Skip to main content

Is Substack Notes a ‘Twitter clone’? We asked CEO Chris Best

Can Substack handle the wrath of Elon Musk and the pain of content moderation?

Share this story

Photo Illustration by Grayson Blackmon / The Verge
Photo Illustration by Grayson Blackmon / The Verge
Photo Illustration by Grayson Blackmon / The Verge

It is fair to say that Substack has had a dramatic week and a half or so, and I talked to their CEO Chris Best about it. The company announced a new feature called Substack Notes, which looks quite a bit like Twitter — Substack authors can post short bits of text to share links and kick off discussions, and people can reply to them, like the posts, the whole thing. Like I said, Twitter.

Twitter, under the direction of Elon Musk, did not like the prospect of this competition, and for several days last week, Twitter was taking aggressive actions against Substack. At one point you couldn’t even like tweets with Substack links in them. At another point, clicking on a Substack link resulted in a warning message about the platform being unsafe. And finally, Twitter redirected all searches for the word Substack to “newsletter.” Musk claimed Substack was somehow downloading the Twitter database to bootstrap Substack Notes, which, well, I’m still not sure what that means, but I at least asked Chris what he thought that meant and whether he was doing it. 

It’s tempting to think of Substack like a rival platform to Twitter, but until the arrival of Substack Notes, it was much more like enterprise software. With Substack Notes, the company is in direct competition with social networks like Twitter. It’s shipping a consumer product that’s designed to be used by Substack readers. It is no longer just a software vendor; it’s a consumer product company. And that carries with it another set of content moderation concerns, that, after talking to Chris, I’m just not sure Substack is ready for. Like, I really don’t know. You’ll just have to listen to his answers — or really, non-answers — for yourself.

This is a wild one. I’m still processing it. Let me know what you think. Okay, Chris Best, CEO of Substack. Here we go.

This transcript has been lightly edited for clarity.

Chris Best, you are the co-founder and CEO of Substack. Welcome back to Decoder.

Thanks for having me.

It has been a little over two years since you’ve been on the show. We were both effectively babies when that conversation happened. Substack had just started, I had just started with Decoder. I feel like we’ve both been on a real ride in that time.

It’s felt like a lot longer than two years, probably.

Yeah, I think for you maybe more than me. I want to ask just a very earthy question to begin with. You were really new as a CEO the last time we talked. I was reading through that interview, and at the time, I was sorting out how to ask Decoder questions and how to structure the thing. And just reading it back, maybe it was a little unfair because you were figuring out how to answer them in a CEO way. Do you feel more settled now as a CEO? 

“As soon as you start to feel like you’re getting okay at it, you just earn the right to go to the next mini-game that you’re newly bad at.”

I’ve learned a ton in the past two years, and I feel like being a CEO of a startup like this, the job is being bad at the current thing. And then, as soon as you start to feel like you’re getting okay at it, you just earn the right to go to the next mini-game that you’re newly bad at. And so I always feel like I’m constantly learning a lot and struggling to be on top of the thing that I currently have to do. But the whole time, I feel like I’m learning a ton, and it’s been very exciting.

Yeah. I think it’s just a question that rarely gets asked of people who are on this kind of journey. Did it click in? What parts of it do you feel more confident in now than you did two years ago?

The story we’re telling hasn’t really changed, but the way that we know how to tell it, and our confidence in it, and our confidence in what we’re building and what that means for the world has gone up and up. I think the team that we’ve built here is incredible — that’s feeling really good. We figured out how to make a team that can do ambitious things and have a group of exceptional people accomplish audacious goals together. All that stuff has come a long way.

I actually want to talk about that story a little bit. So, Substack has told a lot of stories about itself in the past two years since you’ve started. I think you all started with a big story about how social media was a disaster and the attention economy was a mess. You’ve stayed committed to that. But the part of the story that has changed the most, as I look back at it, is what Substack is and what is the product.

I think of Substack as an enterprise software company. You provide enterprise software to writers who then go build businesses with it, and those businesses can look like one of 100 different kinds of things. But at the core of it, you’re their vendor building a subscription email newsletter product. Is that how you think of it, that this is an enterprise software company? Or is it getting more consumer over time — which is the thing that I would candidly say is the change that I’m sensing?

So the thing that I’ve always loved about Substack, the thing that convinced me to work on this in the first place, is that at any given moment we’ve always had a really big, audacious version  of the thing we’re trying to do. There is a big vision here, and that thing that you gestured at at the start really hasn’t changed. We’re building a new economic engine for culture. We’re building a new part of the internet that’s based on different laws of physics, like a different business model — subscription instead of ads. It’s a different way of relating to people, where you subscribe directly to the people you trust rather than signing up for the platform as a whole. And the potential of that thing to really change the world and to inspire people to want to be a part of it is really compelling.

And we’ve always had a very concrete next step of something that we could do that gets us an inch closer to that vision. And at the very start, it wasn’t “build an enterprise software thing”; it was, “help one person turn on page descriptions in their email newsletter.” It was the very smallest instantiation of, “Well, but now that I’ve turned on paid subscriptions in my email newsletter, I own my audience. I have a connection directly to them, I can reach them in their inbox. I can bring my email list with me, I can import it, I can export it. I can get paid directly by my audience. That changes the game of what kind of stuff I’m allowed to create.”

And so, we see that the journey of Substack has been tackling a series of these concrete next steps that get us ever closer to this big thing that we set out to do. We’re still pretty early on if we’re right about the size of the thing that we’re eventually going to be able to create. And you mentioned that this is just a tool — enterprise software, whatever. Is this a tool that is a thing for writers that readers don’t even need to know about? Maybe it’s just Shopify for writers or something like that? Should that be the vision for Substack?

