Hello everyone,

Now that it’s been half a year or so since I “officially” launched Blood in the Machine as a full-time newslettering concern, I thought it time for a status update on the enterprise: How it’s all working, how it stacks up against a 9-5 media job, what the numbers are looking like, and what’s it’s been like living at the whims of platforms, patrons, and algorithms. It wound up being long and detailed and occasionally weirdly personal, so if that’s your thing, buckle up.

First of all, thank you to everyone who’s read, shared, commented on, or otherwise boosted this work. You’ve helped this project grow at a rate that has at times legitimately taken me by surprise. And an extra hearty and machine-rattling thanks to each and every one of you who support this work with your pocketbooks; I really wouldn’t be writing any of this without you. When I was considering what to do next after getting laid off by the LA Times and its tech billionaire owner, I wasn’t sure if going independent was a long-term possibility, and you have demonstrated that it very much is. It’s already been quite a ride.

The tech oligarchs and their AI are taking over. Let's fight back. Brian Merchant · Feb 18 Read full story

As we’ll discuss more a bit further down, this paid subscription/modern patronage model is a weird beast; an unruly assemblage of boons and burdens. One major plus, though, is having concrete knowledge that people care enough about a project to materially support it, and then getting the opportunity to directly engage, discuss, and share that work with them. It’s great. I’d like to find even more ways of expanding this participatory element. For an independent publication focused on the people facing down the maw of the tech oligarchs and their machinery, it’s been a real pleasure interfacing with all of you, whether you’re an artist in West Virginia, an engineer in Silicon Valley, an academic in England, or a farmer in Southeast Asia.

And if you’ve been thinking about becoming a paid supporter, now is a good time, for the reasons I will explicate in greater detail below.

With that, let’s get into it.

The state of the Blood in the Machine

There were really three looming questions that made me most anxious about the prospect of going all-in on a newsletter, the first of which was, predictably, ‘Is this in fact a viable way to generate enough income to pay rent and get my kids the stuff they need for school and such.’ The second was, essentially, ‘Will this work be taken as seriously now that it is detached from an established media outlet?’ The third was along the lines of ‘Do I want to commit myself to a job where I am necessarily, fully, and potentially terminally online and the particular brand of stress that entails?’

And, well, some of these questions have been more definitively answered than others. First things first.

1) The financial situation

With regard to Question #1—is newslettering a viable longterm financial option—so far the answer is an emphatic “maybe.”

One the one hand, after all the overhead is taken into account, I am currently making more than one half of my salary at my last media job. It’s also been uniquely stressful trying to do this journalism, build an email list, and learn how to run a business in uh let’s say uncertain times. (Yes, this is a capital-b Business. I am the proud chief executive officer of Ned Ludd Inc, an S Corp registered in the state of California. I am a small business owner. I am the backbone of America.) As readers of the newsletter already know, I have two young kids, and committing to a *newsletter business* at the crest of an AI bubble, in the midst of tariff mania, and amid the rapid contraction of American democracy, well, let’s just say there’s been a lot of jaw-clenching and teeth-grinding in the offices of Ned Ludd Inc.

On the other hand, it’s just been a little over half a year since I’ve been all-in, and the newsletter has been growing steadily. It’s exciting. For all that uncertainty, as noted above, it’s been extremely rewarding building this community of engaged readers and thinkers, hearing your ideas and messages, corresponding when I can, and seeing so many smart repeat commenters show up to help chew things over. The number of paid readers is ticking up (thank you), especially when I’m better than not at feeding the machine (more on that in a second), and *if* this all works out there’s a future in which the job security is comparable to or even better than all but the best legacy media jobs (all four of them that are left out there).

But it’s still pretty white knuckle, if I’m being honest, and there some anxious lay-awake nights where I’m wondering what the hell I’m doing. I turned down a couple good professional opportunities to focus on BITM and some days—not most, by any stretch, but some—I wonder if that was the right call. My wife has been asking me lately when exactly I can start to chill out. Soon, I say, I hope, soon!

