Cloudflare on the Edge

Matthew Prince, at the end of his prepared remarks after Cloudflare’s recent earnings report, related a story from the company’s earliest days:

Back in 2010, right before Cloudflare’s first Board meeting and our launch, I got some advice from one of our early investors. He said running a company is a bit like flying an airplane. You want to make sure it’s well maintained at all times. And that when you’re flying, you keep the wheel steady and the nose 10 degrees about the horizon. That’s stuck with me, and we’ve designed Cloudflare for consistent and disciplined execution. That shows in quarters like the one we just had.

What is most important of all, though, is the destination that airplane is headed for.

TechCrunch Disrupt

The launch Prince referred to happened at TechCrunch Disrupt 2010; the entire video is worth a watch, but there are three highlights in particular. First, Prince — despite a three-minute technical delay — did an excellent job of laying out Cloudflare’s core value proposition:

Prince, a graduate of Harvard Business School, explicitly invoked HBS Professor Clayton Christensen while answering a question about competition:

The most memorable moment of the presentation, though, was Prince’s response to a seemingly anodyne question about when companies might grow out of Cloudflare’s offering:

Despite the audacity of Prince’s answer — Our vision is that we’re going to power the Internet — the company’s list of competitors in its 2019 S-1 seemed rather aspirational, in both breadth and scale:

Our current and potential future competitors include a number of different types of companies, including:

  • On-premise hardware network vendors, such as Cisco Systems Inc., F5 Networks, Inc., Check Point Software Technologies Ltd., FireEye, Inc., Imperva, Inc., Palo Alto Networks, Inc., Juniper Networks, Inc., and Riverbed Technology, Inc.;
  • Point-cloud solution vendors, including cloud security vendors such as Zscaler, Inc. and Cisco Systems Inc. through Umbrella (formerly known as OpenDNS), content delivery network vendors such as Akamai Technologies, Inc., Limelight Networks, Inc., Fastly, Inc., and Verizon Communications Inc. through Edgecast, domain name system vendors services such as Oracle Corporation through DYN, NeuStar, Inc., and UltraDNS Corporation, and cloud SD-WAN vendors; and
  • Traditional public cloud vendors, such as Amazon.com, Inc. through Amazon Web Services, Alphabet Inc. through Google Cloud Platform, Microsoft Corporation through Azure, and Alibaba Group Holding Limited through Alibaba Cloud.

The first two categories make sense; after all, Cloudflare’s value proposition from the beginning was speed and security, so of course they would grow up to compete with network and security vendors. It was that last bullet point, though, that even now leads to raised eyebrows: Cloudflare’s big quarter entailed $138 million in revenue; AWS, over the same period, made $150 million a day.

Cloudflare Disrupt

To understand why Cloudflare sees public cloud vendors as competitors it helps to go back to what made Cloudflare disruptive; Christensen wrote in The Innovator’s Dilemma:

Occasionally, however, disruptive technologies emerge: innovations that result in worse product performance, at least in the near-term. Ironically, in each of the instances studied in this book, it was disruptive technology that precipitated the leading firms’ failure. Disruptive technologies bring to a market a very different value proposition than had been available previously. Generally, disruptive technologies underperform established products in mainstream markets. But they have other features that a few fringe (and generally new) customers value. Products based on disruptive technologies are typically cheaper, simpler, smaller, and, frequently, more convenient to use.

That was basically Prince’s value proposition: Cloudflare’s CDN would be cheaper (free), simpler (just change DNS servers), smaller (only 5 servers to start), and more convenient (ridiculously easy!). And Cloudflare’s customers were definitely fringe:

What Cloudflare had in its favor, though, was the most potent advantage on the Internet: the service, much like Google a decade-earlier with its link-based ranking system, got better with use. This was because Cloudflare paired its content delivery network with DDoS protection; the latter was extremely attractive to websites, gave Cloudflare an in with ISPs who valued the protection to build point-of-presence servers around the world, and, critically, gave Cloudflare better-and-better data about how data flowed around the world (improving its service) even as it improved its CDN capabilities.

Cloudflare’s focus on security-for-free also meant its CDN was built on general-purpose hardware from the beginning; from the S-1:

To achieve the level of efficiency needed to compete with hardware appliances required us to invent a new type of platform. That platform needed to be built on commodity hardware. It needed to be architected so any server in any city that made up Cloudflare’s network could run every one of our services. It also needed the flexibility to move traffic around to serve our highest paying customers from the most performant locations while serving customers who paid us less, or even nothing at all, from wherever there was excess capacity.

As time went on those general purpose machines were used for more-and-more offerings beyond a CDN and DDoS protection; HHHypergrowth has a fantastic overview of everything Cloudflare is working on, and the article is daunting in length because Cloudflare’s portfolio is so vast. It is Cloudflare Workers, though, that are responsible for the big cloud players being in Cloudflare’s competitive set.

Cloudflare Workers

Cloudflare launched Workers seven years after the company’s launch at Disrupt; from the introductory blog post:

Cloudflare is about to go through a similar transition [as programmable CPUs]. At its most basic level, Cloudflare is an HTTP cache that runs in 117 locations worldwide (and growing). The HTTP standard defines a fixed feature set for HTTP caches. Cloudflare, of course, does much more, such as providing DNS and SSL, shielding your site against attacks, load balancing across your origin servers, and so much else.

But, these are all fixed functions. What if you want to load balance with a custom affinity algorithm? What if standard HTTP caching rules aren’t quite right, and you need some custom logic to boost your cache hit rate? What if you want to write custom WAF rules tailored for your application?

You want to write code.

We can keep adding features forever, but we’ll never cover every possible use case this way. Instead, we’re making Cloudflare’s edge network programmable. We provide servers in 117+ locations around the world — you decide how to use them.

Workers were extremely limited in functionality to start; just a bit of stateless Javascript code running in a V8 isolate, but as close to users as possible. In 2018 Cloudflare added a key-value store, giving Workers access to highly distributed eventually-consistent data storage; in 2020 the company introduced Workers Unbound, dramatically expanding Workers capabilities, and Durable Objects, which not only store data but also state, which means a single source of truth. Once again Cloudflare’s network comes to the rescue:

When using Durable Objects, Cloudflare automatically determines the Cloudflare datacenter that each object will live in, and can transparently migrate objects between locations as needed. Traditional databases and stateful infrastructure usually require you to think about geographical “regions”, so that you can be sure to store data close to where it is used.

Thinking about regions can often be an unnatural burden, especially for applications that are not inherently geographical. With Durable Objects, you instead design your storage model to match your application’s logical data model. For example, a document editor would have an object for each document, while a chat app would have an object for each chat. There is no problem creating millions or billions of objects, as each object has minimal overhead.

In Cloudflare’s example of a chat app, every individual conversation is an object, and that object is moved as close to the participants as possible; two people chatting in the U.S. would utilize a Durable Object in a U.S. data center, for example, while two in Europe would use one there. There is a bit of additional latency, but less than there might be with a centralized cloud provider. That’s ok, though, because the real advantage of Workers isn’t what Cloudflare thought it was.

Public Cloud Economics

The economics of public clouds are very straightforward: it makes far more sense for Amazon or Microsoft or Google to build and maintain data centers all over the world and rent out capacity than it does for companies for whom data centers are not their core competency to duplicate their efforts at a sub-scale level. It’s so compelling I labeled the current state The End of the Beginning:

This last point gets at why the cloud and mobile, which are often thought of as two distinct paradigm shifts, are very much connected: the cloud meant applications and data could be accessed from anywhere; mobile made the I/O layer available anywhere. The combination of the two make computing continuous.

A drawing of The Evolution of Computing

What is notable is that the current environment appears to be the logical endpoint of all of these changes: from batch-processing to continuous computing, from a terminal in a different room to a phone in your pocket, from a tape drive to data centers all over the globe. In this view the personal computer/on-premises server era was simply a stepping stone between two ends of a clearly defined range.

While this view of the omnipresent cloud is true for end users, the story is a bit more complicated for developers; if you want to set up a new instance you need to first select a region. AWS, for example, has twenty-five regions around the world:

AWS regions

Once you choose a region your actual app is geographically contained in that region. In theory that limitation gives an advantage to Cloudflare Workers; Prince wrote in a blog post:

Since we’re unlikely to make the speed of light any faster, the ability for any developer to write code and have it run across our entire network means we will always have a performance advantage over legacy, centralized computing solutions — even those that run in the “cloud.” If you have to pick an “availability zone” for where to run your application, you’re always going to be at a performance disadvantage to an application built on a platform like Workers that runs everywhere Cloudflare’s network extends.

The truth, though, is that this performance doesn’t matter very much for most applications. Stratechery’s podcast service runs in the US East (Ohio) region, for example, and it doesn’t really make a difference for me, despite the fact I’m halfway around the world. Price admitted as such:

But let’s be real a second. Only a limited set of applications are sensitive to network latency of a few hundred milliseconds. That’s not to say under the model of a modern major serverless platform network latency doesn’t matter, it’s just that the applications that require that extra performance are niche…People who talk a lot about edge computing quickly start talking about IoT and driverless cars. Embarrassingly, when we first launched the Workers platform, I caught myself doing that all the time.

Indeed, for almost all applications the public clouds were good enough, and again, the economics made any other choice a bad idea.

The Edge Opportunity

Earlier this year, in the wake of January 6, I wrote Internet 3.0 and the Beginning of (Tech) History; after raising the arguments from The End of the Beginning I noted:

In the case of the Internet, we are at the logical endpoint of technological development; here, though, the impasse is not the nature of man, but the question of sovereignty, and the potential re-liberation of megalothymia is the likely refusal by people, companies, and countries around the world to be lorded over by a handful of American giants.

As long as economics were all that mattered, we would only ever have the centralized cloud providers; the “limited set of applications” that needed minimal latency could pay a bit more to run on those blue AWS edge providers in the maps above. The point of that article, though, is that economics weren’t the only thing that mattered: going forward politics would be even more important.

Prince had the same realization; the blog post I have been quoting is entitled The Edge Computing Opportunity: It’s Not What You Think, and the chief benefits Prince cites are very much about politics:

Most computing resources that run on cloud computing platforms, including serverless platforms, are created by developers who work at companies where compliance is a foundational requirement. And, up until to now, that’s meant ensuring that platforms follow government regulations like GDPR (European privacy guidelines) or have certifications providing that they follow industry regulations such as PCI DSS (required if you accept credit cards), FedRamp (US government procurement requirements), ISO27001 (security risk management), SOC 1/2/3 (Security, Confidentiality, and Availability controls), and many more.

But there’s a looming new risk of regulatory requirements that legacy cloud computing solutions are ill-equipped to satisfy. Increasingly, countries are pursuing regulations that ensure that their laws apply to their citizens’ personal data. One way to ensure you’re in compliance with these laws is to store and process data of a country’s citizens entirely within the country’s borders.

The EU, India, and Brazil are all major markets that have or are currently considering regulations that assert legal sovereignty over their citizens’ personal data. China has already imposed data localization regulations on many types of data. Whether you think that regulations that appear to require local data storage and processing are a good idea or not — and I personally think they are bad policies that will stifle innovation — my sense is the momentum behind them is significant enough that they are, at this point, likely inevitable. And, once a few countries begin requiring data sovereignty, it will be hard to stop nearly every country from following suit.

This potential reality presents a big problem for Amazon, Microsoft, and Google: what scales on their side is the cloud as a whole, from management to interface to purchasing; individual developers are meant to stay in their regions. Yes, all three companies guarantee that data in one region won’t go elsewhere, but it’s a development nightmare: you have to maintain different apps with different data stores in different regions.

Cloudflare, meanwhile, can use the same capabilities that seamlessly transfer Durable Objects to the nearest data center, to follow local compliance data sovereignty laws at a granular level; from an announcement for Jurisdictional Restrictions for Durable Objects:

Durable Objects, currently in limited beta, already make it easy for customers to manage state on Cloudflare Workers without worrying about provisioning infrastructure. Today, we’re announcing Jurisdictional Restrictions for Durable Objects, which ensure that a Durable Object only stores and processes data in a given geographical region. Jurisdictional Restrictions make it easy for developers to build serverless, stateful applications that not only comply with today’s regulations, but can handle new and updated policies as new regulations are added…

By setting restrictions at a per-object level, it becomes easy to ensure compliance without sacrificing developer productivity. Applications running on Durable Objects just need to identify the jurisdictional rules a given Object should follow and set the corresponding rule at creation time. Gone is the need to run multiple clusters of infrastructure across cloud provider regions to stay compliant — Durable Objects are both globally accessible and capable of partitioning state with no infrastructure overhead.

Durable Objects are not, in-and-of-themselves, going to kill the public clouds; what they represent, though, is an entirely new way of building infrastructure — from the edge in, as opposed to the data center out — that is perfectly suited to a world where politics matters more than economics.

Internet 3.0

I actually already covered Cloudflare’s differentiated approach, albeit in passing, and by accident. Back in March, I interviewed Prince in the process of writing Moderation in Infrastructure; one thing that stood out to me was how his response to Internet fragmentation differed from Microsoft President Brad Smith, and Google Cloud CEO Thomas Kurian:

Smith:

I think, it’s a reflection of the fact that if you’re a global technology business, most of the time, it is far more efficient and legally compliant to operate a global model than to have different practices and standards in different countries, especially when you get to things that are so complicated. It’s very hard to have content moderators make decisions about individual pieces of content under one standard, but to try to do it and say, “Well, okay, we’ve evaluated this piece of content and it can stay up in the US but go down in France.” Then you add these additional layers of complexity that add both cost and the risk of non-compliance which creates reputational risk.

Kurian:

So far, we have tried to get to what’s common, and the reality is, Ben, it’s super hard on a global basis to design software that behaves differently in different countries. It is super difficult. And at the scale at which we’re operating and the need for privacy, for example, it has to be software and systems that do the monitoring. You cannot assume that the way you’re going to enforce ToS and AUPs is by having humans monitor everything, I mean we have so many customers at such a large scale. And so that’s probably the most difficult thing is saying virtual machines behave one way in Canada, and a different way in the United States, and a third way…I mean that’s super complicated.

Prince:

Everywhere in the world, governments have some political legitimacy, and they certainly have a lot more political legitimacy than I do…It’s important that we comply with the laws in each jurisdiction in which we operate. We should help our customers comply with the laws in each jurisdiction we operate…Germany can set whatever rules they want for Germany, but it has to be the rules inside of Germany.

And you can manage that okay. You can manage on a per country basis. You feel good about that?

Sure. I mean, for us, that’s easy. And then we can provide that to our customers as a function of what we’re doing. But I think that if you could say, German rules don’t extend beyond Germany and French rules don’t extend beyond France and Chinese rules don’t extend beyond China and that you have some human rights floor that’s in there.

Right. But given the nature of the internet, isn’t that the whole problem? Because, anyone in Germany can go to any website outside of Germany.

That’s the way it used to be, I’m not sure that’s going to be the way it’s going to be in the future. Because, there’s a lot of atoms under all these bits and there’s an ISP somewhere, or there’s a network provider somewhere that’s controlling how that flows and so I think that, that we have to follow the law in all the places that are around the world and then we have to hold governments responsible to the rule of law, which is transparency, consistency, accountability. And so, it’s not okay to just say something disappears from the internet, but it is okay to say due to German law it disappeared from the internet. And if you don’t like it, here’s who you complain to, or here’s who you kick out of office so you do whatever you do. And if we can hold that, we can let every country have their own rules inside of that, I think that’s what keeps us from slipping to the lowest common denominator.

The quotes aren’t perfectly comparable — you can read the full interviews to get the context — but it makes sense that Microsoft and Google (and presumably Amazon) would be very concerned about a world where individual countries make their own laws about what can be put on the Internet, or even seen. Theirs are services predicated on the superior economics that come from centralization; Cloudflare, on the other hand, is already doing all of its computing on the edge — data sovereignty rules are simply a variable. It’s “easy”.

This is why the direction of Cloudflare’s metaphorical plane is so fascinating: Cloudflare’s current addressable market of enterprise security and networking is significant, particularly as remote work has laid bare the problems with traditional approaches; the destination with outsized upside, though, is Internet 3.0, and the resultant need for a service that routes around obstacles, not from nuclear war, but sovereign governments.1

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

  1. As I explained last week, I use “Internet 3.0” instead of Web3 because the world beyond centralization isn’t only going to be about crypto; Cloudflare’s potential is a great example. []

Market-Making on the Internet

Two months ago I bemoaned how Web 2.0 had failed to live up to its promise in one crucial area; from The Web’s Missing Interoperability:

[From] the Wikipedia…definition:

Web 2.0…refers to websites that emphasize user-generated content, ease of use, participatory culture and interoperability (i.e., compatible with other products, systems, and devices) for end users.