The reason I think that is too reductive and would be too small of a vision is that I think the power of the network that we’re building is a big part of the value that we can create for writers. And so, we don’t want to say, “You can just go off and use our publishing-in-a-box software to start your own business.” It’s great that you can do that. That’s a necessary step one. But you want the power of a network. You don’t want to be totally off the grid, having to figure everything else out yourself. You want to be able to plug into this. You want to have the benefit of independence, of owning your work and owning your list, and also the benefit of being part of a network that helps you grow and interact with other people — that helps bring all of that value. And I think we’ve shown now that we can do both of those things and that, together, the whole is greater than the sum of the parts.

So I want to hold onto that as we go through this conversation because figuring out where Substack lives in this stack of technologies that create user experiences is important for a host of reasons, and I think the most important one is that the closer you are to the consumer experience, the more responsible for things like moderation you have to be. 

I don’t think Comcast, or AT&T, or whoever my ISP is should be responsible for content moderation. In fact, I think they should do zero of it. That’s the idea of net neutrality. I don’t know if Cloudflare should do it or another web host or service provider. I definitely think Facebook should do it. I definitely think YouTube should do it. And there’s a gradient there, and Substack is in maybe the grayest part of that zone where it’s a brand and now you want me to put an app on my phone, but you do provide an enterprise software product to people, and I think you would prefer to be just the pipes and let the consumers decide. I think that has been where most of your challenges have come from — applying one set of expectations to a different part of your product — and I’m wondering if that has gotten any clearer for you.

So the big vision is we’re building this new economic engine for culture. What that actually looks like is a subscription network. And the thing that we started out believing and I think have become more convinced of over time is that doing this in a way that’s vertically integrated allows us to make something truly new. So the fact that both we are the tool that the writer uses to compose the thing they’re writing and we’re sending the email the reader sees or that we’re building the app that the reader is using to read it in lets us make a combined experience that is different and better than would be possible at all if you just had a bunch of different things that were one little piece of that stack. And so, we are trying to build this new thing and having Substack be providing the different pieces of that together is a core part of what makes it work.

So I think that brings us, inevitably, to Substack Notes, which is the new feature you all just launched. When we wrote about it, we said, “Hey, this looks a lot like Twitter.” I think the guy who runs Twitter looked at it and said, “Hey, this looks a little too much like Twitter.” We’ll come to that part of it, but tell us what Substack Notes is and what you want it to do.

So Substack Notes is a way for writers on Substack to share shortform posts and recommendations on the Substack network and help them grow. So Substack is a subscription network. People don’t subscribe to Substack; they subscribe to individual writers. And a lot of what writers do to grow is promote their work and get into conversation with other people to share recommendations and ideas. And Notes is a place where you can do that within the Substack network.

So I am very sympathetic to this. We launched a thing on our own site called “Quick Posts” because not all writers want to write a whole story every time. Sometimes you just want to write something that’s really short and say “look at this thing” or “go read this other great story.” So I get that motivation very clearly. Our ideas are probably converging more than they’re not in that sense. But Substack Notes really does pull a lot of ideas from Twitter, Facebook, and other social networks into the Substack app. And I’m just wondering how you think about the tension between that and then all the rhetoric around social networks being bad and the attention economy being unhealthy.

“The readers on Substack are the customers. On a social network, it’s the advertisers.”

So I think that the difference between Substack and a social network is not in how it looks. The difference is the business model. The difference is what you don’t see. You don’t see ads, you don’t see the incentive structure that ads necessarily create. It runs on a totally different business model, it runs on paid subscriptions. The customers are different — the readers on Substack are the customers. On a social network, it’s the advertisers.

Wait, you don’t have to pay to use it? You can just download the Substack app, read everyone’s free Substack posts, and then read their Notes for free?

You can read their Notes for free, but at some point, you may discover that when you find writers who you deeply value and you care about what they say and they have a paid Substack, that might be something that you choose to be a part of and is actually a big driving force for why the whole thing works. And so, to look at Substack Notes and say “well, it looks like other products that I’m familiar with” is looking at a Tesla and saying, “It’s the same as an Aston Martin because they both have a steering wheel.” You drive them, they’ve got four wheels. They’re completely different because the thing that powers them, the fuel, is completely different.

Was it meant to compete with Twitter?

No.

What was it meant to do besides post recommendations? Because it feels like you want writers putting more content in the app so people use the app more, which I totally understand. Again, that’s our motivation for our product and our site: we want our writers to participate more, be more present, build more audience, build more community.

I think that the incentive structure of the social media business model pulls in a certain direction. It pulls in this direction of being maximally cheaply compelling, maximally addictive, and trying to get you to spend more and more of your time there, regardless of how much you value it. And I basically think that the truest instantiation of that today is TikTok. And I think that every company that has this business model is going to get pulled in the direction of getting closer and closer to TikTok and then whatever comes beyond TikTok. TikTok but everything’s AI, or TikTok plugs into your brain, or whatever it is. I don’t know.

There’s some gravitational pull that’s pulling every platform that works that way to be that way whether they want to or not. And I think that opens up an opportunity for something that is in opposition to that that works a totally different way. That says, “Hey, over here, you are the customer. You are going to subscribe directly to things you care about. The job of this app, of this inbox, of this feed is not to keep you here at all costs. It’s to find you things that you value so much that you might want to pay for them.”

“Everybody’s going to either have to turn into TikTok or turn into Substack. We are already Substack.”

And so, my mental model of this is basically that everybody’s going to either have to turn into TikTok or turn into Substack. We are already Substack. In the broad sense, that creates an alternative to the attention economy. Substack, as a whole, is an alternative to Facebook, Instagram, Twitter, and TikTok, and over time, we think this alternative model will grow. But it’s just obviously not the case that we’re going to release one feature and everybody from some other thing is going to jump over. It doesn’t work that way.

