweet, though, was from Netflix:

    That was quite the flex, and Netflix was only getting started.

    The Bird Box Flex

    There is an argument that Bird Box viewership numbers — which, as of Netflix’s earnings report last week, are up to 80 million Netflix member households — are not particularly meaningful. Sure, taking the wildly conservative assumption that one household=one viewer would mean that 80 million viewers was the equivalent of a box office haul of over $700 million2; increasing that to two viewers per household would imply an equivalent box office haul that would rank in the top 10 of all time.

    The problem, of course, is that none of those 80 million households actually paid explicitly for Bird Box: they got the movie for “free” with their Netflix subscription, and it seems like a stretch to think that most of them would have paid box office prices that are roughly as expensive as a month of the streaming service, to see the movie on purpose.

    This critique is both true and misses the point — three points, actually. First, it is not as if Netflix is counting on box office receipts: to point out that the company isn’t earning $700 million or $1.4 billion or whatever is even more of a moot point than the number of people that watched Bird Box. Secondly, and relatedly, Netflix is counting on subscription revenue. To that end, producing a piece of content that 58% of its subscriber base viewed in a single month is by definition a triumph (and yes, worth ~$700 million). Third, and most importantly, the success of Bird Box drives the virtuous cycle that Netflix has as an aggregator in multiple ways.

    Netflix the Aggregator

    Start with the most important side for an Aggregator — the demand side. Bird Box and other successful content does triple duty for Netflix:

    • For current customers, Bird Box provides two hours of entertainment and a pass into popular culture. It is a cost of goods expense.
    • For prospective customers, Bird Box makes Netflix more attractive for the same price. Or, to look at it another way, it lowers Netflix’s customer acquisition cost. It is a marketing expense.
    • For marginal customers, Bird Box is a reason to stay on the platform. It lowers Netflix’s customer retention cost. It is an operating expense.

    The latter two points are critical pieces of what makes an Aggregator an Aggregator; from Defining Aggregators:

    Once an aggregator has gained some number of end users, suppliers will come onto the aggregator’s platform on the aggregator’s terms, effectively commoditizing and modularizing themselves. Those additional suppliers then make the aggregator more attractive to more users, which in turn draws more suppliers, in a virtuous cycle.

    This means that for aggregators, customer acquisition costs decrease over time; marginal customers are attracted to the platform by virtue of the increasing number of suppliers. This further means that aggregators enjoy winner-take-all effects: since the value of an aggregator to end users is continually increasing it is exceedingly difficult for competitors to take away users or win new ones.

    This is in contrast to non-aggregator and non-platform companies that face increasing customer acquisition costs as their user base grows. That is because initial customers are often a perfect product-market fit; however, as that fit decreases, the surplus value from the product decreases as well and quickly turns negative. Generally speaking, any business that creates its customer value in-house is not an aggregator because eventually its customer acquisition costs will limit its growth potential.

    The question, then, is why do suppliers come onto Netflix’s platform?

    The first reason is that Netflix pays the most. From a supplier perspective that is certainly straightforward, but the question as to why Netflix can pay the most is an interesting one. There are multiple reasons:

    • First, Netflix is selling content to the entire world. That means its customer base is larger than other content buyers, giving Netflix greater buying power
    • Second, because of the demand-side dynamics I just described, Netflix is not simply selling to today’s subscribers, but the subscribers it anticipates attracting over the next several years, giving Netflix greater buying power again.
    • Third, because Netflix is not monetizing any particular piece of content in isolation, but rather as part of an overall subscription offering, it can more easily absorb failures on one hand (its customers have other shows to watch), and capture excess value on the other (because the lifetime value of customers is far greater than a single movie ticket). This means that Netflix’s risk, relative to traditional distributors, is significantly biased towards the upside, justifying higher prices.

    Secondly, Netflix has long appealed to the other motivations a supplier might have, particularly creative control. What the success of Bird Box shows, though, is that Netflix is uniquely capable of driving an audience as well. Sure, the company spent money on marketing Bird Box, but the reality is that Bird Box was popular because it was on Netflix. That is what drove views, and what drove Bird Box into the popular consciousness, and while all suppliers like getting paid, artists in particular like to be seen.

    And so we have a virtuous cycle: Netflix’s control of demand draws suppliers, which increases demand, and so it goes.