Seventeen years on and there is more user-generated content than ever, in part because it is so easy to generate: you can type an update on Facebook, post a photo on Instagram, make a video on TikTok, or have a conversation on Clubhouse. That, though, points to Web 2.0’s failure: interoperability is nowhere to be found.

The context for that piece was Twitter’s preview of Super Follows, acquisition of Revue, and imminent launch of Spaces, all of which helped explain how centralization had won the web. Twitter’s latest acquisition, though, leverages centralization in a way that bodes well for realizing the “thousand flowers bloom” vision of Web 2.0.

Web 2.0 and the Aggregators

The first page of O’Reilly’s definition of 2.0 had a “Web 2.0 Meme Map”:

The Web 2.0 Meme Map from O'Reilly

Don’t worry about the low resolution and tiny type; what should feel familiar is this idea of an interlocking web of properties and services between which a user could flit to and fro, tying their content and data together with links. It turned out, though, that centralized services offered better user experiences that gained them critical masses of users, drawing suppliers onto the service on the service’s terms, attracting more users in a virtuous cycle; meanwhile, by virtue of being centralized, the service could gather better data about its users and offer a one-stop shop for advertisers. Aggregation Theory, in other words.

Aggregation was the antithesis of the Web 2.0 promise; the best suppliers could do was either subject themselves to the Aggregator’s terms and try and make the best of it (call it the BuzzFeed strategy) or work to build a direct connection with customers that went around the Aggregators (the New York Times strategy); Twitter, though, may be on the verge of offering a middle path: market-making.

Ad Agencies

Market-making in media isn’t a new concept; perhaps the best example is the traditional advertising agency. I explained back in 2017’s Ad Agencies and Accountability:

Few advertisers actually buy ads, at least not directly. Way back in 1841, Volney B. Palmer, the first ad agency, was opened in Philadelphia. In place of having to take out ads with multiple newspapers, an advertiser could deal directly with the ad agency, vastly simplifying the process of taking out ads. The ad agency, meanwhile, could leverage its relationships with all of those newspapers by serving multiple clients:

A drawing of The Pre-Internet Ad Agency Structure

It’s a classic example of how being in the middle can be a really great business opportunity, and the utility of ad agencies only increased as more advertising formats like radio and TV became available. Particularly in the case of TV, advertisers not only needed to place ads, but also needed a lot more help in making ads; ad agencies invested in ad-making expertise because they could scale said expertise across multiple clients.

In this case the ad agencies gave a single point of contact for advertisers on one side, and ad inventory sellers on the other, creating a market.

That was then, though; aggregation has been terrible for ad agencies for the same reason it has been bad for publishers: the more that advertising becomes centralized on Facebook and Google, whether on their sites or on programmatic exchanges, the fewer advertising dollars are available for the inventory that ad agencies used to abstract away for clients. And, as I noted in that article:

A drawing of The Post-Internet Ad Agency Structure

That’s a problem for the ad agencies: when there are only two places an advertiser might want to buy ads, the fees paid to agencies to abstract complexity becomes a lot harder to justify.

In this world, the more effective of the two strategies I noted above has clearly been the New York Times model, at least for the New York Times and its 7.5 million subscribers; Visual Capitalist put together this striking infographic last week showing how the New York Times dominated the subscription market (click through to see a veritcal version of the infographic that shows every publication listed):

An inforgraphic of news websites ranked by subscriptions
Licensed from Visual Capitalist

The problem is that there are a lot of publications in the world that would like to be supported by subscriptions, and a lot of readers in the world that would prefer to pay for ad-free content, but nobody is making a market. This is where Twitter is making its play.

Twitter Buys Scroll

From the Scroll blog:

Twitter is acquiring Scroll. The service will be going into private beta as we integrate into a broader Twitter subscription later in the year. We’re very excited!

Since launch last year, Scroll has proven that there’s a model that gives consumers a better experience and journalists a better future. Today when Scroll members visit hundreds of top sites like The Atlantic, The Verge, USA Today, The Sacramento Bee, The Philadelphia Inquirer or The Daily Beast, they get a site that feels built solely for them: a blazing fast experience that loads with no ads, no dodgy trackers and no chumboxes of clickbait. At the same time, publishers get to deliver a site that increases engagement and makes them more money than they would make from ads. It’s a better internet and we’ve proven the model works.

I’ve been a subscriber to Scroll for some time now, and it’s a great experience: while ad-blockers are oppositional to publishers, blocking their trackers and advertisements while users take content they refuse to help monetize, Scroll partnered with publications directly, such that Scroll subscribers were never served ads in the first place. Publishers would instead get paid their usage-based share of the Scroll user’s subscription fee, which Scroll claimed was more than a publication would have made by serving that user ads.

Scroll’s list of partner sites has been scrubbed from their website, but I recall it being pretty impressive; the real challenge for the startup, though, was acquiring users, which is the first reason why Twitter is a great match: Twitter not only has 199 million monetizable daily active users, but is also one of the places where users are most likely to click-through to publications.

Twitter between publishers and readers

Now imagine this model, but with Scroll’s business model on top:

Twitter using Scroll to monetize publishers

This is a great example of market-making in action: Twitter is taking its user base, which no one publication could realistically reach or monetize on its own, and re-distributing their subscription fee across publications that no one user could ever support individually.

This doesn’t necessarily mean that every publication will buy in; the New York Times, to take an obvious example, is doing quite well on its own, as are sovereign writers. Twitter, though, because of its outsized role in driving traffic to sites across the web, is uniquely positioned to bundle everything else together. Moreover, there are, as they say levels to this: Twitter could offer one price tier for no ads, and another price tier for getting past paywalls, and perhaps even individual site subscriptions above that.

Multi-Medium Creators

Spotify’s announcement last week, meanwhile, shows that markets need not be exclusive. Start with creators: today someone might build a following on Instagram, before branching out to YouTube or TikTok, much as an artist may have started out singing and ended up making movies. Today this ability to move across mediums extends down to even the smallest creators: even Stratechery is multi-medium, with text available on the web or via email, and podcasts on your phone.

What is new is that the business model can be more than fame (and, naturally, advertising). Subscriptions work for creators just as well as they do for publications (arguably better given an individual creator’s cost structure), but to-date that has meant that creators were limited to mediums that they could fully control — i.e. text and podcasts — and only on the open web; what Spotify made clear is that they want into this world. From Spotify’s Surprise:

For full disclosure, I have been briefed on the Open Access Platform, and Spotify has addressed all of my concerns; no, they won’t support arbitrary RSS feeds, but instead another open technology — OAuth. Some time soon Stratechery and Dithering subscribers will be able to link their subscriptions to their Spotify accounts, and Spotify isn’t going to charge a dime — they will be my customers from email address to credit card. Spotify Chief R&D Officer Gustav Söderström told me, “Having all of audio on Spotify means meeting independent creators on their terms, not ours.”

In this case Spotify isn’t market-making: rather, it is recognizing that creators are going to want to make their own markets, pulling their fans from medium to medium and service to service, and they want to make sure they are plugged in to that:

The creator carrying audiences to different mediums

What is neat about markets is that they create the conditions for win-win outcomes; Spotify aligning with creators doesn’t hurt Spotify’s core business, it enhances it by making sure Spotify’s podcast service is as complete as it can be. Critically, it does this not by fighting over users, but rather by linking them.

Spotify existing in harmony with creators

This is the part that Web 2.0 got wrong; much like the Facebook model of social networking emphasized being your whole self, Web 2.0 assumed that your one identity would connect together the different pieces of your web existence. However, just as the future of social networking is about different identities for different contexts, interoperability via markets is about linking together distinct user bases in a way that is appropriate for different services, all under the control of the user who is paying for the privilege.

Shopify and Platforms

The other place where market-making might seem familiar is that most classic of tech concepts: platforms. I wrote in 2019’s Shopify and the Power of Platforms, in the context of the e-commerce service’s planned foray into logistics:

Notice, though, that Shopify is not doing everything on their own: there is an entire world of third-party logistics companies (known as “3PLs”) that offer warehousing and shipping services. What Shopify is doing is what platforms do best: act as an interface between two modularized pieces of a value chain.

A drawing of Shopify as an Interface

On one side are all of Shopify’s hundreds of thousands of merchants: interfacing with all of them on an individual basis is not scalable for those 3PL companies; now, though, they only need to interface with Shopify.

The same benefit applies in the opposite direction: merchants don’t have the means to negotiate with multiple 3PLs such that their inventory is optimally placed to offer fast and inexpensive delivery to customers; worse, the small-scale sellers I discussed above often can’t even get an audience with these logistics companies. Now, though, Shopify customers need only interface with Shopify.

What makes the Shopify platform so fascinating is that over time more and more of the e-commerce it enables happens somewhere other than a Shopify website. Shopify, for example, can help you sell on Amazon, and in what will be an increasingly important channel, Facebook Shops. In the latter case Facebook and Shopify are partnering to create a fully-integrated market: Facebook’s userbase and advertising tools on one side, and Shopify’s e-commerce management and seller base on the other. The broader takeaway, though, is that Shopify’s real value proposition is working across markets, not creating an exclusive one.

Market-Making and Aggregators

Market-making is certainly a characteristic of Aggregators; Google, for example, is a one-stop shop for users, advertisers, and content suppliers. What makes Aggregators unique, though, is their infinite scalability, driven by the effectively zero marginal and transactional costs necessary to serve one more user, advertiser, or supplier. This characteristic, by necessity, reduces everything to a commodity. In contrast, the commonality between what Twitter appears to be building, the phenomenon that Spotify is seeking to plug into, and Shopify and e-commerce is the inherent friction of transferring money (usually via Stripe), for something that is not flattened, but differentiated.

Some of these plays are certainly more Aggregator-like — Twitter/Scroll appears likely to (non-exclusively) abstract away publishers from subscribers — but I think the distinction from advertiser-driven Super Aggregators is a notable development, and an exciting one. The web started with no economy, then built a commoditized advertising-only one, and now is increasingly a market for all sorts of goods and services — the more differentiated the better. That doesn’t just mean interoperability, in the purest most fungible form of money, but also opportunity.

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

Spotify’s Surprise

This Article is a follow-up in two ways:

Plus, a special announcement about Exponent.

Anchor Subscriptions

Spotify’s first announcement was teased during its Stream On event last month: paid podcasts via Anchor.

The first implication of this is exactly what you might expect: podcasters who use Spotify’s Anchor service for hosting their podcasts can now create subscriber-only podcasts on Spotify. At first glance this seems similar to Apple’s offering: use their subscription service to offer paid podcasts on their app. However, Anchor is different in three important ways:

On-Web versus In-App

First, subscribing to an Anchor show on Spotify is a clumsier process than it is on Apple; note these images from the announcement:

Spotify's subscription podcast screen

Subscribing via Anchor

Spotify can’t put a “Subscribe” button in its app due to Apple’s App Store rules, which means that Anchor podcasters have to hope a would-be subscriber finds their way to the Anchor website.1 Apple, meanwhile, can put a subscribe button front-and-center:

Apple's subscription screen

This is, needless to say, a rather stark example of Apple leveraging its control of the App Store to give itself an advantage in a new market; the obvious analogy is Apple Music, which has the same advantage relative to Spotify itself. Spotify filed a complaint to the EU two years ago, and the EU is expected to charge Apple with anti-competitive behavior this week.

Rates

Second, Anchor is charging less than Apple is; the announcement states:

For the next two years, this program will come at no cost to the creator, meaning that participating creators receive 100% of their subscriber revenues (excluding payment transaction fees). Starting in 2023, we plan to introduce a competitive 5% fee for access to this tool.

That “excluding payment transaction fees” is a pretty important parenthetical; here is what the take-home amount is for $5/month and $10/month podcasts on both Anchor and Apple Podcasts:

$5/month $10/month
Anchor now $4.55 $9.41
Anchor 2023+ $4.31 $8.91
Apple Year One $3.50 $7.00
Apple Year Two $4.25 $8.50

These aren’t directly comparable:

  • Anchor’s rates are the same for all subscribers; all $5 subscriptions earn $4.55 in 2021 and 2022, and then earn $4.31 in 2023, no matter if the subscriber signed up in 2021 or 2023.
  • Apple’s “Year One” and “Year Two” rates apply to subscribers, not podcasts; even if your podcast has been available for two years, for example, new subscribers pay out at the 70/30 rate for the first year they are subscribed.

That caveat is an important one, because after 2023 rates for long-time subscribers are more comparable than you might expect.

The Open Podcast Ecosystem

This was the first big surprise in the announcement; from Anchor’s blog post (emphasis mine):

The most seamless discovery, subscription, and listening experience for your audience equals increased earning potential for you. It starts with a straightforward subscription process for listeners that immediately gets access to paid episodes within your existing show feed in Spotify, meaning less friction and more supporters; your audience won’t need to contend with RSS feeds or downloading a separate third-party app to listen. They’ll still have the option of listening on the platform of their choosing through a private RSS feed, so you don’t lose out on any potential subscribers. And with paid subscription content clearly marked on your show and episode pages in Spotify, listeners can easily see how to support you directly, thus presenting a bigger potential paid audience.

Yes, most of this paragraph is about the Spotify experience, but that sentence is a huge deal to the open podcast ecosystem: all Anchor subscriptions will include per-subscriber private RSS feeds so that you can listen to the podcast you paid for in the app of your choosing — it’s not locked to Spotify. That’s exactly how the Stratechery podcast and Dithering work; I explained last year:

HTTP and SMTP, though, are not the only open protocols available to publishers: RSS is another, and it is the foundation of the podcast ecosystem. Most don’t understand that podcasts are not hosted by Apple, but rather that iTunes is a directory of RSS feeds hosted on servers all over the Internet. When you add a podcast to your podcast player, you are simply adding an RSS feed that includes information about the show, and a link for where to download new episodes.

iTunes is only a directory

This, if you squint, looks a lot like email: create something that listeners find valuable on an ongoing basis, and deliver it into a feed they already check, i.e. their existing podcast player. That is Dithering: while you have to pay to get a feed customized to you, that feed can be put in your favorite podcast app, which means Dithering fits in with the existing open ecosystem, instead of trying to supplant it.

The implications of this are fascinating: you can not only listen to an Anchor podcast in Spotify, or an independent podcast player like Overcast, you can even listen to it in Apple Podcasts; I posted this image last week:

Adding a podcast by URL

I expressed concern on Dithering about whether or not Apple would shut off the ability to add arbitrary RSS feeds in order to force creators to use their subscription offering; I hope not, as it would be shutting off Apple Podcasts from the open web even as Spotify is embracing it.

Whose Customers?

As I previously explained when writing about Spotify, I have multiple perspectives on podcasts: the first is my role as analyst, the second is my role as publisher, and the third is my role as a podcaster.

When it comes to Apple, Analyst Ben thinks that Apple’s Podcast Subscription service makes a lot of sense, and is a good example of how Apple can compete on the user experience; Podcaster Ben, meanwhile, would prefer to have my shows everywhere. Publisher Ben, though, has one big hangup when it comes to Apple’s offering:

As I noted above, I’m actually very open to allowing Apple to be my payment processor; in my experience, though, a critical part of the creator business model is having a direct connection with your customers. That is something Apple simply doesn’t allow. From the Podcasters Program Agreement:

Personal Data. In connection with any Podcaster Content hosted by Apple and made available in Apple Podcasts under this Agreement, You represent and warrant that You and Your personnel, agents, and contractors will not access or otherwise process any information that can be used to uniquely identify or contact an individual (“Personal Data”).

This makes it crystal clear that every subscriber that signs up is Apple‘s customer, not mine, and while the revenue may be nice in the short run, it is fundamentally constraining in the long run. I believe that creators will increasingly monetize across apps and experiences; Apple, though, won’t even let me email folks to let them know about what is happening beyond the podcast.

Anchor says it plans to be better in this regard, promising to add “email subscriber” functionality specifically; that’s a good start, but it’s important to note that when it comes to money the subscribers are ultimately Anchor’s, not the publishers. Publisher Ben isn’t too happy about that.

The Spotify Open Access Platform

This was the second big surprise, and full disclosure, it affects me personally, and in a very positive way. From the announcement:

Are you a creator or publisher who has subscribers elsewhere? We’re also working on technology that will let your listeners hear your content on Spotify using your existing login system. This gives creators with existing subscriber bases the option to deliver paid content to their existing paid audiences using Spotify, retaining direct control over the relationship.