Just to be clear, the users of Notes are expecting that. Just browsing Notes for the past couple days that I’ve had access to it, there’s a lot of hope that everyone will just move over from Twitter. Obviously, there’s some Twitter-related drama that we should get into very directly. But right now, looking at the users of Notes, who are a bunch of Substackers, they’re saying, “Hey, this feels like an early Twitter. I hope we can keep this going and this replaces Twitter.” Do you think that’s a fair or appropriate expectation?

I’ve seen the meme that you’re talking about. That’s not the story that we’re telling, and that’s not the way that we believe it’s going to work. We don’t think that Substack anything is going to be the new Twitter. And we think, even if that were possible, that’s not what we’re trying to do. We’re not trying to build the same thing.

I can understand how people, especially writers who really want to have a reliable platform for their work, a reliable place to share, could be feeling the heat and looking for alternatives. I think that simplistic understanding just doesn’t hold water.

Let’s talk about that part of it for just a quick second. The last time we talked, you said that you were keeping track of various writers’ Twitter followers as you were considering recruiting them to Substack, and there have been a lot of different deals that recruit writers to Substack. But the idea that people with a lot of Twitter followers were attractive targets to recruit to Substack came through loud and clear.

I have certainly heard from a lot of people in Substack over the past two years that their primary source of conversions to paid is Twitter. Twitter is their best marketing engine. Substack maybe monetized Twitter better than Twitter ever monetized itself. This has been a real relationship that we’ve seen play out. And now, Twitter’s going through whatever it’s going through. I think there is a lot of fear from that community that their single best top-of-funnel network is going away or will be inhospitable to them in some meaningful way.

And Notes might be the replacement, or Mastodon might be the replacement, but it’s certainly not going to be TikTok, and it probably isn’t search. Are you thinking, “Okay, we’ve got to build something that looks like the top of the funnel for Substack writers”?

I think the right way to think about this is we’re trying to build ways in Substack where you can use the power of the Substack network to grow, to reach new audiences, to have a place, in a Substack-y way. An example of this is the recommendation feature we launched where writers can recommend each other. Therefore, readers can discover new things they might like, not through a machine that’s just predicting what you’ll click on but somebody that they’ve chosen to trust saying, “Hey, this is worth checking out.”

We’re extending that ethos into how Notes works. It’s just a way that you can recommend even more things. And at the same time, we want people publishing on Substack to be able to publish everywhere. We want people to also share on Twitter. We want them to also share on TikTok, also share everywhere. We think writers should have the power to share their work broadly everywhere that it can go. We think that’s good.

Okay. Now, we’ve got to get into it. I’m going to ask just a foundational question first. Before Elon [Musk] took over, Twitter had a competitor to Substack. It was called Revue. What was Substack’s relationship to Twitter like in the previous administration?

I think it was fine. It was good. One thing that I will say is there’s a lot of Substack users who use Twitter. And we’ve always thought that having that work really well and reliably was good. It’s nice to be able to embed a tweet into your Substack post. It’s nice to be able to look up which of your people you follow has a Substack that you might want to subscribe to. We’ve built a bunch of pieces on the Twitter API that plug in and kind of make life better for people who are using both platforms, and we think that’s good. That’s something we would like to offer.

From what I’m getting from your answer is you weren’t even thinking about your relationship to Twitter in the previous administration. It was just sort of operating the way that we expect the internet to operate. You were using their API, and that was fine.

Yeah.

Okay. I’m just curious because sometimes, for all I know, you and Jack Dorsey had 10,000 screaming matches, but it doesn’t seem to be the case. 

So then Elon buys Twitter. All the things happen. Last week, he made this claim: “Substack was trying to download a massive portion of the Twitter database to bootstrap their Twitter clone, so their IP address is obviously untrusted.” He puts up a warning against Substack links. He starts throttling posts that have Substack links in them. He blocks the word Substack from being searched, which is incredible. What happened?

So last week, we announced Notes, this thing we’ve been talking about. It’s a way for writers to share shortform posts. We were really excited about this launch. We think it’s a big, exciting thing. But we were not expecting it to get anything close to as much attention as it’s since got.

Twitter saw Substack Notes as a threat and reacted very strongly.

And really surprising to us, Twitter saw it as a threat and reacted very strongly. They took a bunch of actions that ultimately hurt writers who are users of both Substack and Twitter in these cases. They’re throttling links. They’re falsely marking them as unsafe and even trying to throttle discussion of the word Substack. And all of that was incredibly disappointing to us. We think writers should be able to freely share their work. It’s one thing to react to us. It’s another thing to kind of take it out on writers, many of whom are your own users.

And so, we were really disappointed. And also, there was quite a strong backlash. A lot of people felt, as we did, that this was not a good path. And fortunately, by Saturday, it seems like this has reversed course. The links have stopped being restricted. A bunch of the stuff has been rolled back. But in our mind, this whole episode has been just a reminder of why it’s so important for writers to have a platform that’s reliable and a platform where they own their audience.

Just to be clear, Elon’s claim is: “Substack was trying to download a massive portion of the Twitter database to bootstrap their Twitter clone.” Were you doing that?

No. And it’s one of several claims that got bandied around during this time. It’s not true.

Why do you think Elon thinks Substack was trying to download a massive portion of the Twitter database?

I don’t know. And if I had to speculate, my guess is that they’re looking for reasons to justify these actions in the face of a lot of pushback.

Have you had any conversations with Elon this past week?

We’ve been trying every route that we can to calm this down and sort of find a peaceful resolution that can help writers. That hasn’t totally worked, and we haven’t had any answers to our specific questions of, “Are there things that we could do to make this better?”

Yeah. Tell me what your specific questions are.

Are there things that we could do to resolve this? They haven’t answered with any specifics.

And they haven’t shown you any evidence of their claim about the Twitter database?

No.

And then, he just flipped it back on by himself? Did you have any warning that, okay, it’s going to stop being throttled?

No.

It’s amazing that you don’t know the answers to these questions. I’m just befuddled. I have a lot of questions to ask you, but I just want to put a pin in this. You experienced this because he just turns off embeds for Substack, so Substack writers can no longer embed Twitter posts. And then he turns off links and throttles it. How do you find out that this is even happening?