    The Pricing Flex

    Between the Bird Box announcement and Netflix’s earnings (where the company announced similar stellar viewership numbers for a number of other shows) came one more piece of news: Netflix is raising the price on U.S. subscribers by $2/month; new subscribers will pay the new price immediately, while existing subscribers will be phased in over the next several months. CEO Reed Hastings said on the company’s earnings interview:3

    With respect to the price changes…you’ll see that impact over the course of the year, and what that means is that will obviously impact the rate of net addition growth in the first half of the year. But commensurately, you also see ASP domestically improve over the course of the year and that’s what we think will drive an acceleration in revenue growth over the course of 2019. And that’s what also we believe drive operating margin higher sequentially over the course of the year to enable us to hit that 13% target for the full year.

    One of the obvious challenges for Netflix, particularly in the United States, is saturation. The company has 60 million subscribers in the U.S., which as of 2017 had 126 million households; given widespread account sharing, the company’s penetration is almost certainly well over 50%. There is still room for growth — around 100 million households have traditional multichannel video programming (i.e. the cable bundle) — but by definition households without Netflix are either exceptionally hard to reach (which is why Netflix has partnered with MVPDs to sell the service) or exceptionally frugal. Raising the price will certainly further inhibit the latter with their presumably high price elasticity.

    At the same time, Netflix is clearly confident that the price elasticity of its existing customers is very low: the company does not appear to expect any undue churn, which is reasonable given that previous price increases went off without a hitch. More broadly, it speaks to the importance of understanding how it is that Bird Box and other Netflix original content affects demand:

    The impact of Netflix's original content

    This is a graphical representation of the point I made above: existing customers are less price elastic, and marginal customers are more likely to stick around or sign up. Critically, this is a win for every part of the value chain: subscribers get more value, Netflix gets more revenue, and there is more money for suppliers.

    The Streaming Value Chain

    Much of this is obvious, at least at this point, but it is particularly noteworthy in the context of Netflix’s competitors. The traditional MVPD value chain, for example, has four participants: suppliers, networks, distributors (cable, satellite, or virtual), and end users. This made sense when the chief constraints were time and the need to actually run a cable into the back of an end user’s television, but it is a significant handicap in a world where there is no time constraint and where distribution is over the Internet.

    Consider the recent announcement from NBC; from CNBC:

    Comcast’s NBCUniversal plans to debut a free, ad-supported streaming service to anyone that subscribes to a traditional pay-TV service, including competitors such as Charter, AT&T, Cox and Dish, in the first quarter of 2020, the company announced Monday. For those that don’t subscribe to a pay-TV service, the streaming product, which will include 1,500 hours of NBC TV shows, such as SNL and Parks and Recreation, and hundreds of hours of Universal movies, will cost somewhere around $12 a month, a person familiar with the company’s plans told CNBC. The service will be run by Bonnie Hammer, the company announced Monday.

    This sounds suspiciously like TV Everywhere, the plan to allow MVPD subscribers to log into dedicated apps with their cable account. The problem is that the MVPD value chain ensured that TV Everywhere would be a complete mess:

    • Instead of there being one app, consumers had to download an app per network
    • Not all networks supported TV Everywhere, or did so inconsistently
    • Not all cable networks supported TV Everywhere, or did so inconsistently

    In short, TV Everywhere was an attempt to apply a value chain that was created around cable television to a fundamentally new paradigm, which introduced massive amounts of misalignment and inefficiency, most of which was borne by the end user. And oh, by the way, the old business model of advertising as well.

    The MVPD value chain

    Contrast that to Netflix which has created a value chain perfectly attuned to the streaming paradigm.

    Netflix's value chain

    Netflix’s integration of production and distribution also dramatically increases its flexibility and addressable markets when it comes to both supply and demand. On the demand side, as noted above, Netflix can reach users both all over the world as well as into the future. Just as importantly, on the supply side Netflix can accommodate all kinds of content on all kinds of deal terms. Hastings said on the earnings call:

    Our main goal is to make the best content. And we’ve said in previous quarters that that is a combination of several different business models depending on who owns the IP. So, what we’re going to do is make the best show and not be stuck on the business model, because the consumer really doesn’t understand that or we even want to spend any time thinking about it.

    So by way of example, last year, we had 140 different shows around the world that premiered on a network somewhere and on Netflix everywhere else in the world. Next year, it’s more to closer to 180. And these are combination of co-producing with local producers in other countries; it shows that then air on a network in that country and then premier on Netflix. But when I say co-production, I mean, we come in at the script stage, we come in at the first money stage, we’re involved creatively with the production of that show.

    Netflix has shows it owns completely, shows it own first-run rights to, hybrid shows like Hastings described, second-run shows — it runs the gamut. Critically, while some models are more profitable than others, all make the service more attractive to Netflix’s customers.

    This will be a particular challenge for a company like Disney: the company is staking a good portion of its future on its own streaming service driven by its own IP, but has not suggested a willingness to scale supply like Netflix has. That, by definition, will limit the company’s upside when it comes to consumer reach and also long-term pricing power.