I am in fact a creator or publisher who has subscribers elsewhere! That’s my entire frustration with Apple and Anchor’s offerings, and why Publisher Ben was so opposed to Spotify’s moves in podcasting, even as Analyst Ben thought they were a great idea; from a Daily Update last year:

Analyst Ben says it is a good idea for Spotify to try and be the Facebook of podcasting…Podcaster Ben certainly sees the allure: having my podcast available to Spotify’s 271 million monthly active users would be great.

Publisher Ben, though, remembers that my business model is predicated on a higher average revenue per user (thanks to subscriptions), not a higher number of users; that means making tradeoffs, and foregoing wide reach is one of them. That, by extension, means not agreeing to Spotify’s terms for Exponent, and accepting that leveraging RSS to have per-subscriber feeds makes having the Daily Update Podcast on Spotify literally impossible. More broadly, owning my own destiny as a publisher means avoiding Aggregators and connecting directly with customers.

For full disclosure, I have been briefed on the Open Access Platform, and Spotify has addressed all of my concerns; no, they won’t support arbitrary RSS feeds, but instead another open technology — OAuth. Some time soon Stratechery and Dithering subscribers will be able to link their subscriptions to their Spotify accounts, and Spotify isn’t going to charge a dime — they will be my customers from email address to credit card. Spotify Chief R&D Officer Gustav Söderström told me, “Having all of audio on Spotify means meeting independent creators on their terms, not ours.”

Needless to say, Publisher Ben is very happy about this news, and Podcaster Ben is excited to have his work available everywhere. Analyst Ben, though, is pleased as well: Spotify isn’t simply ensuring it has all of audio on its app, it is also sending a powerful signal to creators of all types that their corporate incentives are aligned with what matters to creators.

Spotify’s Facebook Play

While most of Spotify’s announcement was devoted to their new subscriptions offerings, the company also announced that the Spotify Audience Network was now available to podcasts hosted on Megaphone (which Spotify acquired last year) and Anchor (on May 1st). This is the Facebook play I was referring to: Spotify wants to accumulate the most podcast listeners and construct a self-serve ad market place that delivers personalized ads at scale.

Again, Analyst Ben thinks this is smart, but shouldn’t Publisher Ben still be a bit nervous about an Aggregator in my space? Not at all — in fact, Twitter and Facebook are great for Stratechery; if your business is based on word-of-mouth, then giving your readers a voice is nothing but upside. And, while I have never advertised Stratechery, Facebook and Twitter would be the obvious — and most accessible — choices if I did (and don’t underrate LinkedIn). I would never use a Twitter or Facebook subscription product — see the part above about owning my users — but that’s ok, because the web is an open alternative.

And, now that Spotify has fixed the openness problem, I see upside in their approach; it will actually be easier to have a mix of free and paid feeds than it is with custom private RSS feeds, which means a new customer acquisition channel, while the Spotify Audience Network might be the first podcast advertising product that is easily accessible for smaller podcasts. Facebook and Twitter would do well to reconsider their subscription plans to accommodate independent creators like Spotify has, instead of trying to capture them (and Apple too, but I’m not holding out hope).

An Exponent Announcement

I apologize if the Analyst/Publisher/Podcast Ben nomenclature was a bit over-the-top; the reality is that writing about Spotify has always been a bit more fraught than most other companies given that it affects my business, and I wanted to be open about that. I have tried, though, to practice what I preach: that is why James Allworth and I made the decision to remove the Exponent podcast we co-host from Spotify. Yes, Exponent is free, but every listener on Spotify was one I couldn’t convert to a customer as an independent creator, which meant saying no to the would-be Aggregator’s audience.

It’s important to point out that the foundation of the Spotify Audience Network is Streaming Ad Insertion on Spotify itself,2 which, as the name suggests, depends on Spotify’s streaming infrastructure being far more capable and flexible than downloadable MP3s served over RSS. That’s why I wasn’t necessarily mad at Spotify for not supporting private RSS feeds: they made their decision for business reasons, and I made mine.

That, though, is why advocates of the open podcast ecosystem should be pleased with Spotify’s adoption of OAuth: there are good reasons that the company will never support private RSS feeds, but when there is a will there is a way, and Spotify has shown the will to support openness and independent creators. To that end, Exponent has an announcement:

Here’s hoping other Aggregators follow Spotify’s lead.

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

  1. Note the link in show notes: show notes are a part of the podcast, not the Spotify app, which means that it should be excluded from Apple’s ban on linking to the web for digital content subscriptions. After all, the link was placed by the podcaster over the web, not Spotify via its app. []
  2. Ads served on Megaphone are inserted on download []

Podcast Subscriptions vs. the App Store

There was a bit of a brouhaha a couple of weeks ago when Senators Amy Klobuchar and Mike Lee sent a letter to Apple accusing the company of failing to provide a witness for the App Store-focused antitrust committee hearing that is happening later today; Apple responded that it was all a misunderstanding due to a scheduling conflict. From Bloomberg:

“We have deep respect for your role and process on these matters and, as we told your staff, we are willing to participate in a hearing in the subcommittee,” Apple said. “We simply sought alternative dates in light of upcoming matters that have been scheduled for some time and that touch on similar issues.”

It seems likely that Apple was referring to Epic’s lawsuit against Apple and its App Store policies, which goes to trial on May 3rd; the Senators’ letter said as much. The fact this hearing is the day after an Apple event, though, is notable in its own right, given how Apple itself just highlighted where the App Store goes wrong.

iTunes 4.9

Apple Podcasts received 75 seconds of attention in Apple’s one hour and one minute presentation; it seems appropriate that 20 of those seconds were spent recounting Apple’s role in popularizing the format:

The 2005 release of iTunes 4.9 and the iTunes podcast directory was indeed a critical step in popularizing podcasts. I explained in 2017’s Podcasts, Analytics, and Centralization:

Centralization occurs in industry after industry for a reason: everyone benefits, at least in the short term. Start with the users: before iTunes 4.9 subscribing and listening to a podcast was a multi-step process, and most of those steps were so obscure as to be effective barriers for all but the most committed of listeners.

  • Find a podcast
  • Get a podcatcher
  • Copy the URL of the podcast feed into the podcatcher
  • Copy over the audio file from the podcatcher into iTunes
  • Sync the audio file to an iPod
  • Listen to the podcast
  • Delete the podcast from the iPod the next time you synced

iTunes 4.9 made this far simpler:

  • Find a podcast in the iTunes Store and click ‘Subscribe’
  • Sync your iPod
  • Listen

Recounting this simplification may seem pedantic, but there is a point: this was the most important improvement for podcast creators as well. Yes, the iTunes Music Store offered an important new discovery mechanism, but it was the dramatic improvement to the user experience that, for the vast majority of would-be listeners, made podcasts even worth discovering in the first place. Centralized platforms win because they make things easier for the user; producers willingly follow.

And then Apple stopped. Yes, the iPhone happened, and podcast management and listening was further centralized into a single app, but given that Apple’s goal was only ever to sell more iPods (and then more iPhones) the company never pursued centralization to its logical conclusion:

Remember, the web was thought to be a wasteland for advertising until Google provided a centralized point that aggregated users and could be sold to advertisers. Similarly, mobile was thought to monetize even worse than the (desktop) web until Facebook provided a centralized point that aggregated users and could be sold to advertisers. I expect a similar dynamic in podcasts: the industry will remain the province of ads for web hosting and underwear absent centralization and aggregation, and the only entity that can accomplish that is Apple.

In fact, it was Spotify that identified the vacuum that Apple had created, aggressively expanding its podcasting business in an attempt to displace Apple’s Aggregator position; eMarketer predicts the streaming service will surpass Apple later this year in podcast listeners. Spotify is pursuing a multi-pronged strategy ranging from exclusive content to open podcast hosting to targeted advertising, and the company’s actions not only promise to dramatically increase podcast monetization but have also stirred Apple to action.

Podcast Subscriptions

The company’s initial response came in the remaining 55 seconds:

Paid podcasts are not a new concept; Stratechery launched a paid version of the Daily Update last February, and Dithering (where we covered Apple’s other product announcements this morning) last May. Both podcasts are predicated on the fundamentally open nature of RSS: every subscriber has a unique feed, which they can add to the podcast player of their choice, including Apple Podcasts (but not Spotify). It’s a little clumsy, but it works:

Apple’s solution looks far easier to use:

The company didn’t actually show the subscription flow in action, but it seems like a safe bet that it will operate similarly to an app subscription flow: hit a button, scan your face, and you’re good to go. Apple’s pricing is the same as apps as well: 30% for the first year of a subscription, and 15% after that.

As a longstanding | critic | of | the App Store, you might expect me to be scandalized by Apple’s podcast subscription offering…and you would be wrong! In fact, Apple’s podcast offering is an excellent example of how the App Store should operate (with one big exception).

What Podcast Subscriptions Gets Right

Apple’s podcast subscription offering gets four big things right, three of which are the complete opposite of the App Store.

A Great Customer Experience with Competitive Creator Economics

This is the part that podcast subscriptions share with the App Store: it really is a great customer experience, from purchase to subscription tracking to cancellations. This accrues to creators as well: increased customer trust means an increased conversion rate.

Second, because Apple controls the entire experience, they can offer things like trials, early access, or the wholesale substitution of ad-supported episodes with ad-free ones. Integration has value!

Third, while 30% is really high, 15% is extremely competitive; a $5/month podcast like Dithering loses 9% of that amount in credit card fees (2.9% + $0.30/charge), plus whatever amount is paid to the subscription management service. Add on the fact that the number one cause of churn is expired or lost credit cards and Apple’s offering — which is far more likely to have an up-to-date credit card attached — is more attractive than it seems.

Unfortunately everything else that I like about Apple’s offering is in stark contrast to the App Store.

Multiple Ways to Subscribe

Some creators may find 30% to be too much (along with the big problem detailed below), but that’s fine: you can still add arbitrary RSS feeds to Apple Podcasts, either via a deep-link like I demonstrated above, or from this screen:

Adding a podcast by URL

This means that creators have a choice, and that Apple has to win on the merits, and again, there is good reason to believe that Apple can do just that. And if they can’t win on the merits, perhaps they will have to lower their price, or increase the attractiveness of their offering. Competition is a good thing!

The App Store, unfortunately, has no alternative. All apps must be installed via the App Store, which Apple says is for security, but is in fact security theater; the primary reason why bad apps don’t mess up your phone is due to the way that iOS is designed. Theoretically the App Store could also protect you from scams, but that appears to be not much of a priority. And why should it be? There is no competition.

Easy Access to Alternative Payment Methods

Podcasts can contain show notes, and show notes can contain links; these links open in a podcast player’s webview (or in Safari). This is great for subscription-based podcasts: you can load a webview to manage your account, or add on a subscription to your members-only feed:

Once again, this is a fair bit clumsier than simply using Apple’s built-in purchase flows (although Apple Pay helps). That’s ok, though: Apple built the iPhone, and it’s reasonable to argue that they can leverage that advantage to have a superior purchasing experience.

Apps, though, can’t load a webview if there is even a hint of a payment option — apps can’t even include words that tell you to visit a website to subscribe. That means that Spotify or Kindle or any number of apps with digital content can do little more than provide users with a login screen, and keep their fingers crossed that users figure out how to sign up on the web on their own. This works for big names, to an extent, but it is much more difficult for smaller players.

There is no technical way that Apple can stop apps from linking to a webpage, of course; this provision is enforced by App Review, which somehow seems far more effective in figuring out how to navigate from a privacy policy on a web page to a purchase page (and subsequently rejecting the app) than it is in rooting out scams. Podcasts are in a much better place because they are based on open standards and the open web.

Availability of Alternative Podcast Players

It is possible that Apple shuts all of these avenues down. The company could end the possibility of adding arbitrary RSS feeds, and it could disable links in show notes. This would, to be clear, be an exceptionally crappy thing to do, but then again:

The good thing about podcasting is that even if Apple locks the Podcast app down, there are plenty of other podcast apps in the App Store, and most of them are free. Make no mistake, it would be bad for my business if I were shut out from Apple’s podcast app, but at least I would have a chance.

That chance doesn’t exist for developers. There are no alternatives to the App Store on the iPhone, and hoping that a customer spends hundreds of dollars and tens of hours switching to Android is completely unrealistic.

Two Big Problems

There does remain two big problems with Apple’s podcast subscription service:

Who Owns the Customer

As I noted above, I’m actually very open to allowing Apple to be my payment processor; in my experience, though, a critical part of the creator business model is having a direct connection with your customers. That is something Apple simply doesn’t allow. From the Podcasters Program Agreement:

Personal Data. In connection with any Podcaster Content hosted by Apple and made available in Apple Podcasts under this Agreement, You represent and warrant that You and Your personnel, agents, and contractors will not access or otherwise process any information that can be used to uniquely identify or contact an individual (“Personal Data”).

This makes it crystal clear that every subscriber that signs up is Apple‘s customer, not mine, and while the revenue may be nice in the short run, it is fundamentally constraining in the long run. I believe that creators will increasingly monetize across apps and experiences; Apple, though, won’t even let me email folks to let them know about what is happening beyond the podcast.

There is an even more problematic angle to this as well: I noted above that Apple might start locking down its podcast app, which might mean that I want to change to a different platform or monetization method. However, the fact I don’t know who my customers are will make that impossible to communicate.

I’m not, in the context of podcasts anyway, saying that what Apple is doing is illegal, and I acknowledge that many customers may prefer this arrangement. As a creator, though, this is a major red flag (developers, meanwhile, also get no contact, but they have no alternatives).

The Anticompetitive Angle

Apple’s podcast offering, as I laid out above, rightfully competes on the merits with alternative ways of paying for subscription podcasts in the Apple Podcast app. Unfortunately there is a meta competition problem, which is that no one else can offer a podcast subscription service like Apple’s.

Spotify is, of course, the other obvious candidate, and the streaming service is currently testing subscription podcasts via Anchor. However, when that product launches Spotify will not be able to upsell customers from within the Spotify app, like Apple is from within the Podcast app. Not because it is technically impossible, but because Apple is leveraging its control of the operating system into control of the App Store into control of apps and now podcast monetization.

Apple’s Flipped Motivations

To go back to yesterday’s presentation, the obvious reason why Podcasts only warranted a minute of Apple’s time is that the company had so many other cool things to announce:

  • AirTags and the anonymous iPhone network they tap into are something that only Apple could create, thanks to their integrated model.
  • The new iMac is gorgeous, and, as Apple was careful to point out, uniquely enabled by their industry-leading chips.
  • The latest iPads are full-blown computers in their own right, and even more capable in ways that creators are still figuring out.

Apple even released a great-looking new iPhone color, and finally fixed (?) the Apple Remote.

Then again, perhaps Apple spent so little time on podcasts for a rather less attractive reason: while iTunes 4.9 was created to make iPods better, the end game of all of these beautiful devices seems ever more focused on locking in services that make Apple richer; that’s a conversation better saved for Congress.

Non-Fungible Taylor Swift

In July 2014 Taylor Swift wrote an editorial in the Wall Street Journal entitled, For Taylor Swift, the Future of Music Is a Love Story:

There are many (many) people who predict the downfall of music sales and the irrelevancy of the album as an economic entity. I am not one of them. In my opinion, the value of an album is, and will continue to be, based on the amount of heart and soul an artist has bled into a body of work, and the financial value that artists (and their labels) place on their music when it goes out into the marketplace.

In recent years, you’ve probably read the articles about major recording artists who have decided to practically give their music away, for this promotion or that exclusive deal. My hope for the future, not just in the music industry, but in every young girl I meet is that they all realize their worth and ask for it.

The Internet commentariat was unimpressed; Nilay Patel wrote in Vox:

Taylor makes a nice little argument in favor of paying for music. “Music is art,” she says, “and art is important and rare. Important, rare things are valuable. Valuable things should be paid for. It’s my opinion that music should not be free, and my prediction is that individual artists and their labels will someday decide what an album’s price point is.”

This is an impressively-constructed syllogism. It is also deeply, deeply wrong…On the internet, there’s no scarcity: there’s an endless amount of everything available to everyone. The laws of supply and demand don’t work terribly well when there’s infinite supply. Swift is right that “important, rare things are valuable,” but she’s failed to understand that the idea of rarity simply doesn’t exist in the digital marketplace.

Three months after that op-ed it looked like Swift’s argument had the upper-hand; 1989 launched as a digital download or on physical media, but was not available on Spotify. It went on to sell 1.2 million copies at launch, the biggest one-week number in over a decade, topped the Billboard 200 for 11 weeks, and has received a ninefold platinum certification from the RIAA.

