There are some wonky audio/video sync issues in the recording of the live stream. Not sure what happened there — something I need to diagnose before I do another one of these. Regardless, you can keep reading for the essay version.
I’m reading Cory Doctorow’s book Enshittification: Why Everything Suddenly Got Worse And What To Do About It. I’m only a little way in, but it has got me thinking about the online platforms that I still use. Specifically, Substack, which powers this newsletter.
We’ve all been burned before, as platforms we’ve poured hours of time and effort into have shifted and disintegrated around us. Enshittification, indeed. If you’re not familiar with the term, here’s how it works:
First, platforms are good to their users
Then they abuse their users to make things better for their business customers
Next, they abuse those customers to claw back all the value for themselves
Finally, they have become a giant pile of shit
‘Enshittification’ as a term has taken root over the last couple of years because it helps to make sense of the world. As soon as you first hear it, the state of the internet comes into sudden focus. It’s not your imagination. It’s happening, everywhere, all at once. Despite the cheekiness of the word itself, enshittification is a precise and accurate diagnosis. It is immediately, obviously, true.
What about Substack, then? I’ve been writing this newsletter since 2021 and have committed an insane amount of time to it. I’ve loved every moment, don’t get me wrong, but that level of commitment makes me nervous: what if Substack morphs into something that no longer makes sense for me?
Spoiler: this isn’t a Substack hit-piece. If you want that, you’ll need to look elsewhere.
If you want to know more about this from the coiner of the word, Brian Merchant has a great interview with Cory over on Blood in the Machine:
Substack’s 3 core features
There are three core feature to the Substack platform. These are the reasons I decided to use the toolset in the first place, and they’re also why I still use Substack. They’re not just useful features, though: they are protections.
We own our subscriber lists
We can export our data at any time
The core of the toolset is writing and sending email newsletters
In the book, Cory talks a lot about ‘switching costs’. By raising the perceived cost of switching, platforms lock-in their users. It’s difficult to leave Facebook because all of your friends are on Facebook, and all of their friends are on Facebook. If you leave, it means losing contact with your friends, unless you can somehow convince them to all leave at the same time.
Substack, by design, chose to minimise the switching cost. This is, to be clear, extremely weird. It’s counter-intuitive and entirely opposite to 20 years of Silicon Valley best practice. Investors must hate it.
When I moved from Wattpad to Substack in 2021, a significant factor in my decision was Wattpad’s general lack of transparency. Authors there don’t own their list, have limited opportunities to communicate with readers, and are entirely at the mercy of algorithms.
Short version is: I wouldn’t advise anyone who takes their creative practice seriously to use a platform in 2026 where you don’t own your own list and content. More here:
As long as those three core features remain intact, I think we’re in a good place. Ironically, the ability to leave is a major factor in why I’m still here, using Substack.
But there’s a catch
There’s always a catch. The switching cost back when Substack was a nicely designed, simple email newsletter service was very low. That isn’t Substack in 2026, though. We now have the ability to do all of this:
Publish videos
Host live streams (with guests)
Publish and distribute podcasts
Engage in short form interactions over on Notes
Run a live chat with subscribers
Take advantage of Recommendations and other network growth effects
Distribute our work through the Substack app, as well as via email, and automatically post relevant content1 to YouTube, LinkedIn and podcast libraries
Sure, I can still take my subscriber list and switch to Ghost, Beehiiv or Buttondown. But none of them offer the same overall toolkit, and the more we lean in to these additional features — the ones that are further away from the core ‘email newsletter’ function — the more locked-in we become. The higher the switching cost.
All the features I just mentioned can be replicated across a mixture of other platforms and services, but that immediately invokes a time and admin cost. The beauty of the Substack toolkit is that I can be very productive without drowning in boring admin tasks. I’m not a full-time writer, and the efficiency of having this all under one roof shouldn’t be underestimated.
We can export our data, and move to another platform, sure, but I’ve lost count of the number of writers who have departed Substack in a huff, only to return six or twelve months later. It’s harsh out there. For all its troubles, Substack does seem to work when it comes to the basics of ‘getting people to read your stuff’. Even if the switching cost is theoretically low, the reality is that moving platform can kill your newsletter; that’s less on Substack, and more on the other platforms to catch up and provide viable alternatives.
The Substack app: life raft or prison ship?
When the Substack app appeared I was deeply suspicious. It seemed to go counter to the idea of owning your list, and of having a reliable decentralised delivery method for subscribers (email).