We start to get complaints that things are broken.

And so, you look at a spike of complaints and you say, “What’s broken with our Twitter situation?” What’s the next thing you did?

The team started looking at how to fix it. What can we do? 

And then, when you discovered it was unfixable because Twitter had taken some technical steps, you’re the CEO. What’s your next move?

Unfortunately, we’ve had to pause all usage of the Twitter API, this API that we’ve been using for years to power these really nice quality-of-life features in Substack. We’ve had to completely pause using it because we want it to be a reliable thing for writers. And other than that, we were left a little bit guessing what was happening.

It has been noted to me by a bunch of reporters on our staff and by people out in the world — Andreessen Horowitz, big investor in Substack, I think they led your last round — they are an investor in a syndicate that now owns Twitter with Elon. Were they of any help to you?

They’ve been a huge help to us in general, but it is the case that they’re an investor in both companies.

Did they convene a meeting? Did you have a dinner at Marc Andreessen’s house?

We’re talking throughout all of this.

And Twitter is just basically radio silent to you this whole time? That’s what it sounds like, that Twitter has not said anything to you of value.

Certainly we don’t know much more than the world at large has been able to see.

What do you think happens next here?

It’s hard to say. I’ll say what I would hope would happen. The thing that I would hope would happen is that the things that happened that have been disrupting to writers stop happening and don’t happen again and it’s made clear that this is not the way to handle these kinds of things.

I would love to get in a position where we could unpause our API usage. I’d love to make any changes that they ask for. I’d love to get in a place where whatever we can support that makes us a reliable partner for writers who also use Twitter could come back on and work. And frankly, I would love to see Notes be a tremendous success. I’d like for people to be having conversations, growing their audiences, sharing this, alongside their use of Twitter, which I think is the best possible scenario.

You mentioned this whole situation has kind of proven the point that being reliant on one company to make decisions, especially a company run by a guy who’s been running around talking about free speech and then he’s literally blocking words from being searched, it just kind of exposes the problem.

Now, you’ve got a big consumer-facing social feed that looks like Notes. You have a bunch of that ability now. If you wanted to block the word Twitter, you could probably do it. Why should we trust you?

I think the best answer to this is we’ve designed the business model of Substack from the ground up to put writers and readers in charge. People don’t follow you on Substack. They subscribe. You have an email. You can email them. You can reach out to them. You own your editorial decisions. You own your relationship with your audience. You could leave Substack if you want to.

And so, on the one hand, it’s easy for me to say, “Well, we’re just not going to do that because it’s not the right thing to do. We’re here to be a reliable partner for writers. It doesn’t make sense for our business.”

But also, we’ve tried to set Substack up in a way that puts readers and writers in charge in a way where, even if we wanted to do that, we’d be constrained from doing it.

Yeah. I think the reason most social networks at any kind of scale don’t do this stuff is because it’s wholly irrational. But now, we are looking at, well, it happened.

Well, listen. They don’t do this version. They don’t do the sharp version where they change the meaning of words on their platform or something like that. But they do do the other version. A lot of people come to Substack — it’s not as remarked upon — but a lot of people come to Substack who use Instagram who worry that their following on Instagram doesn’t mean the same thing that it used to mean. If you talk to people who have worked on these feeds, they’ll tell you if you come up with a way to put something in the Facebook feed that gets you to read a long article or watch a long video and you put that in the feed, it tanks the metrics.

And the team that runs that thing says, “Take this out. It’s costing us a ton of money that you found somebody this piece of content that they love and want to deeply engage with.”

And so, while you’re not going to see, I hope, the overt version of this, you’re seeing, across all of these networks, the more subtle versions of it. And Substack is a place that just, by virtue of the business model, we’re more aligned with the writers.

Is this feed algorithmic today? 

It’s algorithmic, but it’s algorithmic in service of the user instead of the users having to serve the algorithm, by which I mean it’s based off of the choices people are making. It’s based off of, “I subscribe to you. You recommend somebody.” We’re sort of building this trust graph in the network, and everything that goes into the feed is sort of in service of that.

I think I might just be a tick too unsophisticated to understand that. That sounds exactly like what Facebook would say. It’s based on your actions, who you are following, what you like, what your friends like, what Wi-Fi networks you’ve been on that your friends have been on. That’s how the Facebook algorithm works.

I think there’s a difference between the implicit “what Wi-Fi networks your friends have been on,” versus, if I subscribe to your Substack and you choose to recommend, you say, “Hey, this other thing is good. This is part of my network. This is like we’re forming this scene, this community that has an emergent set of norms, that has an emergent discourse that’s happening within it.” And there’s human beings that are making decisions about what that is and how it should be. And the platform is not trying to work against those decisions in order to get you to see as many ads as possible but is instead trying to work with them to help you discover things you deeply value. It’s a completely different model and a completely different outcome.

I’m really interested in the action around the decentralized social networks like Bluesky and Noster and Mastodon. I’m really interested in what’s going on with ActivityPub, which Matt Mullenweg from WordPress was on and he said Tumblr’s going to support ActivityPub and Mozilla’s supporting it. There’s a lot of energy around decentralizing these networks so that one person can’t just block the word Substack from being searched. Is Notes going to be compatible with ActivityPub or Bluesky? Have you considered that stuff?

That stuff is really interesting. I sort of mentioned before that we want stuff that gets published on Substack to go everywhere. So we have RSS feeds for existing content. And we’re really interested in, are there ways that we could help this stuff spread? I don’t think we have a specific plan with any of these protocols, but we’re really interested, in general, in how we can help people on Substack have their work travel everywhere in the world as frictionlessly as possible. You talked about the levels in the stack — the question I might ask on some of these things is, “What’s the right level in the stack to return power to people? What’s the right level in the stack to say you’re going to have control of your experience on these things?” And I think one answer might be the protocol level — and protocol level even implies the hardware level. You could own your own server that runs your own Mastodon instance that then does this thing and yada yada yada. 