Eight years on, though, and Swift has fully embraced streaming: Lover, folklore, and evermore were all available on both Spotify and Apple Music at launch. The most interesting album of all, though, is the one Swift released last week.

Fearless (Big Machine’s Version)

The original Fearless, released in 2008, was Swift’s second studio album and her big commercial breakthrough. It was also the second of six albums Swift owed Big Machine Records, the record label that had made Swift its first signing. In 2018, having fulfilled her contract, Swift switched to Republic Records and Universal Music group; six months later, Scott Borchetta, the founder of Big Machine Records, sold the label — and crucially, the masters to Swift’s first six albums — to an investment group headed by Scooter Braun.

The drama that followed (summarized in this Yashar Ali Twitter thread) pulled in a whole host of Swift history, from her initial signing with Borchetta to Braun client Kanye West, who (in)famously stormed the 2009 MTV Video Music Awards to complain that Beyoncé deserved Swift’s award for Best Video by a Female Artist, launching a feud that would wax and wane for over a decade. Swift accused Borchetta of betrayal; Borchetta claimed Swift misrepresented her negotiations with Big Machine. What is most interesting, though, is this tweet from then Big Machine board member Erik Logan, where he posted an open letter, including the following:

Somewhere you have told a story to yourself that you have the right to change history, facts and re-frame any story you want to fix with any narrative you wish. But, as someone who has been by Scott’s side from before you were born, I’m not going to sit on the sidelines and allow you to re-write history and bend the truth to justify your lack of understanding of a business deal. The facts will come out, you will be proven wrong, and people will begin to see that the world you perpetuate, only through your lens, is not reality…

I don’t know what happened between Swift and Big Machine Records; as I noted in Five Lessons From Dave Chappelle, while it is easy to take the side of creators who signed away their rights to record labels or networks when they were unknowns, few remember that those same record labels and networks also own a bunch of rights for creators that never came close to paying off the investment made in their (failed) careers. It doesn’t matter.

Chappelle’s Redemption Song

As Chappelle made abundantly clear, the fact that Comedy Central lost money on other comics didn’t change the fact that they made a whole bunch of money on Chappelle, and that Chappelle’s ongoing lack of control and cut of royalties made him upset; more importantly, because of the Internet, Chappelle could rally his fans to his side:

This was, to use Logan’s words, re-framing the story, with Chappelle’s narrative, through his lens. And it worked! Chappelle posted an update to Instagram a few months later; here is the key section:

A few weeks ago I put a special out. I called it ‘Unforgiven’. I told people what my beef was with Comedy Central. I never talked about it, I demanded that the network pay me. Many of my peers laughed at me because that’s a ridiculous thing to demand. They said, “You signed the contract so what are you even mad about?”

Here’s the thing: I’m very good at minding my own business. The trick to minding your own business is know what is your business. These people that talk about me, these cowards that rejoice, well they don’t understand what greatness looks like.

I never asked Comedy Central for anything. If you remember, I said “I’m going to my real boss”, and I came to you, because I know where my real power lies. I asked you to stop watching the show and thank God almighty for you, you did. You made that show worthless, because without your eyes it’s nothing. And when you stopped watching it, they called me, and I got my name back, and I got my license back, and I got my show back, and they paid me millions of dollars. Thank you very much.

Swift is doing the exact same thing, which is why the story of her breakup with Big Machine and the question of who was right or wrong ultimately doesn’t matter; Swift, like Chappelle, is taking her masters, whether she owns them or not.

Fearless (Taylor’s Version)

Just look at the art for last week’s release:

The art for Fearless (Taylor's Version)

It’s not just Fearless, it’s Fearless (Taylor’s Version); which version do you think that Swift fans will choose to stream (which, after all, is where most of the residual value of Fearless lies)? That’s the part that Logan forgot: when it comes to a world of abundance the power that matters is demand, and demand is driven by fans of Swift, not lawyers for Big Machine or Scooter Braun or anyone else.

It’s easy to see how this plays out going forward: Swift probably doesn’t even have to remake another album; she has demonstrated the willingness and capability to remake her old records, and her fans will do the rest. It will behoove Shamrock Capital, the current owner of Swift’s masters, to buy-out Braun’s share of future upside and make a deal with Swift, because Swift, granted the power to go direct to fans and make her case, can in fact “change history, facts, and re-frame any story [she] want[s] to fit with any narrative [she] wish[es].”

What is notable about Swift’s tactics is that they are the opposite of what she urged in that 2014 op-ed. Instead of treasuring “Fearless”, Swift devalued it; instead of asking for what her masters are worth, Swift is simply taking them. Patel was right that while art may be important, on the Internet it’s not rare; what he missed is how that makes Swift more powerful than ever.

NFTs

Consider NFTs — non-fungible tokens. I explained NFTs in this Daily Update (and this episode of Dithering); an excerpt:1

Smart contracts are code that is stored on a blockchain, which contain the various conditions entailed in the contract; in the case of an NFT, a smart contract would contain the unique token ID of the piece of digital art and the conditions under which it can be transferred (NFTs can represent anything, including physical assets, but I’m going to assume digital art for now for simplicity sake).

There has been a great deal of excitement in creative industries about NFTs restoring the “value of art”; in addition to digital drawings, there have been music albums sold for millions of dollars. You can see the allure: Patel wrote about the end of scarcity, so technology that brings scarcity back seems like a panacea. Perhaps Swift’s 2014 vision was simply ahead of its time?

It was, in fact, but not because NFTs are inherently valuable, at least not in the way so many aspiring artists wish they were. At the end of the day an NFT is simply an entry in a blockchain, and a blockchain is intrinsically worthless. That, to be clear, does not mean blockchains cannot be a store of value; I wrote in 2017 about Bitcoin:

Bitcoin has been around for eight years now, it has captured the imagination, ingenuity, and investment of a massive number of very smart people, and it is increasingly trivial to convert it to the currency of your choice. Can you use Bitcoin to buy something from the shop down the street? Well, no, but you can’t very well use a piece of gold either, and no one argues that the latter isn’t worth whatever price the gold market is willing to bear. Gold can be converted to dollars which can be converted to goods, and Bitcoin is no different. To put it another way, enough people believe that gold is worth something, and that is enough to make it so, and I suspect we are well past that point with Bitcoin.

Bitcoin’s value is rooted not in the Bitcoin blockchain, but rather in the collective belief of millions that it is in fact valuable; NFTs, to the extent they capture and retain value, will require the same sort of collective belief (this is why I find NBA Top Shot particularly interesting: it is rooted in real world copyright). That means the real power is not the record of belief, but rather the ability to inspire belief in the first place.

Artists, Not Art

This explains what Swift got right in 2014:

A friend of mine, who is an actress, told me that when the casting for her recent movie came down to two actresses, the casting director chose the actress with more Twitter followers. I see this becoming a trend in the music industry. For me, this dates back to 2005 when I walked into my first record-label meetings, explaining to them that I had been communicating directly with my fans on this new site called Myspace. In the future, artists will get record deals because they have fans — not the other way around.

This is the inverse of Swift leveraging her fans to acquire her masters: future artists will wield that power from the beginning (like sovereign writers). It’s not that “art is important and rare”, and thus valuable, but rather that the artists themselves are important and rare, and impute value on whatever they wish.

To put it another way, while we used to pay for plastic discs and thought we were paying for songs (or newspapers/writing or cable/TV stars), empowering distribution over creators, today we pay with both money and attention according to the direction of creators, giving them power over everyone. If the creator decides that their NFTs are important, they will have value; if they decide their show is worthless, it will not. And, in the case of Swift, if she decides that albums are valuable they will be, not because they are now scarce, but because only she can declare an album “Taylor’s Version”.

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

  1. This paragraph is here for reference; as Felix Salmon noted on Twitter, the actual implementation of NFTs leaves a lot to be desired []

Sovereign Writers and Substack

There has been a bit of controversy around Substack over the last week; I’m not going to get into the specifics of various accusations made about various individuals, or their responses; however, I do think that there are a few fundamental issues about the Substack model specifically, and the shift to sovereign writers generally, that are being misunderstood.

I’m going to anchor on this piece from Peter Kafka at Recode:

Substack’s main business model is straightforward. It lets newsletter writers sell subscriptions to their work, and it takes 10 percent of any revenue the writers generate (writers also have to fork over another 3 percent to Stripe, the digital payments company)…

The money that Substack and its writers are generating — and how that money is split up and distributed — is of intense interest to media makers and observers, for obvious reasons. But the general thrust isn’t any different from other digital media platforms we’ve seen over the last 15 years or so.

From YouTube to Facebook to Snapchat to TikTok, big tech companies have long been trying to figure out ways they can convince people and publishers to make content for them without having to hire them as full-time content creators. That often involves cash: YouTube set the template in 2007, when it set up a system that let some video makers keep 55 percent of any ad revenue their clips generated…Like Substack, YouTube and the other big consumer tech sites fundamentally think of themselves as platforms: software set up to let users distribute their own content to as many people as possible, with as little friction as possible.

I’m not sure the connection to Facebook and YouTube hold (even with Substack Pro, which I’ll get to in a moment); As Kafka notes, Substack “lets newsletter writers sell subscriptions to their work”; that, by definition, means that Substack is not “figur[ing] out ways they can convince people and publishers to make content for them without having to hire them as full-time content creators”. Just look at my credit card statement, where I happened to find charges for Casey Newton’s Platformer and Bill Bishop’s Sinocism next to each other:

Credit card charges from Platformer and Sinocism (not Substack)

Notice that the names of the merchant, the phone number of the merchant, and the location are different — that’s because they are different merchants. Substack is a tool for Newton and Bishop to run their own business, no different than, say, mine; Kafka writes:

To be clear: You don’t need to work with a company like Substack or Ghost to create and sell your own newsletter. Ben Thompson, the business and technology writer whose successful newsletter served as the inspiration for Substack, built his own infrastructure cobbling together several services; my former colleague Dan Frommer does the same thing for his New Consumer newsletter. And Jessica Lessin, the CEO of the Information, told me on the Recode Media podcast that she’d consider letting writers use the paid newsletter tech her company has built for free.

Here’s what you see on your credit card statement for Stratechery:

Credit card statement from Stratechery

My particular flavor of membership management software is Memberful, but Memberful is not Stratechery’s publisher; I am. Memberful is a tool I happen to use to run my business, but it has no ownership of or responsibility for what I write. Moreover, Memberful — like Substack — doesn’t hold my customer’s billing data; Stripe does, and that charge is from my Stripe account, just as the first two charges are from Newton and Bishop’s respective Stripe accounts.

This is what makes “the intent interest of media makers and observers” so baffling. There is a very easy and obvious answer to “how that money is split up and distributed”: subscriber money goes to the person or publication the subscriber subscribes to. That’s it! Substack is a tool for the sovereign writer; the sovereign writer is not a Substack employee, creator, contractor, nothing. Users quite literally pay writers directly, who pass on 10% to Substack; Substack doesn’t get any other say in “how that money is split up and distributed.”

But what about Substack Pro?

Substack Pro

Back in 2017 I wrote a post called Books and Blogs that explained why subscriptions were a much better model for writers than books:

A book, at least a successful one, has a great business model: spend a lot of time and effort writing, editing, and revising it up front, and then make money selling as many identical copies as you can. The more you sell the more you profit, because the work has already been done. Of course if you are successful, the pressure is immense to write another; the payoff, though, is usually greater as well: it is much easier to sell to customers you have already sold to before than it is to find customers for the very first time…

Since then it has been an incredible journey, especially intellectually: instead of writing with a final goal in mind — a manuscript that can be printed at scale — Stratechery has become in many respects a journal of my own attempts to understand technology specifically and the way in which it is changing every aspect of society broadly. And, it turns, out, the business model is even better: instead of taking on the risk of writing a book with the hope of one-time payment from customers at the end, Stratechery subscribers fund that intellectual exploration directly and on an ongoing basis; all they ask is that I send them my journals of said exploration every day in email form.

Recurring revenue is much better than selling a book once; however, just as you have to spend time to write a book before you can sell it, you need time to build up a subscriber base that supports a full-time subscription. I accomplished this by writing Stratechery on nights and weekends while working at Microsoft and Automattic, and then, when I started charging, basically jumping off of the deep end, but working writers can’t always do the former (I would bet that publications start being stricter about this going forward). This is where Substack Pro comes in; from Kafka:

But in some cases, Substack has also shelled out one-off payments to help convince some writers to become Substack writers, and in some cases those deals are significant. Yglesias says that when it lured him to the platform last fall, Substack agreed to pay him $250,000 along with 15 percent of any subscription revenue he generates; after a year, Yglesias’s take will increase to 90 percent of his revenue, but he won’t get any additional payouts from Substack.

As Yglesias told me via Slack (he stopped working as a Vox writer last fall but still contributes to Vox’s Weeds podcast), the deal he took from Substack is actually costing him money, for now. Yglesias says he has around 9,800 paying subscribers, which might generate around $860,000 a year. Had he not taken the Substack payment, he would keep 90 percent of that, or $775,000, but under the current deal, where he’ll keep the $250,000 plus 15 percent of the gross subscription revenue, his take will be closer to $380,000.

Substack has been experimenting with this kind of offer for some time, but last week, it began officially describing them as “Substack Pro” deals.

In short, the best analogy to Substack Pro are book advances, which are definitely something that publishers do. In that case publishers give an author a negotiated amount of money in advance of writing a book for reasons that can vary; in the case of famous people the advance represents the outcome of a bidding war for what is likely to be a hit, while for a new or unknown author an advance provides for the author’s livelihood while they actually write the book. The publisher then keeps all of the proceeds of the book until the advance is paid back, and splits additional proceeds with the author, usually in an 85/15 split (sound familiar?); of course we don’t know the exact details of book deals, because they are not disclosed.1

At the same time, Substack Pro isn’t like a book advance at all in a way that is much more advantageous to the writer. Book publishers own the publishing rights and control the royalties as long as it is in print; writers in the Substack Pro program still own their customers and all of the revenue past the first year, of which they can give 10% to Substack for continued use of their tool, or switch to another tool. This is where the comparison to YouTube et al falls flat: YouTube wants to be permanently in the middle of the creator-viewer relationship, while Substack remains to the side; from this perspective Substack Pro is more akin to an unsecured bank loan — success or failure is still determined by the readers.

The Real Scam

Now granted, there may be some number of Substack Pro participants who end up earning less than their advance, particularly if Substack sees Substack Pro as more of a marketing tool to shape who uses Substack; if Substack actually runs Substack Pro like a business, though, I would expect lots of deals like the Yglesias one, which has turned out to be quite profitable for Substack. As Yglesias himself noted:

Substack Pro made it possible for Yglesias to launch Slow Boring without worrying about paying the bills, and is making a profit as a reward for bearing the risk of Yglesias not succeeding or succeeding more slowly than he needed. As for Yglesias, he may end up missing out on several hundred thousand dollars this year, but given he’s not selling a book but rather a subscription he can look forward to a huge increase in revenue next year.

This, needless to say, is not a scam, which is what Annalee Newitz argued:

For all we know, every single one of Substack’s top newsletters is supported by money from Substack. Until Substack reveals who exactly is on its payroll, its promises that anyone can make money on a newsletter are tainted. We don’t have enough data to judge whether to invest our creative energies in Substack because the company is putting its thumb on the scale by, in Hamish’s own words, giving a secret group of “financially constrained writers the ability to start building a sustainable enterprise.” We are, not to put too fine a point on it, being scammed.

It is, for the reasons I laid out above, easier to get started with a subscription business if you have an advance. No question. But this take is otherwise completely nonsensical: Substack’s top newsletters are at the top because they have the most subscribers paying the authors directly. For example, look at the “Politics” leaderboard, where Yglesias is seventh:

Substack's 'Politics' leaderboard

We already know that “Thousands of Subscribers” to Slow Boring is 9,800; given that 9,800 * $8/month = $78,400/month, we can surmise that The Weekly Dish has at least 15,680 subscribers ($78,400/month ÷ $5/month). Those are real people paying real dollars of their own volition, not because Substack is somehow magically making them do it.

Frankly, it’s hard to shake the sense that Newitz and other Substack critics simply find it hard to believe that there is actually a market for independent writers, which I can understand: I had lots of folks tell me Stratechery was a stupid idea that would never work, but the beautiful thing about being my own boss is that they don’t determine my success; my subscribers do, just as they do every author on Substack.