Email is a protocol, and is not owned by a large tech company. Email clients are proprietary, such as Gmail or Outlook, but the underlying technology is universal and reliable.
Email has been the single most reliable form of internet communication for about 40 years. While usenet, blogs, forums and social media have come and gone, email has remained consistent and reliable. That’s why owning your list is so vital, if you’re an author (or anyone else making stuff online).
The tech bros have noticed, and they don’t like it. Email is mostly non-algorithmic. It is user-controlled. There is less intermediation. Someone sends an email, someone receives the email. That’s it. Simple, reliable. It’s anathema to the US tech giants, and that means email is a target, and is vulnerable.
Which brings us to the Substack app, but not necessarily in the way you might think.
By moving readers into the Substack app, we’re slowly conditioning them towards using the app for their long-form reading, rather than email.
The implications for the switching cost should be obvious: if all your readers expect to read your work via the app, it’s going to be pretty painful if you switch to using something like Ghost. Readers won’t necessarily think to check their email for your newsletter. It breaks the habit.
While the theoretical switching cost for leaving Substack is low, the app makes it much higher practically. It traps readers and writers. This is a bad thing.
Except, it’s not quite that simple, because email itself is changing.
Apple have been clamping down on email privacy for years (which is a good thing), making it difficult to know if your subscribers are even receiving your emails (which is an annoying thing). Google are clearly working on incorporating Gemini AI into Gmail, offering LLM summaries of your inbox. The direction of AI travel with Microsoft and Outlook/Copilot is clear. The more that the leading email clients obfuscate our inboxes, the harder it is for individual newsletters to be seen, even by the readers who want to see them.
Which brings us back to, of course: enshittification.
There are solutions, such as switching to an email client that isn’t run by a horrible tech megacorp. Perhaps something like Proton. But most people aren’t going to go through the hassle. Again: switching costs.
Anyway, my point is that in this context, in this possible future, where the big email clients become antagonistic towards their users, the Substack app becomes something of a life raft. As the bottom falls out of email communications, the app could be the only thing keeping us directly connected to our subscribers.
The problem is that it’s a life raft we can never leave, condemned to float endlessly in the storm.
The economic model to the rescue(?)
“Substack only makes money when we do.” That’s the company line, and it’s a pretty good one.
“We want you to grow. We want you to reach as many people as possible, influence as many people as possible, and make as much money as possible. And we’ve built our business model around that—we only make money when you make money. You have the right and ability to leave any time you want.” Hamish McKenzie, over here
It’s a brilliant guard against the usual perverse incentives online platforms fall prey to (or, you know, actively embrace). Meta and Twitter’s any anti-user decisions were driven by the need to please advertisers. See stage 2 of enshitification, above.
I do wonder how much of it is an illusion, or a pretty marketing line. If Substack was entirely solvent through subscription revenue, we’d all be in a good place. That isn’t the case, and a big chunk of funding comes from external investment. Necessary, no doubt, especially given all the new features they’ve been adding, but each dollar of external funding undermines that central concept of “we only make money when you do”, which in turn risks a breakdown of the core motivation and design principles.
Not being an economist or a business person, this is all well outside my area of knowledge. But it’s a shame we live in a society that requires companies and individuals to repeatedly compromise themselves in order to achieve their aims.
An apocalypse of new features
Substack has changed drastically in the last four years. It’s a very different place today than it was when I showed up in 2021. We now have:
Substack Notes, their Twitter-like short form social thing
Podcasts
Video
Livestreams
Chat
It should be noted that each one of these was absolutely, positively the herald of the End Times. The announcement of a new feature would cause a collective teeth gnashing of the sort that could grind entire planets to dust. Much of that fear was caused by the very enshittification we’ve been talking about: after a decade of decline across Facebook, Instagram, Twitter, Tumblr, Google Search, Spotify, Netflix, YouTube, [insert your former favourite thing here], is it any wonder that we have such reactionary responses to even the slightest change on a platform we still enjoy using?
Going somewhat against the tide, I’ve not been particularly worried by any of the changes, because each new feature has been incorporated elegantly and is easily ignored. None have interfered with those core three features I mentioned up at the top of the post.