There’s various places that you could try and introduce freedom and control. And I guess the Substack theory on this in general is that we want that freedom of control to exist at a higher abstraction level. We want you to have editorial freedom over what you write. We want you to have freedom to decide who you’re subscribing to, which communities to be a part of, how the communities federate together but not have to have the responsibility to run your own software or figure out how to do all of the little pieces of it. We kind of want to give you this whole package that lets you do that.

Did you have the product conversation — like, “We should just build this on Mastodon” — and choose against it?

No.

But have you thought about how Substack would hook into a Bluesky or an ActivityPub or whatever it is?

We’re interested in this. I don’t think we have any settled course on all that stuff.

I just hear a lot of energy, and I don’t know if any of them are the answer, but it seems interesting that, in this moment where you are experiencing the most power of a closed platform that can be expressed, your answer isn’t, “Oh, we should make sure ours is open.” Your incentives in the business might prevent you or at least disincentivize you from making these kinds of unilateral censorship moves, but the product still allows you to — and that, it’s just interesting that you haven’t quite had that immediate reaction.

I do think it’s interesting because I’m sort of very philosophically sympathetic to these platforms and these arguments. I’m a big believer in the open internet. I’ve grown up appreciating a lot of this stuff, and I’m reminded a little about the period where everybody would be asking me these same things but about Web3 stuff, right? Things like, “When is Substack going to become a trustless protocol on the blockchain?” And the biggest problem I always came back to is, “Is this great for users? Are we able to use this to make something that’s great for users?”

And often, the answer was no, in that case, or we weren’t able to find a way where the answer was yes. We haven’t finished the exploration, but the question I would ask with all of these protocols is, “Can we make a great experience for users?” And I worry sometimes that people take this leap you’re saying where it’s like, well, there’s something out there that we haven’t been able to trust. And so the answer is a system where you just don’t have to trust anybody. You can only trust in yourself. And I actually think probably a better answer is not a trustless system. A better answer is to have a platform you can trust.

And trust it by leaving, right? That’s actually the biggest expression of trust is, “I could leave, but I choose to stay here.”

And we made that decision very early on in Substack — that you’re going to be able to leave at any time, you take your relationships, you take your payment relationships even, and that put us under a lot of pressure. We take 10 percent as the fee. That’s how we make money. We only make money when you make money.

And so, at some point at the start, that was a great deal. People are like, “Oh, it’s free. I can publish my email. This is a wonderful thing.” And then they start making millions of dollars a year, and they say, “Wait a minute. Outside of all of this heavy philosophical stuff, how much money am I paying you, and is it worth it?” And the thing that’s caused us to do is to be a reliable, trustworthy partner for writers and to serve them, right? We’ve built this network where we can go to them now and say, “Look, you’re paying this 10 percent fee, but look over here. You’re getting 15, 20, 30 percent of your new paid subscriptions from this network.” It’s forced us to have to serve the people who are using the platform. That’s the thing we try to do. We try to make our business model aligned so that you can trust that we’ll do it, not only because we hope that we’re good people but because we’ve set up a system that just works that way.

I don’t want to dwell on this too long because Substack Notes is new.

It’s one day old at this point.

But it’s notable to me that Substack Notes is closed. You can’t leave it if you build up a substantial following there, and I think there’s a tension there that you will have to sort out over time.

Well, it is still based on the exact same subscription network, so you can leave it in the sense that you can go and take your audience with you.

But not in Notes. That social graph stays with you. Again, it’s one day old. If you end up having this problem–

Well, let me actually correct you. That’s not true.

It’s not true?

It is based on the subscription graph. Substack is a subscription network, and so, when people subscribe to you on Notes, you can leave.

You can leave and you take your set of subscribers and you get all their emails. Is that why you can’t follow someone? When I asked my team what questions they have, the feature requests came in fast and furious, and one of them was, “Can I just follow someone without subscribing to the newsletter?” And you’ve consciously connected those two things.

I think Casey [Newton] put this as, “I wish I could flirt with people before I had to make a commitment.”

Casey Newton.

Yes. Casey’s substack is Platformer.news. It’s a good Substack. You should subscribe. Obviously, we’re still figuring out how these things work. We’re still building it. But this is the thing that we ultimately want to do. We do think that the subscription network is the thing that sets Substack apart. We want you to have a direct relationship. This is a good example, actually. People come to Notes and say, “Well, this looks like everything else. Why can’t I follow people?” Well actually, this is pretty different, and we’re looking at ways to let people configure their experience and all this stuff, but the root of it for us is still the subscription graph because, at the end of the day, that’s the thing the writers want, and that’s the way that they’re going to get paid and we’re going to get paid.

A quick note — Chris and I are about to talk about Substack’s content moderation guidelines. Like I said at the top of the show, it makes sense to me that you have looser guidelines the more you provide infrastructure, like an email service to people, and stricter guidelines the more your product looks like a consumer service. Substack has pretty loose guidelines, and I wanted to know if it would tighten those guidelines for Substack Notes, the new feature that looks like Twitter.

I want to call out that I got something wrong here — I came up with what I thought was an easy hypothetical, about whether posts calling to kick brown people out of the country would be moderated on Substack Notes. I thought it was a gimme because, well, obviously, but also because I read Substack’s content guidelines a little too loosely. Here’s the relevant section, under the topic of “Hate”:

“Substack cannot be used to publish content or fund initiatives that incite violence based on protected classes. Offending behavior includes credible threats of physical harm to people based on their race, ethnicity, national origin, religion, sex, gender identity, sexual orientation, age, disability or medical condition.”