Still, that doesn’t change the fact there is a real unfair deal in publishing, and it has nothing to do with Substack. Go back to Yglesias: while I don’t know what he was paid by Vox (it turns out that Substack, thanks to their leaderboards, is actually far more transparent about writers’ income than nearly anywhere else), I’m guessing it was a lot less than the $380,000 he is on pace for, much less the $775,000 he would earn had he forgone an advance.2 It was Vox, in other words, that was taking advantage of Yglesias.

This overstates things, to be sure; while Yglesias built his following on his own, starting with his own blog in 2002, Vox, which Yglesias co-founded, was a team effort, including capital from Vox Media. Still, if we accept the fact that Yglesias charging readers directly is the best measurement of the value those readers ascribe to his writing, then by definition he was severely under-compensated by Vox. The same story applies to Andrew Sullivan, the author of the aforementioned The Weekly Dish; Ben Smith reported:

But Mr. Sullivan is, as his friend Johann Hari once wrote, “happiest at war with his own side,” and in the Trump era, he increasingly used the weekly column he began writing in New York magazine in 2016 to dial up criticism of the American left. When the magazine politely showed him the door last month, Mr. Sullivan left legacy media entirely and began charging his fans directly to read his column through the newsletter platform Substack, increasingly a home for star writers who strike out on their own. He was not, he emphasizes, “canceled.” In fact, he said, his income has risen from less than $200,000 to around $500,000 as a result of the move.

Make that nearly $1 million, i.e. $800,000 of surplus value that New York Magazine showed the door.

Substack Realities

Of course things aren’t so simple; Sullivan, like several of the other names on that leaderboard, are, to put it gently, controversial. That he along with other lightning-rod writers ended up on Substack is more a matter of where else would they go? Again, the entire point is that Sullivan’s readers are paying Sullivan, which means Substack was an attractive option precisely because they don’t decide who gets paid what — or if they get paid at all.

Just because Sullivan was forced to be a sovereign writer, though, doesn’t change the fact that writers who can command a paying audience have heretofore been significantly underpaid. That points to the real reason why the media has reason to fear Substack: it’s not that Substack will compete with existing publications for their best writers, but rather that Substack makes it easy for the best writers to discover their actual market value.

This is where Substack really is comparable to Facebook and the other big tech companies; the media’s revenue problems are a function of the Internet unbundling editorial and advertising. The fact that Google and Facebook now make a lot of money from advertising is unrelated. Similarly, media’s impending cost problem — as in they will no longer be able to afford writers that can command a paying audience — is a function of the Internet making it possible to go direct; that Substack is one of many tools competing to make this easier will be similarly unrelated.

This explains three other Substack realities:

  • First, Substack is going to have a serious problem retaining its most profitable writers unless it substantially reduces its 10% take.
  • Second, Substack is less threatened by Twitter and Facebook than many think; the problem with the social networks is that they want to own the reader, but the entire point of the sovereign writer is that they own their audience. Substack’s real threat will be lower-priced competitors.
  • Third, it would be suicidal for Substack to kick any successful writers off of its platform for anything other than gross violations of the law or its terms of service. That would be a signal for every successful writer to seek out a platform that is not just cheaper, but also freer (i.e. open-source).

This is also why Substack Pro is a good idea. To be honest, I was a tad bit generous above: signing up someone like Yglesias is closer to the “popular author bidding war” side of the spectrum, and may not be worth the trouble; what would be truly valuable is helping the next great writer build a business, perhaps in exchange for more lock-in or the rights to bundle their work. Ideally these writers would be the sort of folks who would have never gotten a shot in traditional media, because they don’t fit the profile of a typical person in media, and/or want to cover a niche that no one has ever thought to cover (these are almost certainly related).

The Sovereign Writer

I am by no means an impartial observer here; obviously I believe in the viability of the sovereign writer. I would also like to believe that Stratechery is an example of how this model can make for a better world: I went the independent publishing route because I had no other choice (believe me, I tried).

At the same time, I suspect we have only begun to appreciate how destructive this new reality will be for many media organizations. Sovereign writers, particularly those focused on analysis and opinion, depend on journalists actually reporting the news. This second unbundling, though, will divert more and more revenue to the former at the expense of the latter. Maybe one day Substack, if it succeeds, might be the steward of a Substack Journalism program that offers a way for opinion writers and analysts to support those that undergird their work.3

What is important to understand, though, is that Substack is not in control of this process. The sovereign writer is another product of the Internet, and Substack will succeed to the extent it serves their interests, and be discarded if it does not.

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

  1. Substack reportedly pays 15% of all revenue, not just revenue above-and-beyond the advance. []
  2. There is an anonymous Google Doc with self-reported media salaries; only three individuals make more than $380,000, and none more than $775,000 []
  3. I am still bullish on the Local News Business Model. []

Moderation in Infrastructure

It was Patrick Collison, Stripe’s CEO, who pointed out to me that one of the animating principles of early 20th-century Progressivism was guaranteeing freedom of expression from corporations:

If you go back to 1900, part of the fear of the Progressive Era was that privately owned infrastructure wouldn’t be sufficiently neutral and that this could pose problems for society. These fears led to lots of common carrier and public utilities law covering various industries — railways, the telegraph, etc. In a speech given 99 years ago next week, Bertrand Russell said that various economic impediments to the exercise of free speech were a bigger obstacle than legal penalties.

Russell’s speech, entitled Free Thought and Official Propaganda, acknowledges “the most obvious” point that laws against certain opinions were an obvious imposition on freedom, but says of the power of big companies:

Exactly the same kind of restraints upon freedom of thought are bound to occur in every country where economic organization has been carried to the point of practical monopoly. Therefore the safeguarding of liberty in the world which is growing up is far more difficult than it was in the nineteenth century, when free competition was still a reality. Whoever cares about the freedom of the mind must face this situation fully and frankly, realizing the inapplicability of methods which answered well enough while industrialism was in its infancy.

It’s fascinating to look back on this speech now. On one hand, the sort of beliefs that Russell was standing up for — “dissents from Christianity, or belie[f]s in a relaxation of marriage laws, or object[ion]s to the power of great corporations” — are freely shared online or elsewhere; if anything, those Russell objected to are more likely today to insist on their oppression by the powers that be. What is certainly true is that those powers, at least in terms of social media, feel more centralized than ever.

This power came to the fore in early January 2021, when first Facebook, and then Twitter, suspended/banned the sitting President of the United States from their respective platforms. It was a decision I argued for; from Trump and Twitter:

My highest priority, even beyond respect for democracy, is the inviolability of liberalism, because it is the foundation of said democracy. That includes the right for private individuals and companies to think and act for themselves, particularly when they believe they have a moral responsibility to do so, and the belief that no one else will. Yes, respecting democracy is a reason to not act over policy disagreements, no matter how horrible those policies may be, but preserving democracy is, by definition, even higher on the priority stack.

This is, to be sure, a very American sort of priority stack; political leaders across the world objected to Twitter’s actions, not because they were opposed to moderation, but because it was unaccountable tech executives making the decision instead of government officials:

I do suspect that tech company actions will have international repercussions for years to come, but for the record, there is reason to be concerned from an American perspective as well: you can argue, as I did, that Facebook and Twitter have the right to police their platform, and, given their viral nature, even an obligation. The balance to that power, though, should be the openness of the Internet, which means the infrastructure companies that undergird the Internet have very different responsibilities and obligations.

A Framework for Moderation

I have made the case in A Framework for Moderation that moderation decisions should be based on where you are in the stack; with regards to Facebook and Twitter:

At the top of the stack are the service providers that people publish to directly; this includes Facebook, YouTube, Reddit, and other social networks. These platforms have absolute discretion in their moderation policies, and rightly so. First, because of Section 230, they can moderate anything they want. Second, none of these platforms have a monopoly on online expression; someone who is banned from Facebook can publish on Twitter, or set up their own website. Third, these platforms, particularly those with algorithmic timelines or recommendation engines, have an obligation to moderate more aggressively because they are not simply distributors but also amplifiers.

This is where much of the debate on moderation has centered; it is also not what this Article is about;

It makes sense to think about these positions of the stack very differently: the top of the stack is about broadcasting — reaching as many people as possible — and while you may have the right to say anything you want, there is no right to be heard. Internet service providers, though, are about access — having the opportunity to speak or hear in the first place. In other words, the further down the stack, the more legality should be the sole criteria for moderation; the further up, the more discretion and even responsibility there should be for content:

A drawing of The Position In the Stack Matters for Moderation

Note the implications for Facebook and YouTube in particular: their moderation decisions should not be viewed in the context of free speech, but rather as discretionary decisions made by managers seeking to attract the broadest customer base; the appropriate regulatory response, if one is appropriate, should be to push for more competition so that those dissatisfied with Facebook or Google’s moderation policies can go elsewhere.

The problem is that I skipped the part between broadcasting and access; today’s Article is about the big piece in the middle: infrastructure. As that simple illustration suggests, there is more room for action than the access layer, but more reticence is in order relative to broadcast platforms. To figure out how infrastructure companies should think about moderation, I talked to four CEOs at various layers of infrastructure:

  • Patrick Collison, the CEO of Stripe, which provides payment services to both individual companies and to platforms. link to interview
  • Brad Smith, the President of Microsoft, which owns and operates Azure, on which a host of websites, apps, and services are run. link to interview
  • Thomas Kurian, the CEO of Google Cloud, on which a host of websites, apps, and services are run. link to interview
  • Matthew Prince, the CEO of Cloudflare, which offers networking services, including free self-serve DDoS protection, without which many websites, apps, and services, particularly those not on the big public clouds, could not effectively operate. link to interview

What I found compelling about these interviews was the commonality in responses; to that end, instead of my making pronouncements on how infrastructure companies should think about issues of moderation, I thought it might be helpful to let these executives make their own case.

The Line in the Stack

The first overarching principle was very much in line with the argument I laid out above: infrastructure is very different from user-facing applications, and should be approached differently.

Microsoft’s Brad Smith:

The industry is looking at the stack and almost putting it in two layers. At the top half of the stack are services that basically tend to meet three criteria or close to it. One is it is a service that has control over the postings or removal of individual pieces of content. The second is that the service is publicly facing, and the third is that the content itself has a greater proclivity to go viral. Especially when all three or even when say two of those three criteria are met, I think that there is an expectation, especially in certain content categories like violent extremism or child exploitative images and the like that the service itself has a responsibility to be reviewing individual postings in some appropriate way.

Cloudflare’s Matthew Prince:

I think that so long as each of the different layers above you are complying with law and doing the right thing and cooperating with law enforcement…it should be an extremely high bar for somebody that sits as low in the stack as we do…I think that that’s different than if you’re Facebook firing an individual user, or even if you’re a hosting provider firing an individual platform. The different layers of the stack, I think do have different levels of responsibility and that’s not to say we don’t have any responsibility, it’s just that we have to be very thoughtful and careful about it, I think more so than what you have to do as you move up further in the stack.

Stripe’s Patrick Collison:

We’re different from others. We’re financial services infrastructure, not a content platform. I’m not sure that the kind of neutrality that companies like Stripe should uphold is necessarily best for Twitter, YouTube, Facebook, etc. However, I do think that in some of the collective reckoning with the effects of social media, the debate sometimes underrates the importance of neutrality at the infrastructure level.

Your Layer = Your Responsibility

The idea of trusting and empowering platforms built on infrastructure to take care of their layer of the stack was a universal one. Collison made the case that to try and police the entire stack was unworkable in a free society (and that this explained why Stripe kicked the Trump campaign off after January 6th, but still supported the campaign indirectly):

This gets into platform governance, which is one of the most important dimensions of all of this, I think. We suspended the campaign accounts that directly used Stripe — the accounts where we’re the top-of-the-stack infrastructure. We didn’t suspend all fundraising conducted by other platforms that benefitted his campaign. We expect platforms that are built on Stripe to implement their own moderation and governance policies and we think that they should have the latitude to do so. This idea of paying attention to your position in the stack is obviously something you’ve written about before and I think things just have to work this way. Otherwise, we’re ultimately all at the mercy of the content policies of our DNS providers, or something like that.

Google’s Thomas Kurian noted there were practical considerations as well:

Imagine somebody wrote a multi-tenant software as a service application on top of GCP, and they’re offering it, and one of the tenants in that application is doing something that violates the ToS but others are not. It wouldn’t be appropriate or even legally possible for us to shut off the entire SaaS application because you can’t just say I’m going to throttle the IP addresses or the inbound traffic, because there’s no way that the tenant is below that level of granularity. In some cases it’s not even technically feasible, and so rather than do that, our model is to tell the customer, who we have a direct relationship with, “Hey, you signed an agreement that agreed to comply with our ToS and AUP [Acceptable Use Policy], but now we have a report of a potential violation of that and it’s your responsibility to pass those obligations on to your end customer.”

Still, Smith didn’t completely absolve infrastructure companies of responsibility:

The platform underneath doesn’t tend to meet those three criteria, but that doesn’t absolve the platform of all responsibility to be thinking about these issues. The test for, at the platform level, is therefore whether the service as a whole has a reasonable infrastructure and is making reasonable efforts to fulfill its responsibilities with respect to individualized content issues. So whether you’re a GCP or AWS or Azure or some other service, what people increasingly expect you to do is not make one-off decisions on one-off postings, but asking whether the service has a level of content moderation in place to be responsible. And if you’re a gab.ai and you say to Azure, “We don’t and we won’t,” then as Azure, we would say, “Well look, then we are not really comfortable as being the hosting service for you.”

Proactive Process

This middle way — give responsibility to the companies and services on top of your infrastructure, but if they fail, have a predictable process in place to move them off of the platform — requires a proactive approach. Smith again:

Typically what we try to do is identify these issues or issues early on. If we don’t think there’s a natural match, if we’re not comfortable with somebody, it makes more sense to let them know before they get on our service so that they can know that and they can find their own means of production if that’s what they want. If we conclude that somebody is reliant on us for their means of production, and we’re no longer comfortable with them, we should try to manage it through a conversation so they can find a means of production that is an alternative if that’s what they choose. But ideally, you don’t want to call them up at noon and tell them they have two hours before they’re no longer on the internet.

Kurian made the same argument against arbitrary decision-making and in favor of proactive process:

We evolve our Acceptable Use Policies on a periodic basis. Remember, we need to evolve it in a thoughtful way, not react to individual crisis. Secondly, we also need to evolve it in a way with a certain frequency, otherwise customers stop trusting the platform. They’d be like “Today, I thought I was accepted and tomorrow you changed it, and now I’m no longer accepted and I have to migrate off the platform”. So we have a fairly well thought out cadence, typically once every six months to once every twelve months, when we reevaluate that based on what we see…

We try to be as prescriptive as possible so that people have as much clarity as possible with what can they do and what they can’t do. Secondly, when we run into something that is a new circumstance, because the boundary of these things continue to move, if it’s a violation of what is considered a legally acceptable standard, we will take action much more quickly. If it’s not a direct violation of law but more debatable on should you take action or not, as an infrastructure provider, our default is don’t take action, but we will then work through a process to update our AUP if we think it’s a violation, and then we make that available through that.

What About Parler?

This makes sense as I write this on March 16, 2021, while the world is relatively calm; the challenge is holding to a commitment to this default when the heat is on, like it was in January. Prince noted:

We are a company that operates, we have equipment running in over 100 countries around the world, we have to comply with the laws in all those countries, but more than that, we have to comply with the norms in all of those countries. What I’ve tried to figure out is, what’s the touchstone that gets us away from freedom of expression and gets us to something which is more universal and more fundamental around the world and I keep coming back to what in the US we call due process, but around the rest of the world, they’d call it, rule of law. I think the interesting thing about the events of January were you had all these tech companies start controlling who is online and it was actually Europe that came out and said, “Whoa, Whoa, Whoa. That makes us super uncomfortable.” I’m kind of with Europe.

Collison made a similar argument about AWS’s decision to stop serving Parler:

Parler was a good case study. We didn’t revoke Parler’s access to Stripe. They’re a platform themselves and it certainly wasn’t clear to us in the moment that Parler should be held responsible for the events. (I’m not making a final assessment of their culpability — just saying that it was impossible for anyone to know immediately.) I don’t want to second guess anyone else’s decisions — we’re doing this interview because these questions are hard! — but I think it’s very important that infrastructure players are willing to delegate some degree of moderation authority to others higher in the stack. If you don’t, you get these problematic choke points…These sudden, unpredictable flocking events can create very intense pressure and I think responding in the moment is usually not a recipe for good decision making.