One of the more recent additions did cause my lip to curl, though:
That lightbulb icon is a direct integration of a thing called Polymarket. For some more context, here’s Substack’s slightly weird announcement:
Polymarket, as I understand it, enables people to bet on the likelihood of real world events. In other words, it’s a monetised and overly complex way to find out what most people think is going on. Quite why this requires an entire gambling framework I’ve no idea.2
The addition of a specific Polymarket icon alongside and of equal importance to adding a video or image is intrusive and unusually wonky design from the Substack team.3
It is an especially silly situation, given that betting on Polymarket is illegal where I live:
It’s the equivalent of putting a betting shop in the foyer of your public library. Strange design choice, awful moral choice. I thought this was an interesting post on it:
But, as distasteful as I find it, it doesn’t impinge on those three core features. A somewhat worrying direction of travel, though.
Bad actors
There are deeply nasty people on Substack, because Substack is unfortunately populated by humans. There isn’t space to get into it here, so instead I’ll point to an earlier post I made on the subject of community, censorship and moderation:
The TL;DR is: there are grim people here, just as there are grim people in the city where I live. Grim people share the same pubs and cafes and supermarkets as me. My country has a lot of grim people living in it. There are also a ton of brilliant people; far more of them, in fact. I’m not willing to retreat and cede every space until there is nowhere left to go and all that is left is grimness.
Conclusion: I’ve no fecking idea
As you can probably tell by now, it’s a tricky, complicated thing. Has Substack become enshittified? Personally, I don’t think so. As a writer, it is still an excellent toolset that helps me communicate directly with readers. As a reader myself, I can say without exaggeration that I discover new writers and fascinating writing every single week. No other platform has managed to do that.4
When talking with the people who founded and work at Substack, I sense that they are well-meaning. They have good intentions, despite various stumbles. In the stormy sea that is the internet in 2026, Substack still feels like a lush island — even if the waves are crashing on the shore and sharks are circling in the shallows.
The short version is that being a creator of any sort in the 2020s is a massive pain in the arse. The ground is continually shifting beneath our feet. Perhaps that’s how it’s always been, even pre-internet?
The critical thing we can all do is to stay vigilant: we have to forge the online spaces that we want. Set an example through our behaviour and interactions. When the billionaires try to reshape the world according to their whims, we have to push back, deny them our time and money.
And, of course, make sure you have an escape plan. Don’t commit to any platform that doesn’t let you take your audience with you, if you decide to leave. If you’re still on platforms that hide your audience from you, which have high switching costs, now is the time to consider leaving. Avoid the sunk cost fallacy.
Meanwhile.
Thanks for reading. That was a lot, it turns out.
While we’re doing some Substack navel-gazing, don’t miss this superb post from Scoot:
It’s one of the most clear-eyed examinations of how this newsletter business works, neatly sidestepping all the usual made-up guruslop. It’s not actually specific to Substack, either. It’s good content strategy in general.
Sometimes I’m really enjoying an essay and then it ends, even though there’s still so much to talk about. One such example is this:
So much more to dig into there, from both the casting end and the billionaire end. Good stuff, though.
Yesterday I read issue 16 of Kieron Gillen’s The Power Fantasy, which was a quite astonishing finale.5 I wish I had a tenth of Gillen’s writing skill. He manages to mess about with formal structures and be highly indulgent, pretentious and self-referential, but without ever becoming annoying. I don’t know how he does it. I read his stuff and feel slightly more clever by the end; and not a little bit jealous.
It’s superheroes on the surface, but don’t let that put you off. You’ll find instead a meditation on power, free will, ethics, war and nuclear power.
By definition, ‘classics’ tend to come from the past, and be anointed retrospectively. “I wonder what it was like to read this at the time?” Reading Gillen’s work answers that question for me.
Elsewhere, still reading The Hobbit with the boy. It’s my first time, and I’ve been surprised at how early the trolls and Gollum show up. Those are the scenes everyone knows about, and I rather expected them to arrive later in the story. Obviously, doing Gollum’s voice was a particular delight.
Everything else I’ve been reading has been about the state of the world, so the less said about that, the better.
Have good weeks, all. Thanks again.
As a general rule, I loathe the overuse of the word ‘content’. It’s the ultimate homogeniser and should be avoided. In this specific context, it seemed appropriate.
I’ve never been especially interested in what most people think. That tends to be obvious and a bit dull. Give me the weird outlier opinions, the unexpected, the niche, please.
If we could customise which icons appeared in the Note editor, I’d have less of a problem. That way people could add/remove icons based on their relevance. A quick way to find and embed YouTube videos, for example, or Wikipedia entries. One of the options could be Polymarket, and I wouldn’t object. Forcing it in there for everyone is daft.
There was a brief period in its very earliest days when Twitter was always bringing me something new and stimulating. Veeeery brief.
Sort of finale.