Now, I think it’s debatable whether calling to kick brown people out of the country incites violence — I think it does, but I can see the argument that, in my example, it literally does not. I wish I had used a clearer example. That’s on me. But I think it’s more notable that Chris didn’t correct me either way and actually didn’t engage the question at all, which… well, you’ll see how that went. Alright. Back to the interview. 

All right, last question on Notes. Notes is the most consumer-y feature. You’re saying it’s inheriting a bunch of expectations from the consumer social platforms, whether or not you really want it to, right? It’s inheriting the expectations of Twitter, even from Twitter itself. It’s inheriting the expectations that you should be able to flirt with people and not have to subscribe to their email lists.

In that spectrum of content moderation, it’s the tip of the spear. The expectations are that you will moderate that thing just like any big social platform will moderate. Up until now, you’ve had the out of being able to say, “Look, we are an enterprise software provider. If people don’t want to pay for this newsletter that’s full of anti-vax information, fine. If people don’t want to pay or subscribe to this newsletter where somebody has harsh views on trans people, fine.” That’s the choice. The market will do it. And because you’re the enterprise software provider, you’ve had some cover. When you run a social network that inherits all the expectations of a social network and people start posting that stuff and the feed is algorithmic and that’s what gets engagement, that’s a real problem for you. Have you thought about how you’re going to moderate Notes?

We think about this stuff a lot, you might be surprised to learn.

I know you do, but this is a very different product.

Here’s how I think about this: Substack is neither an enterprise software provider nor a social network in the mold that we’re used to experiencing them. Our self-conception, the thing that we are attempting to build, and I think if you look at the constituent pieces, in fact, the emerging reality is that we are a new thing called the subscription network, where people are subscribing directly to others, where the order in the system is sort of emergent from the empowered — not just the readers but also the writers: the people who are able to set the rules for their communities, for their piece of Substack. And we believe that we can make something different and better than what came before with social networking.

The way that I think about this is, if we draw a distinction between moderation and censorship, where moderation is, “Hey, I want to be a part of a community, of a place where there’s a vibe or there’s a set of rules or there’s a set of norms or there’s an expectation of what I’m going to see or not see that is good for me, and the thing that I’m coming to is going to try to enforce that set of rules,” versus censorship, where you come and say, “Although you may want to be a part of this thing and this other person may want to be a part of it, too, and you may want to talk to each other and send emails, a third party’s going to step in and say, ‘You shall not do that. We shall prevent that.’”

And I think, with the legacy social networks, the business model has pulled those feeds ever closer. There hasn’t been a great idea for how we do moderation without censorship, and I think, in a subscription network, that becomes possible.

Wow. I mean, I just want to be clear, if somebody shows up on Substack and says “all brown people are animals and they shouldn’t be allowed in America,” you’re going to censor that. That’s just flatly against your terms of service.

So, we do have a terms of service that have narrowly prescribed things that are not allowed.

That one I’m pretty sure is just flatly against your terms of service. You would not allow that one. That’s why I picked it.

So there are extreme cases, and I’m not going to get into the–

“We should not allow as many brown people in the country.” Do you allow that on Substack?

Wait. Hold on. In America in 2023, that is not so extreme, right? “We should not allow as many brown people in the country.” Not so extreme. Do you allow that on Substack? Would you allow that on Substack Notes?

I think the way that we think about this is we want to put the writers and the readers in charge–

No, I really want you to answer that question. Is that allowed on Substack Notes? “We should not allow brown people in the country.”

I’m not going to get into gotcha content moderation.

This is not a gotcha... I’m a brown person. Do you think people on Substack should say I should get kicked out of the country?

I’m not going to engage in content moderation, “Would you or won’t you this or that?”

That one is black and white, and I just want to be clear: I’ve talked to a lot of social network CEOs, and they would have no hesitation telling me that that was against their moderation rules.

Yeah. We’re not going to get into specific “would you or won’t you” content moderation questions.

Why?

I don’t think it’s a useful way to talk about this stuff.

But it’s the thing that you have to do. I mean, you have to make these decisions, don’t you?

The way that we think about this is, yes, there is going to be a terms of service. We have content policies that are deliberately tuned to allow lots of things that we disagree with, that we strongly disagree with. We think we have a strong commitment to freedom of speech, freedom of the press. We think these are essential ingredients in a free society. We think that it would be a failure for us to build a new kind of network that can’t support those ideals. And we want to design the network in a way where people are in control of their experience, where they’re able to do that stuff. We’re at the very early innings of that. We don’t have all the answers for how those things will work. We are making a new thing. And literally, we launched this thing one day ago. We’re going to have to figure a lot of this stuff out. I don’t think…

You have to figure out, “Should we allow overt racism on Substack Notes?” You have to figure that out.

No, I’m not going to engage in speculation or specific “would you allow this or that” content.

You know this is a very bad response to this question, right? You’re aware that you’ve blundered into this. You should just say no. And I’m wondering what’s keeping you from just saying no.

I have a blanket [policy that] I don’t think it’s useful to get into “would you allow this or that thing on Substack.”

If I read you your own terms of service, will you agree that this prohibition is in that terms of service?

I don’t think that’s a useful exercise.

Okay. I’m granting you the out that when you’re the email service provider, you should have a looser moderation rule. There are a lot of my listeners and a lot of people out there who do not agree with me on that. I’ll give you the out that, as the email service provider, you can have looser moderation rules because that is sort of a market-driven thing, but when you make the consumer product, my belief is that you should have higher moderation rules. And so, I’m just wondering, applying the blanket, I understand why that was your answer in the past. It’s just there’s a piece here that I’m missing. Now that it’s the consumer product, do you not think that it should have a different set of moderation standards?

You are free to have that belief. And I do think it’s possible that there will be different moderation standards. I do think it’s an interesting thing. I think the place that we maybe differ is you’re coming at this from a point where you think that because something is bad… let’s grant that this thing is a terrible, bad thing...

Yeah, I think you should grant that this idea is bad.