    The Competition Flex

    These two points are related: tighter integration in the middle of the value chain means more flexibility and modularity on the edges. Netflix knows this, which is why the company didn’t even bother labeling Comcast or Disney its competitors. From the company’s letter to investors (emphasis mine):

    In the US, we earn around 10% of television screen time and less than that of mobile screen time. In other countries, we earn a lower percentage of screen time due to lower penetration of our service. We earn consumer screen time, both mobile and television, away from a very broad set of competitors. We compete with (and lose to) ​Fortnite​ more than HBO…

    Our growth is based on how good our experience is, compared to all the other screen time experiences from which consumers choose. Our focus is not on Disney+, Amazon or others, but on how we can improve our experience for our members.

    This is perhaps the biggest flex of all: Netflix is so confident in its position it is effectively stating that if customers choose to watch TV, they will choose Netflix. The company knows its model is that much better.

    I wrote a follow-up to this article in this Daily Update.

    1. Bird Box premiered at the AFI Fest on November 12, 2018, and had a very limited theatrical release on December 14 ↩

    2. Based on an average 2018 ticket price of $9.14 ↩

    3. As an aside, Netflix’s “Earnings Interview”, in which one analyst is allowed to ask questions (as opposed to Q&A from a number of analysts), seems like an unnecessary attempt to control the narrative from a company that, as this optimistic analysis suggests, doesn’t seem to have anything to hide ↩

  • AWS, MongoDB, and the Economic Realities of Open Source

    In 1999, music industry revenue in the United States peaked at $14.6 billion (all numbers are from the RIAA). It is important to be precise, though, about what was being sold:

    • $12.8 billion was from the sale of CDs
    • $1.1 billion was from the sale of cassettes
    • $378 million was from the sale of music videos on physical media
    • $222.4 million was from the sale of CD singles

    In short, the music industry was primarily selling plastic discs in jewel cases; the music encoded on those discs was a means of differentiating those pieces of plastic from other ones, but music itself was not being sold.

    This may sounds like a stupid distinction, but it explains what happened after that peak:

    U.S. music industry sales over time

    Music industry revenue plummeted, even as the distribution and availability of music skyrocketed: the issue is that people were no longer buying plastic discs, which is what the music industry was selling; they were simply downloading music directly.

    Selling Convenience

    The problem is that recorded music has always been worthless: once a recording is made, it can be copied endlessly, which means the supply is effectively infinite; it follows that to capture value from a recording depends on the imposition of scarcity. That is exactly what plastic discs were: a finite supply of a physical good differentiated by their being the most convenient way to get music. Pirating MP3s from sites like Napster or its descendants, though, was even more convenient — and cheaper.

    As you can see from the chart, the industry started to stabilize in 2010, and in 2016 returned to growth; 2018 looks to be up around 10% from 2017’s $8.7 billion number, and it seems likely the industry will pass that 1999 peak in the not-too-distant future.

    What happened is that the music industry — prodded in large part by Spotify, and then Apple — found something new to sell. No, they are still not selling music; in fact, they are beating piracy at its own game: the music industry is selling convenience. Get nearly any piece of recorded music ever made, for a mere $10/month.

    DocumentDB (with MongoDB compatibility)

    Last week, from the AWS blog:

    Today we are launching Amazon DocumentDB (with MongoDB compatibility), a fast, scalable, and highly available document database that is designed to be compatible with your existing MongoDB applications and tools. Amazon DocumentDB uses a purpose-built SSD-based storage layer, with 6x replication across 3 separate Availability Zones. The storage layer is distributed, fault-tolerant, and self-healing, giving you the the performance, scalability, and availability needed to run production-scale MongoDB workloads.

    The specifics of MongoDB and now DocumentDB are not particularly important to this article; basically, MongoDB created a type of database that is more flexible and better suited to large1 amounts of both structured and unstructured data, making it useful for large scale applications that traditional relational databases were never designed to accommodate.

    And now you can run it on AWS. Kind of.

    Open Source Licensing

    Like an increasing number of such projects, MongoDB is open source…or it was anyways. MongoDB Inc., a venture-backed company that IPO’d in October, 2017, made its core database server product available under the GNU Affero General Public License (AGPL).2

    AGPL is a close relative of the GPL, the copyleft license created by Richard Stallman. “Copyleft” means that the license allows for the free distribution, use, and modification of copyrighted material (in this case software), with the stipulation that those same rights extend to all derivative works; that means that any project built using GPL code must itself have a GPL license. This is in contrast to “permissive” open source licenses that allow others to use the copyrighted material however they wish, without a stipulation that derivative works also be open-sourced. AGPL extended the GPL to apply to software accessed over a network; since the software is only being used, not copied, the GPL would not triggered, but the end result is even more onerous than the GPL.