I was unable to speak with anyone from AWS; the company said in a court filing that it had given the social networking service multiple warnings that it needed more stringent moderation from November 2020 on. AWS said in its filing:

This case is not about suppressing speech or stifling viewpoints. It is not about a conspiracy to restrain trade. Instead, this case is about Parler’s demonstrated unwillingness and inability to remove from the servers of Amazon Web Services (“AWS”) content that threatens the public safety, such as by inciting and planning the rape, torture, and assassination of named public officials and private citizens. There is no legal basis in AWS’s customer agreements or otherwise to compel AWS to host content of this nature. AWS notified Parler repeatedly that its content violated the parties’ agreement, requested removal, and reviewed Parler’s plan to address the problem, only to determine that Parler was both unwilling and unable to do so. AWS suspended Parler’s account as a last resort to prevent further access to such content, including plans for violence to disrupt the impending Presidential transition.

This echoes the argument that Prince made in 2019 in the context of Cloudflare’s decision to remove its (free) DDoS service from 8chan. Prince explained:

The ultimate responsibility for any piece of user-generated content that’s placed online is the user that generated it and then maybe the platform, and then maybe the host and then, eventually you get down to the network, which is where we are. What I think changed was, there were certain cases where there were certain platforms that were designed from the very beginning, to thwart all legal responsibility, all types of editorial. And so the individuals were bad people, their platforms themselves were bad people, the hosts were bad people. And when I say bad people, what I mean is, people who just ignore the rule of law…[So] every once in a while, there might be something that is so egregious and literally designed to be lawless, that it might fall on us.

In other words, infrastructure companies should defer to the services above them, but if no one else will act, they might have no choice; still, Prince added:

What was interesting was as we saw all of these other platforms struggle with this and I think Apple, AWS, and Google got a lot of attention, and it was interesting to see that same framework that we had set out, being put out. I’m not sure it’s exactly the same. Was Parler all the way to complete lawlessness, or were they just over their skis in terms of content moderation?

This is where I go back to Smith’s arguments about proactive engagement, and Kurian’s focus on process: if AWS felt Parler was unacceptable, it should have moved sooner, like Microsoft appears to have done with Gab several years ago. The seeming arbitrariness of the decision was directly related to the lack of proactive management on AWS’s part.

U.S. Corporate Responsibility

An argument I made about the actions of tech companies after January 6 is that they are best understood as a part of America’s checks-and-balances; corporations have a wide range of latitude in the U.S. — see the First Amendment, for example, or the ability to kick the President off of their platform — but as Spider-Man says, “With great power comes great responsibility”. To that end U.S. tech companies were doing their civic duty in ensuring an orderly transfer of power, even though it hurt their self-interest. I put this argument to Collison:

I think you’re getting at the idea that companies do have some kind of ultimate responsibility to society, and that that might occasionally lead to quite surprising actions, or even to actions that are inconsistent with their stated policies. I agree. It’s important to preserve the freedom of voluntary groups of private citizens to occasionally act as a check on even legitimate power. If some other company decided that the events of 1/6 simply crossed a subjective threshold and that they were going to withdraw service as a result, well, I think that’s an important right for them to hold, and I think that the aggregate effect of such determinations, prudently and independently applied, will ultimately be a more robust civic society.

Prince was more skeptical

That is a very charitable read of what went on…I think, the great heroic patriotic, “We did what was right for the country”, I mean, I would love that to be true, I’m not sure that if you actually dig into the reality of that, that it was that. As opposed to succumbing to the pressure of external, but more importantly, internal pressures.

Smith took the question in a different direction, arguing that tech company actions in the U.S. are increasingly guided by expectations from other countries:

I do think that there is, as embedded in the American business community, a sense of corporate social responsibility and I think that has grown over the last fifteen years. It’s not unique to the United States because I could argue that in certain areas, European business feels the same way. I would say that there are two factors that add to that sense of American corporate responsibility as it apply to technology and content moderation that are also important. One is, during the Trump administration, there was a heightened expectation by both tech employees and civil society in the United States that the tech sector needed to do more because government was likely to do less. And so I think that added to pressure as well as just a level of activity that grew over the course of those four years, that was also manifest on January 6th…

There is a second factor that I also think is relevant. There’s almost an arc of the world’s democracies that are creating common expectations through new laws outside the United States that I think are influencing then expectations for what tech companies will do inside the United States.

Kurian immediately jumped to the international implications as well:

Cloud is a global utility, so we’re making our technology available to customers in many, many countries, and in each country we have to comply with the sovereign law of that country. So it is a complex question because what is considered acceptable in one country may be considered non-acceptable in another country.

Your example of First Amendment in the United States and the way that other countries may perceive it gets complicated, and that’s one of the questions we’re working through as we speak, which we don’t have a clear answer on, which is what action do you take if it’s in violation of one country’s law when it is a direct contradiction in another country’s law? And for instance, because these are global platforms, if we say you’re not going to be allowed to operate in the United States, the same company could just as well use our regions in other countries to serve, do you see what I mean? There’s no easy answer on these things.

The Global Internet?

If there was universal agreement on the importance of understanding the different levels of the stack, the question of how to answer these questions globally provided the widest range of responses.

Collison argued that a bias towards neutrality was the only way a service could operate globally:

When it comes to moderation decisions and responsibilities as internet infrastructure, that pushes you to an approach of relative neutrality precisely so that you don’t supersede the various governmental and democratic oversight and civil society mechanisms that will (and should) be applied in different countries.

Kurian highlighted the technical challenges in any other approach:

We have tried to get to what’s common, and the reality is it’s super hard on a global basis to design software that behaves differently in different countries. It is super difficult. And at the scale at which we’re operating and the need for privacy, for example, it has to be software and systems that do the monitoring. You cannot assume that the way you’re going to enforce ToS and AUPs is by having humans monitor everything, I mean we have so many customers at such a large scale. And so that’s probably the most difficult thing is saying virtual machines behave one way in Canada, and a different way in the United States, and a third way, I mean that’s super complicated.

Smith made the same argument:

If you’re a global technology business, most of the time, it is far more efficient and legally compliant to operate a global model than to have different practices and standards in different countries, especially when you get to things that are so complicated. It’s very hard to have content moderators make decisions about individual pieces of content under one standard, but to try to do it and say, “Well, okay, we’ve evaluated this piece of content and it can stay up in the US but go down in France.” Then you add these additional layers of complexity that add both cost and the risk of non-compliance which creates reputational risk.

You can understand Google and Microsoft’s desire for consistency: what makes the public cloud so compelling is its immense scale, but you lose many of those benefits if you have to operate differently in every country. Prince, though, thinks that is inevitable:

I think that if you could say, German rules don’t extend beyond Germany and French rules don’t extend beyond France and Chinese rules don’t extend beyond China and that you have some human rights floor that’s in there.

But given the nature of the internet, isn’t that the whole problem? Because, anyone in Germany can go to any website outside of Germany.

That’s the way it used to be, I’m not sure that’s going to be the way it’s going to be in the future. Because, there’s a lot of atoms under all these bits and there’s an ISP somewhere, or there’s a network provider somewhere that’s controlling how that flows and so I think that, that we have to follow the law in all the places that are around the world and then we have to hold governments responsible to the rule of law, which is transparency, consistency, accountability. And so, it’s not okay to just say something disappears from the internet, but it is okay to say due to German law it disappeared from the internet. And if you don’t like it, here’s who you complain to, or here’s who you kick out of office so you do whatever you do. And if we can hold that, we can let every country have their own rules inside of that, I think that’s what keeps us from slipping to the lowest common denominator


Prince’s final conclusion is along the same lines of where I landed in Internet 3.0 and the Beginning of Tech History. To date the structure of the Internet has been governed by technological and especially economic incentives that drove towards centralization and Aggregation; after the events of January, though, political considerations will increasingly drive decision-making.

For many internet service providers this provides an opportunity to abstract away this complexity for other companies; Stripe, to take a pertinent example, adroitly handles different payment methods and tax regimes with a single API. The challenge is more profound for the public clouds, though, which achieve their advantage not by abstracting away complexity, but by leveraging the delivery of universal primitives at scale.

This is why I take Smith’s comments as more of a warning: a commitment to consistency may lead to the lowest common denominator outcome Prince fears, where U.S. social media companies overreach on content, even as competition is squeezed out at the infrastructure level by policies guided by non-U.S. countries. It’s a bad mix, and public clouds in particular would be better off preparing for geographically-distinct policies in the long run, even as they deliver on their commitment to predictability and process in the meantime, with a strong bias towards being hands-off. That will mean some difficult decisions, which is why it’s better to make a commitment to neutrality and due process now.

You can read the interviews from which this Article was drawn here.

The Roblox Microverse

The degree to which Stratechery Weekly Articles are pre-planned varies; Eddy on Twitter, though, had this article pegged:

Tomorrow, after a number of delays, including switching from an IPO to a Direct Listing, is finally the day that Roblox goes public; it’s a company I can’t wait to write about.

The Evolution of Video Games

The article Eddy was replying to was Clubhouse’s Inevitability, particularly this chart:

The most obvious difference between Clubhouse and podcasts is how much dramatically easier it is to both create a conversation and to listen to one. This step change is very much inline with the shift from blogging to Twitter, from website publishing to Instagram, or from YouTube to TikTok.

A drawing of Clubhouse's Similarity to Twitter, Instagram, and TikTok

Secondly, like those successful networks, Clubhouse centralizes creation and consumption into a tight feedback loop. In fact, conversation consumers can, by raising their hand and being recognized by the moderator, become creators in a matter of seconds.

Compare this to how Roblox describes their business in their S-1:

An average of 36.2 million people from around the world come to Roblox every day to connect with friends. Together they play, learn, communicate, explore, and expand their friendships, all in 3D digital worlds that are entirely user-generated, built by our community of nearly 7 million active developers. We call this emerging category “human co-experience,” which we consider to be the new form of social interaction we envisioned back in 2004. Our platform is powered by user-generated content and draws inspiration from gaming, entertainment, social media, and even toys.

Here’s the question, though: with regard to Eddy’s question, how might you fit video games onto this evolution of media framework? You could start with analog games, progress to video games, and then to casual games, as Eddy suggested, but it’s worth noting that video games preceded the web by many years; I think it makes more sense to make traditional video games the base unit. More importantly, that graphic was about creation, not consumption, and in that regard, video games fit quite nicely:

  • Step 0 — Pre-Internet: The primary way to distribute video games was on consoles, which were controlled by the console makers; computer gaming was more open, but still required significant distribution capabilities. This was the newspaper era of video game publishing.

  • Step 1 — Democratization: The Internet made it possible to distribute games directly, meaning that anyone could be a publisher; over time the increase in broadband penetration made casual cloud-gaming (originally via Flash and later on Facebook) much more accessible.

  • Step 2 — Aggregation: Mobile dramatically increased the market for video games by ensuring nearly everyone had a video game device in their pockets. Note that this increased the market in two ways: first, there were more potential players, and two, all potential players had more opportunities to play games. Mobile, though, meant App Stores. This was a boon in that it made it easy to distribute video games in a way that customers were willing to trust and experiment with, and built-in payments unlocked entirely new ways of making money. It also meant that App Stores were the only way to reach customers, and you had to pay 30% for the privilege (these advantages and disadvantages are, of course, the exact same).

  • Step 3 — Transformation: This step is when the medium in question becomes something fundamentally different because of the Internet. I explained in Clubhouse’s Inevitability:

Even with the explosion of content resulting from democratizing publishing, what was actually published was roughly analogous to what might have been published in the pre-Internet world. A blog post was just an article; an Instagram post was just a photo; a YouTube video was just a TV episode; a podcast was just radio show. The final step was transformation: creating something entirely new that was simply not possible previously.

Along those lines, a lot of video games, particularly mobile games, are just mobile versions of what has been available for decades; at the same time, video games have incorporated a lot of things that are only possible on the Internet. Multi-player games have been widespread for over twenty years, and the entire concept of in-app purchases has transformed business models. Both are uniquely enabled by the Internet. Even with that caveat, though, Roblox is something entirely new.

The Metaverse

Back to the Roblox S-1:

Some refer to our category as the metaverse, a term often used to describe the concept of persistent, shared, 3D virtual spaces in a virtual universe. The idea of a metaverse has been written about by futurists and science fiction authors for over 30 years. With the advent of increasingly powerful consumer computing devices, cloud computing, and high bandwidth internet connections, the concept of the metaverse is materializing.

The term “Metaverse” was coined by Neal Stephensen in his seminal 1992 science fiction novel Snow Crash:

Like any place in Reality, the Street is subject to development. Developers can build their own small streets feeding off of the main one. They can build buildings, parks, signs, as well as things that do not exist in Reality, such as vast hovering overhead light shows, special neighborhoods where the rules of three-dimensional spacetime are ignored, and free-combat zones where people can go to hunt and kill each other.

The only difference is that since the Street does not really exist — it’s just a computer-graphics protocol written down on a piece of paper somewhere — none of these things is being physically built. They are, rather, pieces of software, made available to the public over the worldwide fiber-optics network. When Hiro goes into the Metaverse and looks down the Street and sees buildings and electric signs stretching off into the darkness, disappearing over the curve of the globe, he is actually staring at the graphic representations — the user interfaces — of a myriad different pieces of software that have been engineered by major corporations.

The Roblox is but one company — more on this in a moment — but the way in which it describes its platform certainly approaches Stephenson’s vision:

The Roblox Platform has a number of key characteristics:

  • Identity: All users have unique identities in the form of avatars that allow them to express themselves as whoever or whatever they want to be. These avatars are portable across experiences.
  • Friends: Users interact with friends, some of whom they know in the real world and others who they meet on Roblox.
  • Immersive: The experiences on Roblox are 3D and immersive. As we continue to improve the Roblox Platform, these experiences will become increasingly engaging and indistinguishable from the real world.
  • Anywhere: Users, developers and creators on Roblox are from all over the world. Further, the Roblox Client operates on iOS, Android, PC, Mac, and Xbox, and supports VR experiences on PC using Oculus Rift, HTC Vive and Valve Index headsets.
  • Low Friction: It is simple to set up an account on Roblox, and free for users to enjoy experiences on the platform. Users can quickly traverse between and within experiences either on their own or with their friends. It is also easy for developers to build experiences and then publish them to the Roblox Cloud so that they are then accessible to users on the Roblox Client across all platforms.
  • Variety of Content: Roblox is a vast and expanding universe of developer and creator-built content. As of September 30, 2020, there were over 18 million experiences on Roblox, and in the twelve months ended September 30, 2020, over 12 million of these were experienced by our community. There are also millions of creator-built virtual items with which users can personalize their avatars.
  • Economy: Roblox has a vibrant economy built on a currency called Robux. Users who choose to purchase Robux can spend the currency on experiences and on items for their avatar. Developers and creators earn Robux by building engaging experiences and compelling items that users want to purchase. Roblox enables developers and creators to convert Robux back into real-world currency.
  • Safety: Multiple systems are integrated into the Roblox Platform to promote civility and ensure the safety of our users. These systems are designed to enforce real-world laws, and are designed to extend beyond minimum regulatory requirements.

Growth at Roblox has been driven primarily by a significant investment in technology and two mutually reinforcing network effects: content and social.

In short, Roblox isn’t a game at all: it is world in which one of the things you can do is play games, with a persistent identity, persistent set of friends, persistent money, all disconnected from the device that you use to access the world. That is the transformational change.

Think about all the previous evolutions of gaming:

  • Video games required a console, PC, or phone
  • Networked games connected your console or computer or phone to another console or computer or phone
  • In-app purchases transformed the business model of games by leveraging the payment system provided by the console, PC, or phone

It should be pointed out that while consoles and phones have fairly similar models, the open nature of the PC left room for Steam to capture the distribution and payment functionality; still, the device was the center of your gaming experience, and most games were silos. Some games have sought to break these walls down; Fortnite has been particularly aggressive in enabling cross-platform play.

Roblox, though, isn’t simply the same game everywhere, it’s the same persistent world everywhere, from PC to console (Xbox, not PlayStation) to smartphone, in which games happen to exist. It’s a metaverse…kind of.

The Microverse

The problem with invoking the “Metaverse” in the context of Roblox is that the traditional conception was a virtual world that rivaled the real world; anyone could plug into it from anywhere, with full interoperability. Roblox, though, is only Roblox.

That’s actually a benefit: by controlling everything Roblox can bring all of the disparate parts of gaming into one place; instead of one app for social interactions, another app for purchases, and a different app for every different game, everything is all in the same place. This also makes Roblox easier to develop for: by constraining graphics to a consistent toolbox it is very easy to build something new.

Three Roblox games

This creates the conditions for the interlocking feedback loops that characterize transformational products; by reducing the prominence and feature set of games, Roblox made it possible to create something bigger. A microverse.