That therefore censorship of it is the most effective tool to prevent that. And I think we’ve run, in my estimation over the past five years, however long it’s been, a grand experiment in the idea that pervasive censorship successfully combats ideas that the owners of the platforms don’t like. And my read is that that hasn’t actually worked. That hasn’t been a success. It hasn’t caused those ideas not to exist. It hasn’t built trust. It hasn’t ended polarization. It hasn’t done any of those things. And I don’t think that taking the approach that the legacy platforms have taken and expecting it to have different outcomes is obviously the right answer the way that you seem to be presenting it to be. I don’t think that that’s a question of whether some particular objection or belief is right or wrong.

I understand the philosophical argument. I want to be clear. I think government speech regulations are horrible, right? I think that’s bad. I don’t think there should be government censorship in this country, but I think companies should state their values and go out into the marketplace and live up to their values. I think the platform companies, for better or worse, have missed it on their values a lot for a variety of reasons. When I ask you this question, [I’m asking], “Do you make software to spread abhorrent views, that allows abhorrent views to spread?” That’s just a statement of values. That’s why you have terms of service. I know that there’s stuff that you won’t allow Substack to be used for because I can read it in your terms of service. Here, I’m asking you something that I know is against your terms of service, and your position is that you refuse to say it’s against your terms of service. That feels like not a big philosophical conversation about freedom of speech, which I will have at the drop of a hat, as listeners to this show know. Actually, you’re saying, “You know what? I don’t want to state my values.” And I’m just wondering why that is.

I think the conversation about freedom of speech is the essential conversation to have. I don’t think this “let me play a gotcha and ask this or that”–

Substack is not the government. Substack is a company that competes in the marketplace.

Substack is not the government, but we still believe that it’s essential to promote freedom of the press and freedom of speech. We don’t think that that is a thing that’s limited to…

So if Substack Notes becomes overrun by racism and transphobia, that’s fine with you?

We’re going to have to work very hard to make Substack Notes be a great place to have the readers and the writers be in charge, where you can have the kinds of conversations that you find valuable. That’s the exciting challenge that we have ahead of us.

What do you want it to be?

I want it to be not any one thing. This is something that I see of Substack as a whole. If you look at the leaderboards on Substack, if you look at what’s successful on Substack, it looks to me like sort of an index fund of culture. It’s not the place for anyone. Everybody has ideas of what Substack is — it’s for famous journalists who struck it on their own, or it’s for this or that. They have some pocket of people that they know of, but I think if Substack can live up to the grand ambition that we have for it, it will not feel like one monolithic place that has one set of rules, that has one vibe or one overarching thing, but it’ll feel like a place where every Substack is its own island, as it is today, where you can have your own rules for your community, where you can set quite strict rules. You’re operating within the very loose sort of overall platform-wide rules and where writers and readers exercise their freedom to make those spaces feel very different and to have different experiences.

How does that express itself in Substack Notes? If I post a Note and I don’t want people to participate in a certain way, do I have the moderation tools to prevent that from happening?

Yes. Today, they’re pretty basic. You can mute people. You can block people. We’re very shortly going to be adding the ability to limit replies to paying subscribers of your Substack. I think there’s a lot more that we can do beyond those basic tools as we build this thing out in partnership with the people that are using it to give you more and more power to shape your experience and the experience around the things you make.

If someone wants to set a set of moderation standards on Substack Notes for their community that runs right into your terms of service, how do you make that determination?

We review it compared to the terms of service.

Who does that review?

We have a team and, ultimately, we, the founders, do it if it is a big question.

I mean this runs into the, I would say, the standard Decoder questions. Substack, the last time I talked to you, was 20 people. I think it’s around 90 now. You had some layoffs, but it’s around that size. Right?

Eighty.

Eighty. How many of those people are trust and safety folks?

A handful.

There’s a bunch of stuff that any standard-size platform has to really contend with, like copyright infringement. I’m assuming I can’t just post the full Moana movie to Substack Notes, right?

Of course.

How are you handling that stuff?

We’re following the DMCA [Digital Millennium Copyright Act] process.

Is that automated, or do you have a set of people doing it? For example, YouTube has an army of people that manage the DMCA.

Yeah, it’s a combination.

Do you think that’s going to scale as quickly as it has to scale for every other platform? This is an escalating set of moderation costs. DMCA compliance, all on its own, is an ever-escalating set of compliance costs for every social network.

Yeah, I think, as we scale, there’s going to continue to be all of these things. We’re going to have to solve these problems. We’re not going to have them, I think, in the exact same way that others experience. A big one is spam. A lot of people want to come and spam. That’s a real thing that we have to deal with. We have the option to look at what’s come before and structure the network in ways that make that structurally harder. Even given that, there’s no silver bullet solution. We’re going to have to have mitigation efforts that happen.

The reason I asked specifically about costs and how they might rise is because, the last time you were on, we talked about the Y Combinator idea of “default alive,” and you said, “Look, if we just left everything alone right now, we’d be profitable. We want to be default alive. We have to invest in order to grow.” The last funding round you had was led by Andreessen Horowitz. You raised $65 million at a $650 million valuation. There’s reporting that you tried to raise again last year and you scrapped it. Then, you recently launched a retail investor program where you’ve raised about $7 million. As part of that, you had to release some financial statements from 2021, and your costs are higher than your revenues. In fact, you have negative revenue because you’re paying out to your authors before their revenue pays you back. How did we get in that situation?

“We’ve kind of bootstrapped this thing.”

The way that we thought about this, especially in 2021, was we knew we wanted to bootstrap this network. We wanted to start this flywheel going, and we wanted to accelerate that. We were like, “People are joining Substack, this is happening, but we want to find ways to use money to make that happen as quickly as possible.” We raised a bunch of money, we spent a bunch of money doing that, kick-starting that thing. The short version of this is that it worked. We started a network that’s very powerful. A bunch of the people who were attempting to copy Substack, you mentioned Revue, Facebook made a Substack clone competitor thing, which I think missed the point as well.