    Both GPL and especially AGPL tend to be very problematic for companies: Apple, for example, does not allow software licensed with the GPL on the App Store, because the App Store requires that apps be licensed for a single user; apps with permissive licenses are fine — their license can be replaced — but the GPL, once applied, cannot be removed. AGPL is worse, because its provisions are triggered by users simply using the software; that’s why Google bans its use internally. The company notes in its open source documentation:

    The license places restrictions on software used over a network which are extremely difficult for Google to comply with. Using AGPL software requires that anything it links to must also be licensed under the AGPL. Even if you think you aren’t linking to anything important, it still presents a huge risk to Google because of how integrated much of our code is. The risks heavily outweigh the benefits.

    There is one addendum to the policy:

    In some cases, we may have alternative licenses available for AGPL licensed code.

    This is MongoDB’s business.3

    MongoDB’s Business Model

    MongoDB explained in their S-1:

    We believe we have a highly differentiated business model that combines the developer mindshare and adoption benefits of open source with the economic benefits of a proprietary software subscription business model. To encourage developer usage, familiarity and adoption of our platform, we offer Community Server as an open source offering, analogous to a “freemium” offering. Community Server is a free-to-download version of our database that does not include all of the features of our commercial platform. This allows developers to evaluate our platform in a frictionless manner, which we believe has contributed to our platform’s popularity among developers and driven enterprise adoption of our subscription offering…

    Unlike software companies built around third-party open source projects, we own the intellectual property of our offerings since we are the creators of the software, enabling our proprietary software subscription business model…Our primary subscription package is MongoDB Enterprise Advanced, our comprehensive offering for enterprise customers that can be run in the cloud, on-premise or in a hybrid environment. MongoDB Enterprise Advanced includes our proprietary database server, advanced security, enterprise management capabilities, our graphical user interface, analytics integrations, technical support and a commercial license to our platform. We also offer MongoDB Atlas, our cloud hosted database-as-a-service, or DBaaS, offering that includes comprehensive infrastructure and management of our Community Server offering.

    Basically, MongoDB sells three things on top of its open source database server:

    • Additional tools for enterprise companies to implement MongoDB
    • A hosted service for smaller companies to use MongoDB
    • Legal certainty

    The importance of this last one can not be overstated: MongoDB’s enterprise version and hosted service are not governed by the AGPL — or, as of late last year, a new MongoDB-created license called the Server Side Public License (SSPL). The SSPL is like the AGPL on steroids: it compels companies selling MongoDB-as-a-service to not only open-source their modifications, but also open-source their entire stack.4

    What AWS Sells

    The largest company selling software-as-a-service is, of course, Amazon. That, though, does not mean that Amazon is selling “software.” The reality is that software is no different than music: it is infinitely reproducible, and thus, in isolation, worth nothing.

    Instead, the value of software is typically realized in three ways:

    • First is hardware. The most famous example is the iPhone, which is the only way to obtain iOS, but there are countless other examples.
    • Second is licenses. This was Microsoft’s core business for decades: licenses sold to OEMs (for the consumer market) or to companies directly (for the enterprise market). Indeed, there is a bit of irony in that both Microsoft and open source, for all their historical opposition to each other, both depended on copyright, strong legal regimes, and companies doing the right thing.
    • Third is software-as-a-service. This is Microsoft’s new model, as well as Amazon’s, and almost all new enterprise software companies.5 In this case what is being sold is not the software per se, but rather the utility of the software: the company doing the selling does everything else, including making the software available reliably.

    With that in mind, read again what AWS announced last week:

    The storage layer is distributed, fault-tolerant, and self-healing, giving you the the performance, scalability, and availability needed to run production-scale MongoDB workloads.

    AWS is not selling MongoDB: what they are selling is “performance, scalability, and availability.” DocumentDB is just one particular area of many where those benefits are manifested on AWS.

    Make no mistake: these benefits are valuable. There is a secular shift in enterprise computing moving to the cloud, not because it is necessarily cheaper (although costs are more closely aligned to usage), but because performance, scalability, and availability are hard problems that have little to do with the core competency and point of differentiation of most companies.

    Those are, though, the core competency of AWS, which can bring unmatched scale to bear on solving them: by effectively operating the servers for millions of customers Amazon can apply more resources to all of those issues than any one company could on its own, as well as develop its own customer architecture, from datacenter software down to custom chips (and drive a hard bargain for hardware from suppliers like Intel).