This actually fits the patterns of other transformational products. The feed, for instance, relies on reducing all types of content, from posts to photos to links, to the same format, such that they can all be incorporated into a greater whole. It’s a reason why I think that Clubhouse being all audio actually gives it an advantage relative to Twitter: that leaves more room for user entrepreneurship, both in the kinds of rooms created and also norms around behavior (Twitter realized the same benefits relative to blogs with its 140-character constraint). Similarly, Roblox games aren’t really competitive with non-Roblox games: they’re “worse” in any sort of traditional sense, because the things that make them “better” are the parts that are enabled by imposing constraints.

Roblox and the App Store

That leads to perhaps the most surprising thing about Roblox:

The Roblox "App Store"

That’s the screen you see when you launch the app, and I have to say, it looks an awful lot like an App Store! That’s a problem because Apple states in its App Store Guidelines that “Creating an interface for displaying third-party apps, extensions, or plug-ins similar to the App Store or as a general-interest collection” is “unacceptable”.

On one hand, perhaps Roblox is fine because these are not 3rd-party App Store apps, unlike, say, the rejected Facebook Gaming app. But then again, Xbox Game Pass wants to launch 3rd-party games that run in the cloud, not on the iPhone at all, and Apple also said no. I guess it is enough that these are, without question, Roblox games. They are sui generis.

Of course Apple (and Google) is still taking its share; Roblox has to pay 30% on every purchase of Robux, its in-game currency, as will every other would-be platform on top of their platform (like Clubhouse or Twitter). I very much hope, though, that Apple will be content with that: the reality is that Roblox apps really are something different, simpler in isolation than any one iOS app, yet something new and innovative as a collective. It’s a perfect example of the opportunity I wrote about in 2020’s The End of the Beginning:

The implication of this view should at this point be obvious, even if it feels a tad bit heretical: there may not be a significant paradigm shift on the horizon, nor the associated generational change that goes with it. And, to the extent there are evolutions, it really does seem like the incumbents have insurmountable advantages: the hyperscalers in the cloud are best placed to handle the torrent of data from the Internet of Things, while new I/O devices like augmented reality, wearables, or voice are natural extensions of the phone.

In other words, today’s cloud and mobile companies — Amazon, Microsoft, Apple, and Google — may very well be the GM, Ford, and Chrysler of the 21st century. The beginning era of technology, where new challengers were started every year, has come to an end; however, that does not mean the impact of technology is somehow diminished: it in fact means the impact is only getting started.

Indeed, this is exactly what we see in consumer startups in particular: few companies are pure “tech” companies seeking to disrupt the dominant cloud and mobile players; rather, they take their presence as an assumption, and seek to transform society in ways that were previously impossible when computing was a destination, not a given. That is exactly what happened with the automobile: its existence stopped being interesting in its own right, while the implications of its existence changed everything.

Roblox is the exact sort of platform that is only possible when you accept the reality that the platforms on which it rests aren’t going anywhere. The responsibility of those foundational platforms is to give room to let these microverses flourish, without legislating or taxing them to death.

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

The Web’s Missing Interoperability

It was 2004, and Tim O’Reilly needed a name for his new conference, and so “Web 2.0” was hatched in a brainstorming session devoted to finding a name. A year later O’Reilly would flesh the concept out with his definitive 2005 post What is Web 2.0, but given the fact so many Web 2.0 applications were about creating platforms that were then made valuable with user-generated content, it feels appropriate that O’Reilly made a name first and added the content to justify it later. And, in the spirit of Web 2.0, I’m not going to quote O’Reilly’s article but rather the Wikipedia entry for a definition:

Web 2.0…refers to websites that emphasize user-generated content, ease of use, participatory culture and interoperability (i.e., compatible with other products, systems, and devices) for end users.

Seventeen years on and there is more user-generated content than ever, in part because it is so easy to generate: you can type an update on Facebook, post a photo on Instagram, make a video on TikTok, or have a conversation on Clubhouse. That, though, points to Web 2.0’s failure: interoperability is nowhere to be found. Twitter, which has awoken from its years-long stupor to launch or announce whole host of new products, provides an excellent lens with which to examine the drivers of this centralization.

Super Follows and the Content-Creation Loop

While the specific details of Super Follows are still hazy, the idea of Twitter users being able to charge followers for special access makes all kinds of sense for Twitter the company. First, the items of value, particularly exclusive tweets or the ability to interact, are by definition exclusive to Twitter; only Twitter has tweets! Second, the best place to find Super Follows will be amongst your regular followers; access to the ideal user acquisition channel is built in. Third, Twitter will be able to make money coming and going: the obvious advertising mechanism for finding new Super Follows would be Twitter’s own ad products; Twitter could even make advertising “free”, collecting its payment from future subscription revenue.

This points to the first factor driving centralization: as I explained two weeks ago in Clubhouse’s Inevitability, the most compelling apps for user-generated content tie creation and consumption into a tight loop, bound together with network effects and presented with a feed that neither creators nor consumers want to leave. This isn’t nefarious, it’s simply good product design.

Revue and Distribution

The logic of Twitter getting into newsletters with its purchase of Revue is not quite as obvious, but still compelling: long-form content is very different from 280 characters, and there is a certain allure to a company like Substack that is completely focused on newsletters. At the same time, the most challenging part of building a subscription business is customer acquisition, and Twitter is an obvious channel to do so.1

This is the second factor driving centralization: in a world where distribution is free the real cost is user acquisition, which means it is often easier to give existing users new functionality than it is for a new service based on the functionality to acquire new users. The canonical example of this dynamic is Stories: while Instagram didn’t kill Snapchat, the audacity with which Facebook copied Stories stopped the latter’s growth, at least for a few years.

Business concerns are obviously a major driver here as well: ad-based services want to keep users on their own platform instead of sending them somewhere else, even if it incurs short-term costs; engagement is the recipe for long-term revenue growth.

Spaces and Social Graphs

I’m less convinced than many that Spaces, Twitter’s Clubhouse competitor, will crush the startup like Periscope, Twitter’s live streaming service, once crushed Meerkat. I think the audio streaming market is much larger, for one, and Clubhouse has much more of a head start. Still, I can understand the argument: when it comes to a social network the most compelling feature is if you know people using it, and Twitter already has an existing social graph, as well as a good idea of what you are already interested in based on whom you follow.

That is also why it is so important that Clubhouse incentivizes its users to import their contacts: the startup is bootstrapping a social network off of your phone’s address book, in stark contrast to Meerkat, which directly imported your Twitter contacts, right up until the date when Twitter cut it off. Twitter had learned its lesson from Instagram, which booted up its social network on top of the Twitter graph; Twitter eventually cut off Instagram in-line image sharing, but by then it was too late. To that end, the most Twitter could do to Clubhouse is stop its links from unfurling; it can’t stop Clubhouse’s notifications or use of contacts.

Censorship and Competition

New features are a welcome respite to the reason that Twitter is usually in the news: controversy and charges/pleas for censorship. Some folks want Twitter to control more speech, some want Twitter to control less, but nearly all are convinced that Twitter is on the side of their enemies. Still, to be fair, Trump supporters have the stronger argument in that regard, given the fact that Twitter suspended the former President’s account (a decision that I ultimately argued for, even though it was very close).

The problem is that the solution proposed by many Republicans — revoking Section 230 — makes absolutely zero sense. Section 230 doesn’t protect the rights of private companies to make their own moderation decisions; the First Amendment does! In fact, removing Section 230 protection from tech platforms would lead them to censor far more, the better to avoid the inevitable flood of pointless lawsuits that would follow.

A better solution is more competition,2 but the reality of social networking is that new services that succeed do so by focusing on a different aspect of human communication. Twitter is primarily text, for example, while Instagram is pictures, and TikTok is video (this is another reason why I am bullish on Clubhouse relative to Twitter: I think being focused on audio is a big advantage). Facebook is the record of your life, while Snapchat is about ephemerality, and messaging apps about groups and logistics. The first time I tried to map the gamut of social media experiences, I opened with that famous Walt Whitman line from Song of Myself:3

Do I contradict myself?
Very well then I contradict myself,
(I am large, I contain multitudes.)

This is why the FTC’s hilarious attempt to define Facebook’s market in its absurd antitrust lawsuit was so misguided; of course Facebook has a monopoly on being Facebook, just as Twitter has a monopoly on being Twitter. It is quite clear, though, that no service has a monopoly on user-generated content or human interaction.

Still, that doesn’t help the “competition as an answer to censorship” problem, because the entire point is that if Twitter stops you from tweeting the ban is absolute; the issue isn’t simply the form of the tweets, but the network undergirding the service. That, though, is why the Instagram and Meerkat stories are so interesting: we already have evidence about just how powerful it is when a service lets its social graph be exported, and how limiting it is when it doesn’t. To that end, it seems clear to me that the only way to build a direct competitor to any of these services, like Twitter, is to have direct access to the Twitter network.

Interoperability and Privacy

To that end, to the extent regulators want to engender competition in the social networking space, whether for political reasons or simply because they think Facebook is too powerful, the solution is clear: force existing social networks to open up their social graphs. Clubhouse shouldn’t have needed to upload your contacts, which aren’t even that great of a resource, filled as they are with your dentist from ten years ago at best and your abusive ex at worst; Facebook and Twitter (and Snapchat and all the rest) should have had to expose their social graphs to Clubhouse just like Twitter once did for Instagram. There is nothing else that would do more to spur competition.

The problem, of course, is privacy; European lobbyist Alexander Hanff described on LinkedIn how Clubhouse violated GDPR (all punctuation and capitalization from the original):

Clubhouse is a REALLY bad idea for private users, companies and investors. As private users they are asking you to break the law by providing access to your address book in order to invite your friends to use the platform with you…Clubhouse is a shining example of HOW TO BREAK EU LAW — they are so good at it they could and probably should, write a book on the subject.

Hanff isn’t wrong! One of the reasons why GDPR is such a disaster is that it makes it all but impossible for a new social media company to ever be started in Europe; I explained in 2017:

Several folks have suggested that the GDPR’s requirements around data portability, including that it be machine accessible (i.e. not just a PDF) will help new networks form, but in fact the opposite is the case. Note this section from the Guidelines on the right to data portability…

This forbids what I proposed: the easy re-creation of one’s social graph on other networks. Moreover, it’s a reasonable regulation: my friend on Facebook didn’t give permission for their information to be given to Snapchat, for example. It does, though, make it that much more difficult to bootstrap a Facebook competitor: the most valuable data (from a business perspective, anyways) is the social graph, not the updates and pictures that must now be portable, which means that again, thanks to (reasonable!) regulation, Facebook’s position is that much more secure.

I get the argument around banning contact exports; unsurprisingly, there are calls that Apple do exactly that in the wake of Clubhouse’s rise (never mind the fact that contacts have been accessible — and thus have been accessed! — in this way for years). What people making these calls — and these laws — need to be more honest about, though, is that they killing competition. If you want to ensure that Twitter wins in audio, or that Facebook wins everywhere else, then elevating privacy over everything else, ignoring both tradeoffs (like killing competition in social networks) and facts on the ground (like the reality that your contacts have long since ceased to be private), is an excellent way to accomplish exactly that.

Look no further than e-commerce.

The Sad Story of Shopify and Facebook

Two years ago I wrote about one of the most exciting examples of competition on the Internet: Shopify and the Power of Platforms. After explaining how Walmart had failed in its futile attempt to challenge Amazon head-on, I highlighted the fact that Shopify was pursuing a very different strategy:

At first glance, Shopify isn’t an Amazon competitor at all: after all, there is nothing to buy on Shopify.com. And yet, there were 218 million people that bought products from Shopify without even knowing the company existed. The difference is that Shopify is a platform: instead of interfacing with customers directly, 820,000 3rd-party merchants sit on top of Shopify and are responsible for acquiring all of those customers on their own.

A drawing of The Shopify Platform

This means they have to stand out not in a search result on Amazon.com, or simply offer the lowest price, but rather earn customers’ attention through differentiated product, social media advertising, etc. Many, to be sure, will fail at this: Shopify does not break out merchant churn specifically, but it is almost certainly extremely high. That, though, is the point.

Unlike Walmart, currently weighing whether to spend additional billions after the billions it has already spent trying to attack Amazon head-on, with a binary outcome of success or failure, Shopify is massively diversified. That is the beauty of being a platform: you succeed (or fail) in the aggregate. To that end, I would argue that for Shopify a high churn rate is just as much a positive signal as it is a negative one: the easier it is to start an e-commerce business on the platform, the more failures there will be. And, at the same time, the greater likelihood there will be of capturing and supporting successes.

This is how Shopify can both in the long run be the biggest competitor to Amazon even as it is a company that Amazon can’t compete with: Amazon is pursuing customers and bringing suppliers and merchants onto its platform on its own terms; Shopify is giving merchants an opportunity to differentiate themselves while bearing no risk if they fail.

The strategy is working: while Shopify’s Gross Merchandise Volume (GMV) was about a quarter the size of Amazon’s in 2020, it was only 18% in 2019, and a mere 15% in 2018;4 in other words, even during the pandemic Shopify grew faster than Amazon, surely welcome news to those concerned about Amazon’s power, and validation to people like me who believe in the power of the Internet to unlock niche businesses that were never before possible.

That is also why this news in the Wall Street Journal kind of bums me out:

Shopify Inc., a commerce platform for businesses, is bringing its checkout and payment processing system, Shop Pay, to some Facebook Inc. platforms. The Shop Pay option will first be available to Instagram users on Tuesday and will roll out on Facebook Shops, the social-media company’s platform for small businesses, in the next few weeks.

Consumers will be able to use Shop Pay to complete purchases, expanding on existing options to use PayPal Holdings Inc.’s PayPal or manually enter credit or debit card information. All these methods are offered via the Facebook Pay payment system. Shop Pay, which stores credit card and shipping information to speed online checkout, hasn’t previously been available outside the e-commerce stores of Shopify clients.

Narrowly speaking, this is a huge win for Shopify; I fretted last year in Platforms in an Aggregator World that Facebook Shops would be bad for Shopify, in large part because it would limit usage of Shop Pay, itself a part of “Merchant Solutions”, the company’s biggest growth driver. I’m very happy to have gotten this wrong!

At the same time, from a big picture perspective this is clearly a case of Shopify, one of the most exciting companies in tech and the seeming leader of The Anti-Amazon Alliance, effectively moving into Facebook’s garden, because the web is increasingly a barren wasteland for small businesses. The cause is Apple: its approach to cookies makes platform-based web storefronts increasingly difficult to monetize effectively (Shop Pay performed magic in this regard), and its attack on “tracking” — which goes far beyond the IDFA — makes it increasingly impossible to acquire users in one place and convert them in another. The solution is to do user acquisition and user conversion all in one app — i.e. on Facebook — which is why Shopify is helping merchants move off the web and onto Facebook.

Again, it’s a good solution for Shopify, and Facebook deserves credit for recognizing that Shopify is a complement to their service, not a competitor, but I find it disappointing that once again elevating privacy above every other tradeoff is entrenching Facebook, the biggest incumbent of all.


The most frustrating aspect of the entire privacy debate is that the most ardent advocates of an absolutist position tend to describe anyone who disagrees with them as a Facebook defender. My motivation, though, is not to defend Facebook; quite the opposite, in fact: I want to see the social networking giant have more competition, not less, and I despair that the outcome of privacy laws like GDPR, or App Store-enforced policies from Apple, will be to damage Facebook on one hand, and destroy all of its long-term competitors on the other.

I worry even more about small businesses uniquely enabled by the Internet; forcing every company to act like a silo undoes the power of platforms to unlock collective competition (a la Shopify versus Amazon), whether that be in terms of advertising, payments, or understanding their users. Regulators that truly wish to limit tech power and unlock the economic potential of the Internet would do well to prioritize competition and interoperability via social graph sharing, alongside a more nuanced view of privacy that reflects reality, not misleading ads; I would settle for at least admitting there are tradeoffs being made.

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

  1. In my experience word-of-mouth is more powerful than Twitter; both benefit from and build on Stratechery’s freemium approach where Articles like this are free-to-read on the web []
  2. More on questions of infrastructure, which apply to the Parler situation, soon []
  3. Be kind; the map in that article is from 2013! []
  4. Shopify reports GMV; Amazon’s are based on estimates []

Clubhouse’s Inevitability

To what extent are new companies, particularly those in new spaces, pushed versus pulled into existence? Last week I wrote about how Tesla is a Meme Company:

It turned out, though, that TSLA was itself a meme, one about a car company, but also sustainability, and most of all, about Elon Musk himself. Issuing more stock was not diluting existing shareholders; it was extending the opportunity to propagate the TSLA meme to that many more people, and while Musk’s haters multiplied, so did his fans. The Internet, after all, is about abundance, not scarcity. The end result is that instead of infrastructure leading to a movement, a movement, via the stock market, funded the building out of infrastructure.