In our estimation, we’ve kind of bootstrapped this thing and got it started in a way where, today, Substack is clearly the best place if you’re starting one of these subscription things in the Substack mold. There’s this tremendous effect of having all these people here, and we don’t have to spend money like that anymore in order to grow and to sustain that.

Those are the 2021 financials — negative revenue. What’s your revenue this year?

We’re not going to release any new revenue data that we haven’t otherwise. Some things that I can tell you are some of the things we have already shipped. There’s now today more than 35 million active subscriptions on the Substack network. The Substack network drives over 40 percent of all subscriptions and over 15 percent of paid subscriptions, of which there are about 2 million. The top 10 publishers on Substack are collectively making more than $25 million per year.

Would you invest in a company if you couldn’t see their last year’s financial data?

Yes.

Why is that?

For the right company, for a company that is doing something ambitious and… The way to look at a company like this, startups that I have invested in, and Substack with what we’re doing, is not looking at…

Wait. You’ve invested in companies where you haven’t seen their financial information?

Yes.

Okay, because I just have a quote. Ben Thompson’s a friend of mine. He wrote about your retail investment strategy, and the word he used was “shameful” to not give people this information. Our own Liz Lopatto called it “a cynical ploy” to tell people they’re helping writers in the absence of real financial information. Why not give people last year’s financial information before you ask them for money?

We’re a private company. We are doing a regulated community funding process. We’ve given all of the financial information that goes with that process, and we’ve given the information that we think is the important information for telling the story of what we’re doing at Substack. Remember, we’re just giving them the option. We’re giving them the chance to invest and have a piece of it if they choose to, if they think this thing is something they want to be a part of, if it’s something they believe in.

What do you think people should expect their returns to be?

I can’t comment on what people should expect their returns to be. I can’t comment on any of the specific things about the community round because we’re doing it carefully to comply with all of the regulated things. If you want to look at that stuff, you have to go to the wefunder.com/substack page where that stuff is available.

Just to end on the idea of costs, you now run something that you don’t want to be a social network, but it looks like a social network. You will inherit the problems of other social networks: people spamming it; people posting copyrighted stuff that Disney wants you to take down. Weird stuff is going to happen. Those costs will escalate. Are you still doing the sort of Substack Pro deals where you’re fronting the money? Are you doing advances to writers to come onto the Substack platform?

No, not really.

Do you think that program worked? Do you think that’s paying for itself over time?

I think it absolutely worked as a way to kick-start the network and get the company on the trajectory where it is. I think it unquestionably worked.

As your costs go up — just like legal, regulatory, copyright compliance costs go up, where is the next set of subscriptions going to come from if the big top of funnels like Twitter seem to be dwindling or outright blocking you and you’re not doing the kick-starting Substack Pro deals anymore?

I think people have this misconception that Twitter is this giant source of traffic. I think the reality is more like it’s a giant source of mindshare.

I agree with that.

I’m sure that if you look at your stats, and you look at what percentage is coming from Twitter, it’s not actually a huge chunk.

It’s nothing. It could go away tomorrow, and it would mean nothing to us.

I think that’s actually a pretty common experience. The fact of the matter is Substack has a business model that works. We make money when writers make money. We’re growing. We have a path to profitability, we’re default alive. We have control of our destiny as a company. I think we’re in really good shape, and I think that the potential for Substack Notes to accelerate growth for writers only helps all of that.

Do you think the app is kind of the heart of the future of Substack? You’re bringing people more into it with Notes. I will tell you right now — we mentioned Casey Newton — I get a notification about Platformer in the app three minutes on the dot before I get that email every single time. It really does feel like you want to pull people into the app. Is that a part of the growth story here?

I think the app is an important piece of the Substack story. I wouldn’t call it the heart, but I do think it’s important. If part of the Substack vision is this new economic engine for culture, it’s not just an alternative place for writers to come and make money and own their audience and have editorial freedom — all the stuff. It’s also a place where readers can come to be a part of that ecosystem to choose to spend their time and attention in a different way. I think the Substack app and substack.com are an important manifestation of that thing for readers.

Do you foresee a future where that becomes more of a closed ecosystem or more of an open ecosystem? You’re built on email, which is massively open, but you’re saying the app is going to get more important and, right now, it’s fairly closed. Do you think that’s going to continue?

The app has all of your subscriptions, just like your email. This is kind of how the thing works. It’s based off the same subscription network. Actually, I think the great strength of Substack is that we want that subscription network to be open in the ways that it’s open today, which is, “I can bring my audience to Substack, I can leave, I have exit rights. Part of the reason I know that I can trust Substack and that it’s a reliable partner is that I can also leave if I ever don’t feel that way.” We want content published to Substack and have it travel everywhere. We want it to spread on every network. We want it to go into every podcast app. We want it to be on every platform where you as a writer care to be. We’ve designed the business model of the company to incentivize us for that. We wouldn’t want to hinder those things because those are part of what makes the whole thing work.

You’ve given me a lot of time. I want to thank you just for engaging and sticking with it. What’s next for Substack here? What should we be looking for?

We launched Notes a day ago. We have a tremendous amount to learn. Substack, as a whole, this network, is a fundamentally new thing. We’re very much in the process of figuring out how it works. We’re going to keep building it, and we’re going to keep being a reliable partner for writers and building things that help them grow, help them do the work they believe in, help them make a living, sometimes a fortune.

If Twitter never comes back, is that fine for you?

We would be sad about that because it would hurt writers. It would be annoying for writers. A lot of writers use and love Twitter, but it would be fine for us.

All right, Chris, thank you so much for getting into it. I do appreciate it. We’ll have you back sooner than two years next time.

Cheers.

Decoder with Nilay Patel /

A podcast from The Verge about big ideas and other problems.

SUBSCRIBE NOW!