    The result is that “performance, scalability, and availability” is a tremendously attractive business: the more customers AWS has not only drive that much more recurring revenue, but also deepen AWS’ moat by allowing the company to bring that many more resources to bear on ever more obscure use cases, making AWS that much more attractive to new customers. Microsoft is competing but is a distant second; Google is even further behind. In fact, even MongoDB’s managed service runs on the three giants: it simply makes no sense to go it alone.

    The Open Source Conundrum

    Thus we have arrived at a conundrum for open source companies:

    • MongoDB leveraged open source to gain mindshare.
    • MongoDB Inc. built a successful company selling additional tools for enterprises to run MongoDB.
    • More and more enterprises don’t want to run their own software: they want to hire AWS (or Microsoft6 or Google) to run it for them, because they value performance, scalability, and availability.

    This leaves MongoDB Inc. not unlike the record companies after the advent of downloads: what they sold was not software but rather the tools that made that software usable, but those tools are increasingly obsolete as computing moves to the cloud. And now AWS is selling what enterprises really want.

    Worse, because AWS doesn’t have access to MongoDB (it is only matching the API) it only supports MongoDB 3.6; the current version is 4.0.5. It is possible that if AWS’ service becomes popular MongoDB will effectively stagnate: sure, you can get a better version from MongoDB Inc., but then you have to manage it yourself or go the effort to tie in all of your AWS services with MongoDB’s offering (then again, the potential for differentiation may be MongoDB’s salvation, and an important lesson for other companies).

    Not that permissive licensing would necessarily help: Redis Labs offers its Redis database under a permissive license; that means that AWS’ offering is usually up-to-date, which is good for Redis development, but doesn’t help Redis Labs make any money. That compelled Redis Labs to change the licensing on its add-on modules to add the “Commons Clause”; this compels service providers to pay for their use, effectively making them proprietary software.

    It’s hard to not be sympathetic to MongoDB Inc. and Redis Labs: both spent a lot of money and effort building their products, and now Amazon is making money off of them. But that’s the thing: Amazon isn’t making money by selling software, they are making money by providing a service that enterprises value, and both MongoDB and Redis are popular in large part because they were open source to begin with.

    Economic Realities and the Future

    Little of what I wrote is new to folks in the open source community: the debate over the impact of cloud services on open source has been a strident one for a while now. I think, though, that the debate gets sidetracked by (understandable) discussions about “fairness” and what AWS supposedly owes open source. Yes, companies like MongoDB Inc. and Redis Labs worked hard, and yes, AWS is largely built on open source, but the world is governed by economic realities, not subjective judgments of fairness.

    And that is why I started with music: it wasn’t necessarily “fair” that music industry sales plummeted, and yes, companies like Apple with its iPod business made billions off of piracy. The only reality that mattered, though, was that music itself, thanks to its infinite reproducibility, was as pure a commodity as there could be.

    It’s the same situation with software: bits on a disk are fundamentally free — just ask Richard Stallman. In his seminal essay Why Software Should Be Free Stallman wrote:7

    A copy of a program has nearly zero marginal cost (and you can pay this cost by doing the work yourself), so in a free market, it would have nearly zero price. A license fee is a significant disincentive to use the program. If a widely useful program is proprietary, far fewer people will use it.

    It is easy to show that the total contribution of a program to society is reduced by assigning an owner to it. Each potential user of the program, faced with the need to pay to use it, may choose to pay, or may forego use of the program. When a user chooses to pay, this is a zero-sum transfer of wealth between two parties. But each time someone chooses to forego use of the program, this harms that person without benefiting anyone. The sum of negative numbers and zeros must be negative.

    But this does not reduce the amount of work it takes to develop the program. As a result, the efficiency of the whole process, in delivered user satisfaction per hour of work, is reduced.

    This tradeoff is inescapable, and it is fair to wonder if the golden age of VC-funded open source companies will start to fade (although not open source generally). The monetization model depends on the friction of on-premise software; once cloud computing is dominant, the economic model is much more challenging.

    That, though, should give pause to AWS, Microsoft, and Google. It is hard to imagine them ever paying for open source software, but at the same time, writing (public-facing) software isn’t necessarily the core competency of their cloud businesses. They too have benefited from open-source companies: they provide the means by which their performance, scalability, and availability are realized. Right now everyone is winning: simply following economic realities could, in the long run, mean everyone is worse off.

    I wrote a follow-up to this article in this Daily Update.