Electrification of personal vehicles would have happened at some point; it seems fair to argue that Musk accelerated the timeline significantly. Clubhouse, meanwhile, Silicon Valley’s hottest consumer startup, feels like the opposite case: in retrospect its emergence feels like it was inevitable — if anything, the question is what took so long for audio to follow the same path as text, images, and video.

Step 1: Democratization

The grandaddy of independent publishing on the Internet was the blog: suddenly anyone could publish their thoughts to the entire world! This was representative of the Internet’s most obvious impact on media of all types: democratization.

  • Distributing text no longer required a printing press, but simply blogging software:
    From print to blogs
  • Distributing images no longer required screen-printing, but simply a website:
    From magazines to Instagram
  • Distributing video no longer required a broadcast license, but simply a server:
    From TV to YouTube
  • Distributing audio no longer required a radio tower, but simply an MP3:
    From radio to podcasts

Businesses soon sprang up to make this process easier: Blogger for blogging, Flickr for photo-sharing, YouTube for video, and iTunes for podcasting (although, in a quirk of history, Apple never actually provided centralized hosting for podcasts, only a directory). Now you didn’t even need to have your own website or any particular expertise: simply pick a username and password and you were a publisher.

Step 2: Aggregation

Making anyone into a publisher resulted in an explosion of content; this shifted value to entities able to help consumers find what they were interested in. In text the big winner was Google, which indexed pre-existing publications, independent blogs, and everything in-between. The big winner in photos, meanwhile, ended up being Instagram: users “came for the tool and stayed for the network”, as Chris Dixon memorably put it:

Instagram’s initial hook was the innovative photo filters. At the time some other apps like Hipstamatic had filters but you had to pay for them. Instagram also made it easy to share your photos on other networks like Facebook and Twitter. But you could also share on Instagram’s network, which of course became the preferred way to use Instagram over time.

The Internet creates a far tighter feedback loop between content creation and consumption than analog media; Instagram leveraged this loop to become the dominant photo network. YouTube accomplished a similar feat, although the relative difficulty in creating video meant that the ratio of viewers to creators was much more extreme than in the case of photo-sharing. That, though, is exactly what made YouTube so dominant: creators knew that that was where all of their would-be viewers were.

Spotify is trying to do something similar for audio, particularly podcasts. I wrote in a Daily Update after the streaming service signed Joe Rogan to an exclusive contract:

Spotify, meanwhile, has its eyes on an absolute maxima — a podcast industry that monetizes at a rate befitting its share of attention — but as I have explained, that will only be possible with a Facebook-like model that dynamically matches advertisers and listeners in real-time, as they are streaming a podcast…This, by extension, means that Spotify needs a much larger share of the market, so that they can start generating advertising payouts that are better than the current stunted model, thus convincing podcasters to give up their current ads and use Spotify’s platform to monetize instead.

In this view the motivation for the Rogan deal is obvious: Spotify doesn’t just want to capture new listeners, it wants to actively take them from Apple and other podcast players. And, if it can take a sufficient number, the company surely believes it can create a superior monetization mechanism such that the rest of the podcast creator market shifts to Spotify out of self interest.

Capture enough of the audience and the creators will follow.

Step 3: Transformation

Still, even with the explosion of content resulting from democratizing publishing, what was actually published was roughly analogous to what might have been published in the pre-Internet world. A blog post was just an article; an Instagram post was just a photo; a YouTube video was just a TV episode; a podcast was just radio show. The final step was transformation: creating something entirely new that was simply not possible previously.

Start with text: Twitter is not discrete articles but a stream of thoughts, 280 characters long. It was the stream that was uniquely enabled by the Internet: there is no real world analogy to being able to ingest the thoughts of hundreds or thousands of people from all over the world in real-time, and to have the diet be different for every person.

From blogging to Twitter

What is interesting is the effect this transformation had on blogging; Twitter all but killed it, for three reasons:

  • First, Twitter was even more accessible than blogging ever was. Just type out your thoughts, no matter how half-formed they may be, and hit tweet.
  • Second, because blogging was so distributed and imperfectly aggregated it was hard to build an audience; Twitter, on the other hand, combined creation and consumption like any other social network, which dramatically increased the reward and motivation for posting your thoughts there instead of on your blog.
  • Third, Twitter, thanks to the way it combined a wide variety of creators in an easily-consumable stream, was just a lot more interesting than most blogs; this completed a virtuous cycle, as more consumers led to more creators which led to more consumers.

Instagram, meanwhile, had always had that transformational feed, which carried the service to its first 500 million users; it was Stories, though, that re-ignited growth:

Instagram's Monthly Active Users

Stories — which Instagram audaciously copied from Snapchat — combined the customized nature of the feed with the ephemerality inherent in digital’s abundance; the problem with posting what you had for lunch was not that it was boring, but that no one wanted it to stick around forever.

From feed to stories

This too appears to have reduced usage of what came before; while Facebook has never disclosed Stories usage relative to feed viewing, that chart above is from this August 2018 Article about Facebook’s Story Problem — and Opportunity, where I observed:

While more people may use Instagram because of Stories, some significant number of people view Stories instead of the Instagram News Feed, or both in place of the Facebook News Feed. In the long run that is fine by Facebook — better to have users on your properties than not — but the very same user not viewing the News Feed, particularly the Facebook News Feed, may simply not be as valuable, at least for now.

The opportunity came from the fact that dramatically increasing inventory would surely lead to significant growth in the long run, which is exactly what has happened. It didn’t matter that Stories were not nearly as well-composed as pictures in the Instagram feed; in fact, that made them even more valuable, because Stories were easier to both produce and consume.

TikTok is doing the same thing with video; in this case the transformative technology is its algorithm. I explained in The TikTok War:

All of this explains what makes TikTok such a breakthrough product. First, humans like video. Second, TikTok’s video creation tools were far more accessible and inspiring for non-professional videographers. The crucial missing piece, though, is that TikTok isn’t really a social network…

ByteDance’s 2016 launch of Douyin — the Chinese version of TikTok — revealed another, even more important benefit to relying purely on the algorithm: by expanding the library of available video from those made by your network to any video made by anyone on the service, Douyin/TikTok leverages the sheer scale of user-generated content to generate far more compelling content than professionals could ever generate, and relies on its algorithms to ensure that users are only seeing the cream of the crop.

YouTube has invested heavily in its own algorithm to keep you on the site, but its level of immersion is still gated by its history of serving discrete videos from individual creators; TikTok, on the other hand, drops you into a stream of videos that quickly blur together into a haze of engagement and virality.

From YouTube to TikTok

There is nothing like it in the real world.

Podcasts and Blogs

What is striking about audio is how stunted its development is relative to other mediums. Yes, podcasts are popular, but the infrastructure and business model surrounding podcasts is stuck somewhere in the mid-2000’s, a point I made in 2019 in Spotify’s Podcast Aggregation Play:

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

One of the takeaways of that piece was that monetization was holding podcasts back, and that Spotify appeared to be positioning itself to expand the podcast advertising market via centralization. Looking back, though, I should have realized that but for a few exceptions, advertising never ended up working out for blogs; the premise behind 2015’s Blogging’s Bright Future was that subscriptions made far more sense as a business model:

Forgive me if this article read a bit too much like an advertisement for Stratechery; the honest truth is my fervent belief in the individual blog not only as a product but also as a business is what led to my founding this site, not the other way around. And, after this past weekend’s “blogging-is-dead” overdose, I almost feel compelled to note that my conclusion — and experience — is the exact opposite of Klein’s and all the others’: I believe that Sullivan’s The Daily Dish will in the long run be remembered not as the last of a dying breed but as the pioneer of a new, sustainable journalism that strikes an essential balance to the corporate-backed advertising-based “scale” businesses that Klein (and the afore-linked Smith) is pursuing.

Interestingly enough, of the three authors cited in that paragraph, both Ezra Klein — formerly of Vox — and Ben Smith — formerly of BuzzFeed — are now at the New York Times, which is thriving with a subscription model. Sullivan, meanwhile, is at Substack — itself modeled after Stratechery — where within a month of launch he had reached a $500,000 run rate.

When you think about the Twitter-driven shake-out of blogging this evolution makes sense: Twitter captured the long-tail of blogs, in the process dramatically expanding the market for publishing text, but that by definition meant that the blogs that remained popular had readers that would jump through hoops — or at least click a link — to consume their content. It makes sense that the most sustainable way for those bloggers to pay the bills was by directly charging their readers, who already had demonstrated an above-average interest in their content.

My personal bet is that podcasts will follow a similar path. Podcasts, even more than blogs, require a commitment on the part of the listener, but that commitment is rewarded by a connection to the podcast host that feels even more authentic; host-read podcast advertising leverages this authenticity, but for most medium-sized podcasts charging listeners directly will make more sense in the long run.

Implicit in this prediction, though, is that podcasts actually fade in relative importance and popularity to an alternative that doesn’t simply further democratize audio publishing, but also transforms it. Enter Clubhouse.

Clubhouse’s Opening

The most obvious difference between Clubhouse and podcasts is how much dramatically easier it is to both create a conversation and to listen to one. This step change is very much inline with the shift from blogging to Twitter, from website publishing to Instagram, or from YouTube to TikTok.

Clubhouse is similar to Twitter, Instagram, and TikTok

Secondly, like those successful networks, Clubhouse centralizes creation and consumption into a tight feedback loop. In fact, conversation consumers can, by raising their hand and being recognized by the moderator, become creators in a matter of seconds.

This capability is enabled by the “only on the Internet” feature that makes Clubhouse transformational: the fact that it is live. In many mediums this feature would be fatal: one isn’t always free to watch a live video, and believe me, it is not very exciting to watch me type. However, the fact that audio can be consumed while you are doing something else allows the immediacy and vibrancy of live conversation to shine.

Being live also feeds back into the first quality: Clubhouse is far better suited than podcasts to discuss events as they are happening, or immediately afterwards. For example, both Clubhouse and Locker Room, its sports-focused competitor, have become go-to destinations for sports reaction conversations, both during and after games; it’s only a matter of time before secondary market of play-by-play announcers develops, and not only for sports: anything that is happening can be narrated and discussed.

Make no mistake, most of these conversations will be terrible. That, though, is the case for all user-generated content. The key for Clubhouse will be in honing its algorithms so that every time a listener opens the app they are presented with a conversation that is interesting to them. This is the other area where podcasts miss the mark: it is amazing to have so much choice, but all too often that choice is paralyzing; sometimes — a lot of times! — users just want to scroll their Twitter feed instead of reading a long blog post, or click through Stories or swipe TikToks, and Clubhouse is poised to provide the same mindless escapism for background audio.

COVID, China, and Controversy

Much of what I’ve written is perhaps obvious; to me that lends credence to the idea that Clubhouse is onto something substantial. To that end, though, why now?

One reason is hardware:

The fact that Clubhouse makes it so easy to drop in and out of conversation is matched by how easy AirPods make it to drop into and out of audio-listening mode.

An even more important reason, though, is probably COVID. Clubhouse launched last April in the midst of a worldwide lockdown, and despite its very rough state it provided a place for people to socialize when there were few other options. This was likely crucial in helping Clubhouse achieve its initial breakthrough. At the same time, just because COVID helped Clubhouse get off the ground does not mean its end will herald the end of the audio service, any more than improved iPhone cameras heralded the end of Instagram simply because its filters were no longer necessary; the question is if the crisis was sufficient to bootstrap the network.

I suspect so. For one there is the brazenness with which Clubhouse is leveraging the iPhone’s address book to build out its network; getting on the app requires an invitation, or signing up for the waiting list and hoping someone in your address book is already on the service, which lets you “jump the line”. This incentivizes both existing and prospective members to allow Clubhouse to ingest their contacts and get their friends on as quickly as possible.

Secondly, any suggestion that Clubhouse is limited to Silicon Valley is very much off the mark. I almost fell out of my chair while playing board games when my not-at-all-technical sister-in-law started listening to a Clubhouse while we were playing board games over the weekend, and by all accounts Taiwan is one of a whole host of markets where the app has taken off. Locker Room, as noted, appears to be the app of choice for NBA Twitter, but I suspect that is a function of Clubhouse being both gated and iPhone-only; I expect both to be rectified sooner-rather-than-later. And, of course, there is the fact the service has been banned in China.

Unfortunately, that is not the only China angle when it comes to Clubhouse; the service is powered by Agora, a Shanghai-based company. The Stanford Internet Observatory investigated:

The Stanford Internet Observatory has confirmed that Agora, a Shanghai-based provider of real-time engagement software, supplies back-end infrastructure to the Clubhouse App. This relationship had previously been widely suspected but not publicly confirmed. Further, SIO has determined that a user’s unique Clubhouse ID number and chatroom ID are transmitted in plaintext, and Agora would likely have access to users’ raw audio, potentially providing access to the Chinese government. In at least one instance, SIO observed room metadata being relayed to servers we believe to be hosted in the PRC, and audio to servers managed by Chinese entities and distributed around the world via Anycast. It is also likely possible to connect Clubhouse IDs with user profiles.

That certainly puts Clubhouse’s aggressive contact collection in a more sinister light; it also very much fits the stereotype of a new social network scrambling to capture the market first, and worrying about potential downsides later. Given the importance of network effects, I’m not surprised, but the choice of a Chinese infrastructure provider in particular is disappointing for a service launching in 2020.

The perhaps sad reality, though, is that most users probably won’t care: the payoff from uploading contacts is clear, and even if you don’t, you still need a phone number to register, which means that Clubhouse is probably reconstructing your contact list from your friends who did. The company has been far more aggressive in implementing blocking and user-reported content violations mechanism; I suspect this reflects the reality that content controversies are, in the current environment, more damaging than China connections, despite the fact that the former are an inescapable reality of user-generated content, while the latter is a choice.

Whither Facebook?

The one social network that I have barely mentioned in this Article is the social network that the FTC has sued for being a monopoly. That sentence, on close examination, certainly seems to raise some rather obvious questions about the strength of the FTC’s case.

Still, the discussion of all of these different networks really does highlight how Facebook is unique: while Twitter, Instagram, YouTube, and TikTok are all first and foremost about the medium, and only then the network, Facebook is about the network first. That is how the service has evolved from text to images to video and, I wouldn’t be surprised, to audio. This also explains why Facebook managed the shift to mobile so well; for these other networks, meanwhile, it was mobile that was the foundation for their transformative breakthroughs.

That is why I would actually give Facebook’s upcoming Clubhouse competitor a better chance than Twitter’s already-launched offering. Facebook takes innovations developed in different apps for interest-based networks and adds them to its relationship-based network; at the same time, this also means that Facebook is never going to be a real competitor for Clubhouse, which seems more likely to recreate Twitter’s interest-based network than Twitter is likely to recreate the vibrancy of Clubhouse.

The other way that Facebook looms large in the social networking discussion is monetization: it is obvious that there is an endless human appetite for social networks, but advertisers would much rather focus on Facebook’s integrated suite of properties. It is not clear that Clubhouse will even pursue advertising, though; the company has announced its intention to help creators monetize via mechanisms like tipping. This has already been proven out on platforms like Twitch in the West, and is a massive success in China (there is a reason, I should note, why the best available live streaming technology was offered by a Chinese company). It’s a smart move for Clubhouse to move in this direction early, both as a means of locking in creators, and also going where Facebook is less likely to follow.

One potential loser, meanwhile, is Spotify; the company has bet heavily on podcasts, which could be similar to betting on blogs in 2007. Still, the fact the company’s most important means of monetization is subscriptions may be its saving grace; it may turn out that Spotify is the obvious home for highly produced content, available in a more consumer-friendly bundle than the a la carte pricing that followed from blogging’s decentralized nature.


For now I don’t expect Clubhouse to be too concerned about the competition; the company said on its website when it reportedly became a unicorn:

We’ve grown faster than expected over the past few months, causing too many people to see red error messages when our servers are struggling. A large portion of the new funding round will go to technology and infrastructure to scale the Clubhouse experience for everyone, so that it’s always fast and performant, regardless of how many people are joining.

That is, obviously, the best sort of problem to have, and one that evinces product-market fit (the only thing missing is a fail whale); the fact it all seems so obvious is simply because we have seen this story before.

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


May 6, 2021: It looks like this analysis may have missed this mark; I wrote another follow-up in this Daily Update — which you can read for free — examining what I got wrong.