    1. “Mongo” comes from the word humongous ↩

    2. I’m sorry, but this next bit is going to be dry; bear with me please ↩

    3. To be clear, I’m not saying that Google has a license; rather, that MongoDB offers alternative licenses ↩

    4. I’m not going to get into the SSPL, but it is very controversial: many detractors argue it is not an open source license because it does not abide by the freedom to run a program for any purpose, and it may not not be enforceable ↩

    5. I’m using software-as-a-service as an umbrella term for infrastructure-as-a-service and platform-as-a-service ↩

    6. Which, by the way, has its own MongoDB compatible offering ↩

    7. To be clear, I don’t agree with Stallman on a whole host of things; that doesn’t diminish his importance as a thinker or influence on the industry, though, or his insights on the nature of software ↩

  • Apple’s Errors

    As rare as last week’s Apple revenue warning from CEO Tim Cook may have been — the company last issued a revenue warning in June 2002 — the company has had other bad quarters in the iPhone era. Look no further than the stock chart:

    Apple's stock price in the iPhone era

    Three troughs stand out:

    • In fiscal year 2013 (the iPhone 5 cycle),1 Apple’s year-over-year revenue growth slowed to 18%, then 11%, 1%, and 4%; this was a dramatic slowdown from 73%, 59%, 23%, and 27% the year before. Worse, net income growth actually went negative (0%, -18%, -22%, -9%) thanks to a significant drop in margin.
    • In fiscal year 2016 (the iPhone 6S cycle), Apple’s year-over-year revenue growth went negative (2%, -13%, -15%, -9%); again, net income was worse (2%, -22%, -27%, -19%), thanks in part to a $2 billion inventory write-off.
    • This year does project to be the worst first quarter of all three: a -5% revenue decline, and -1% net income decline; this decline comes after last quarter’s announcement that Apple would no longer disclose unit sales, which precipitated the current slide in the stock price.

    What makes this quarter seem so much worse was both the already negative sentiment surrounding the shift in Apple’s reporting (the presumption being the company wanted to hide declining unit sales), and also the fact that Apple’s management forecast was so off: here is CFO Luca Maestri on last quarter’s earnings call:

    As we move ahead into the December quarter, I’d like to review our outlook, which includes the types of forward-looking information that Nancy referred to at the beginning of the call. We have the strongest lineup ever as we enter the holiday season and we expect revenue to be between $89 billion and $93 billion, a new all-time record.

    In fact, the only record, such that there was, was the size of the miss. So what went wrong?

    On Confirmation Bias

    If you will forgive a brief aside, this article requires a few very large caveats: first, Apple has not yet released its final quarter numbers, had its earnings call, or filed it’s 10-Q; there is a lot of information still to come.

    Secondly, thanks in part to the lack of information, this miss is catnip for confirmation bias: everyone has their pet theory about what Apple is doing wrong or how they will ultimately fail, and it has been striking the degree to which this revenue warning has been breezily adapted to show that said critics were right all along (never mind that many of those critics trotted out the exact same explanations in 2013 and 2016).

    Third, well, I happen to think that I am right as well: I believe that Apple’s management made three critical errors in their forecast for this last quarter that were predictable precisely because they had made the same errors before — errors that I wrote about at the time. In other words, I am very much susceptible to confirmation bias as well.

    That noted, if indeed I am right, then that is good news for Apple: I suspect the company is in better shape than the last week of hysteria suggests.

    Error 1: China and ‘S’ Cycles

    The most important takeaway from the revenue warning is that the vast majority of the problem in Apple’s forecast comes from Greater China. From Cook’s letter:

    While we anticipated some challenges in key emerging markets, we did not foresee the magnitude of the economic deceleration, particularly in Greater China. In fact, most of our revenue shortfall to our guidance, and over 100 percent of our year-over-year worldwide revenue decline, occurred in Greater China across iPhone, Mac and iPad…

    The problem was specifically around the iPhone:

    Lower than anticipated iPhone revenue, primarily in Greater China, accounts for all of our revenue shortfall to our guidance and for much more than our entire year-over-year revenue decline. In fact, categories outside of iPhone (Services, Mac, iPad, Wearables/Home/Accessories) combined to grow almost 19 percent year-over-year.

    That this exact quarter would be challenging for Apple is exactly what I predicted in May 2017 in Apple’s China Problem; specifically:

    • In most of the world, Apple is differentiated first-and-foremost by its integration between hardware and software; the company has a “monopoly” on iOS, which allows it to sell its hardware at much higher prices than the competition.
    • However, in China iOS is much less of a lock-in, thanks to the dominance of cross-platform Chinese-specific services, particularly WeChat (WeChat, while the most important factor, is not the only one: indeed, given that Android in China is specifically tuned to the Chinese market by Chinese OEMs, iOS is if anything a hindrance).
    • The net result is that Apple in China competes not on the basis of integration, but rather on the attractiveness of its hardware; in other words, Apple is, to far greater degree in China than anywhere else, simply another OEM.

    I wrote at the time:

    For the day-to-day lives of Chinese there is no penalty to switching away from an iPhone. Unsurprisingly, in stark contrast to the rest of the world, according to a report earlier this year only 50% of iPhone users who bought another phone in 2016 stayed with Apple:

    This is still better than the competition, but compared to the 80%+ retention rate Apple enjoys in the rest of the world, it is shockingly low, and the result is that the iPhone has slid down China’s sales rankings: iPhone sales were only 9.6% of the market last year, behind local Chinese brands like Oppo, Huawei and Vivo. All of those companies sold high-end phones of their own; the issue isn’t that Apple was too expensive, it’s that the iPhone 6S and 7 were simply too boring.

    At the end I concluded that Apple’s next phone — what turned out to be the iPhone X — would return the company to growth in China, and so it did: Apple was up 11%, 21%, 19%, and 16% last fiscal year, after declining by double digits six of the previous eight quarters. The other half of that prediction, though, was that the next ‘S’ model, with only component upgrades in the same form factor, would struggle; that is exactly what appears to have happened.

    To be sure, there are absolutely other issues in China, particularly the country’s significant economic slowdown as well as the possibility of anti-U.S. company sentiment thanks to the ongoing trade war and the arrest of Huawei’s CFO. I strongly suspect, though, that those macroeconomic factors made what would have been a tough quarter for Apple regardless that much worse; to put it another way, Apple is far more exposed to challenging macroeconomic conditions in China than they are elsewhere thanks to their relative lack of a moat.

    There are two adjustments Apple needs to make to avoid this error in the future: first, and most obviously, the company needs to be far more pessimistic with regard to its China forecasts in ‘S’ model years. Second, management needs to appreciate that the plane of competition in China is different than the rest of the world: the company is a luxury brand, but only in terms of hardware. If anything, iOS in China needs to cater more to the local market; as far as hardware, perhaps it is time for the ‘S’ strategy to be retired.

    Error 2: Non-Flagship iPhones

    There is one complicating factor in the last piece of analysis: Apple’s provided their last forecast on November 1, a full month into the quarter; did they not see this massive China miss coming? Perhaps the economy simply crashed in the last two months?

    That may be true, but I don’t think it is the entire explanation; Apple also tripped itself up with the staggered release of iPhones both this year and last:

    • In September 2017 (FY2017 Q4) Apple released the non-flagship iPhones 8 and 8 Plus
    • In November 2017 (FY2018 Q1) Apple released the flagship iPhone X
    • In September 2018 (FY2018 Q4) Apple released the flagship iPhones XS and XS Max
    • In (late) October 2018 (FY2019 Q1) Apple released the iPhone XR

    This schedule resulted in two blind spots for Apple: first, the company’s FY2018 Q4 results in China were almost certainly artificially high, thanks to Error 1. Of course the iPhone XS should have a strong year-over-year comp to the iPhone 8, but that comp was likely particularly extreme in China.

    Second, to the extent that iPhone XS sales slowed in October, Apple likely expected the iPhone XR to pick up the slack;2 I strongly suspect the XR failed to live up to expectations.

    This too, though, should have been predictable: sure, from a feature perspective the XR seemed remarkably competitive with the XS, but we have ample evidence that iPhone buyers want the best possible iPhone. After this year’s iPhone keynote I wrote:

    There is, of course, the question of cannibalism: if the XR is so great, why spend $250 more on an XS, or $350 more for the giant XS Max? This is where the iPhone X lesson matters. Last year’s iPhone 8 was a great phone too, with the same A11 processor as the iPhone X, a high quality LCD screen like the iPhone XR, and a premium aluminum-and-glass case (and 3D Touch!). It also had Touch ID and a more familiar interface, both arguably advantages in their own right, and the Plus size that so many people preferred.

    It didn’t matter: Apple’s best customers, not just those who buy an iPhone every year, but also those whose only two alternatives are “my current once-flagship iPhone” or “the new flagship iPhone” are motivated first-and-foremost by having the best; price is a secondary concern. That is why the iPhone X was the best-selling smartphone, and the iPhone 8 — which launched two months before the iPhone X — a footnote.

    It remains to be seen the extent to which this is the case globally, but the market where having the flagship matters most has always been China. iPhone XS sales slowing and not being picked up by the just-launched XR certainly explain the timing of the missed forecast.

    Error 3: iPhone Destiny

    This gets at the third error made by Apple management, and arguably the most concerning: the assumption that iPhone growth is inevitable.

    This was seen most clearly during the iPhone 6 cycle, when Cook insisted on earnings call after earnings call — I documented his statements in this Daily Update — that Apple’s record-breaking sales were not an abnormally large number people buying new iPhones sooner than they would have thanks to the large screen. In fact, it turned out that is exactly what was happening, which is why 6S sales were such a disappointment. I concluded:3

    I know I’m kind of harping on this point, but in fact I find any possible explanation for this inconsistency very troubling: either Cook was purposely overselling the upgrade narrative last year, which would not only be duplicitous but also dumb, given that he would only be setting up Apple for a fall this cycle; or even as late as last year Cook was out of touch with how the iPhone upgrade cycle actually works, or how it may have changed over time.

    I strongly suspect it is the latter explanation, and while that is concerning, it’s also understandable; the implication of my ongoing contention that the iPhone has now picked all of the “low-hanging fruit” of growth is that iPhone growth had multiple causes: certainly the inherent quality and new features of each annual iPhone model played a role, but an arguably bigger factor was simply distribution — getting the iPhone onto more carriers in more countries. Indeed, I strongly suspect the predictable impact of increased distribution helps explain why Apple’s earnings forecasts were so eerily accurate for so many years.

    Apple, though, has been a lot less accurate for the last five quarters: the iPhone 6 sold better than they expected in nearly every quarter, and now the iPhone 6S is selling worse (note the Maestri forecast errors I highlighted above); along similar lines, Apple seems to have underestimated iPhone SE demand to a significant degree. Ultimately, while I think Apple still has the advantage when it comes selling people their second (or third or nth) smartphones, what I think we are seeing is that it’s a lot more difficult to determine when exactly that sale will occur, and Apple itself is only now coming to grips with that.

    This gets at the rest of the miss — the non-China parts, especially. Cook cited the end of carrier subsidies (largely an old story at this point, to be fair), a stronger U.S. dollar, and customers getting new batteries instead of new iPhones. It’s a bit of a hodgepodge with one primary takeaway: convincing customers to upgrade “good enough” phones is both challenging and unpredictable, and Apple can’t simply assume it will happen at the rate it has previously.

    Reasons for Optimism

    The good news for Apple is that, to the extent these errors really were predictable, there is nothing structurally different about the company’s competitive position today versus six months ago, when the current stock slide began.

    • The next iPhone hardware revision should sell better in China, simply by virtue of being new (and the implication of it being easy to switch away from iOS is that it’s easy to switch back).
    • Customers still prefer Apple’s flagship iPhones, no matter how expensive they are.
    • Headwinds like currency and battery replacement programs will go away, and phones, thanks to their centrality in people’s lives as well as the greater likelihood of harm, will always have a faster replacement cycle than PCs.

    Meanwhile, the company’s Services business continues to grow, along with its installed base (including in China); the company is clearly putting more strategic emphasis in this area, effectively abandoning also-ran hardware products like HomePod and Apple TV to increase the reach of its services. I would expect significant announcements in this area through 2019.

    That is not to say the company is finished with hardware: wearables are a huge area of growth, as both AirPods and Apple Watch are big successes, and it seems likely that an augmented reality product is coming in the next few years. Nothing will match the iPhone, but that’s ok; the sky is not falling, only the stock.

    I wrote a follow-up to this article in this Daily Update.

    1. Keep in mind that Apple’s fiscal years start on October 1st ↩

    2. The company had less than a week’s worth of data about how the XR was selling by the time management made its forecast ↩

    3. I first made the case for abandoning the ‘S’ strategy in this Daily Update, writing:

      To that end, I do question how much longer Apple can afford to stick with the ‘S’ strategy. Again, big screens were such an important feature that it’s difficult to take away too much from the 6S’ poor year-over-year comparisons, but it seems reasonable to wonder if the structural expansions that increased the iPhone’s addressable market papered over the fundamentally weaker value proposition presented by the ‘S’ lines. To put it another way, one could argue the ‘S’ lines are introducing a holiday quarter-like dynamic into Apple’s earnings but on a two-year basis: we may not really tell know the iPhone is doing until a completely new model that will drive upgrades comes out.

      In fact, if you look at a two-year comparison, Apple’s revenue last quarter was up 7%, a perfectly acceptable result ↩

  • Holiday/Vacation Break: Weeks of December 24 and December 31

    Stratechery is taking a holiday and vacation break the weeks of December 24 and December 31. There will be no Weekly Article or Daily Updates. The Daily Update will resume on January 7.

    All new subscriptions made since November 26, including during this break, will have two weeks added to their subscriptions. See you in 2019!