To the stats:

Since the launch announcement in February, BLOOD has tripled in size, and grown to 27,600 subscribers. I’ve published 1-3 times a week since then, barring one week I took off for summer vacation. Conventional Substack Wisdom says that 10% of your subscribers will become paid supporters; for me, it’s closer to 4.5%. My email open rate is consistently between 40-45%, which I am told is pretty good, and even the less-read editions get 20k page views. More popular posts are in the 75k view range. I’m enthused about that. I did not and do not have, say, Mehdi Hasan or Matt Yglesias-numbers of followers on social media to blast announcements to in order to get this thing off the ground, and one thing I’ve learned is that your followers in other arenas—on Twitter or Facebook, say, or readers of your books—won’t necessarily sign up for your newsletter. Based on my experience, one quarter to one-third or so of your social media following will sign up; a little less than I anticipated. From there, you have to get in the trenches and write, week in, week out.

So I’m happy that it’s working, ish, especially because I’ve gotten here by doing what I’ve always done—reporting on and writing critically about Silicon Valley, tech, labor and culture—but it’s tough. There’s been no growth hacking, no cheap rehashes, and, it probably goes without saying, absolutely no AI-generated, AI-brainstormed, or AI-edited content. Yes, I write some TAKES, but they’re informed and well-researched takes, I swear. All told, this has essentially been an extension and/or permutation of my tech column at the LA Times; a blend of reportage, analysis, historical context, and 100% correct opinions.

AI Killed My Job: Tech workers Brian Merchant · Jun 25 Read full story

I’ve also started the AI Killed My Job project, hosted live chats with authors like Karen Hao, published some stellar dystopian fiction, and written a number of Critical AI reports.

I’ve done some of the work that I’m as proud of as anything I’ve written in my career here in this newsletter. I’ve published stories I think few other places would. I broke the story that Duolingo workers had already been pushed out of their company by a CEO who was all-in on AI. I pulled back the curtain on DOGE’s ransacking of federal tech workers and the politically motivated elimination of 18F. I’ve heard from hundreds of workers whose jobs have been impacted by AI, and published dozens of their stories. I’ve sifted through academic research, policy papers, and tech manifestos. I reported on a California therapists’ hunger strike over job automation and degradation. I went to the scene of the torched Waymos in downtown LA to report on an incendiary protest. I broke down Trump’s dystopic AI Action plan, and Silicon Valley’s lobbying assault on AI laws. To name some highlights.

And folks are responding; it’s finding an audience. This is good news. For those of us who are not take kingpins who can delight/infuriate the masses with contrarian analysis of current events and/or routinely instigate feuds with other semi-famous Substackers, it means the old-fashioned reporting out of a good story and putting together of feature packages can differentiate you from the blog hordes here.

2) Newsletters are perfectly respectable amid the ruins of the journalism industry

This is perhaps a good cue to answer the second question I had: whether publishing my work on Substack would prevent it from being taken as seriously, whether it’d travel as far, and whether I’d be able to have as much ‘impact’ with an independent newsletter. This turned out to be by far the easiest of the three to answer. That’s because for the most part there’s been no discernible difference on any of those fronts between publishing independently and publishing with a legacy media outlet.

As I mentioned above, readership and traffic has been great, even better than a lot of places I’ve published before. There were stories I wrote for VICE that crapped out in the triple digits of page views. Three days of writing, hit pub, cross fingers—729 views. Congrats. After all, you’d often get one shot down the feed among the two dozen-plus other stories that might publish in a given day across the website; if the socials didn’t take off and the algorithmic gods were vengeful that day, that was that. The LA Times had a rigid paywall that dramatically limited who could see and share a story online. Blood in the Machine stories, meanwhile, are sent direct to inboxes, (generally) free of onerous paywalls, receive all my love and care, and remain the sole focal point of the website. The average story at BITM now reaches many more readers than my average story at the Times did.

And publishing as an independent newsletter hasn’t put up any real reputational roadblocks, either. Stories I’ve run in the newsletter here have been cited by outlets like the New York Times, the Atlantic, the Guardian, the Financial Times, WIRED, and Futurism, and are frequently picked up by aggregators like TechMeme, Reddit, and Hacker News. I’ve discussed my Blood work on shows like KQED’s Forum, the Majority Report, Decoder, Better Offline, and in a More Perfect Union documentary short. And the newsletter is read by policymakers, legislators, filmmakers, organizers, academics, and activists. That it’s published on Substack, independently, I’ve found, has been no barrier to audiences, editors, or institutions taking it seriously.

Part of this is certainly because I was fortunate to log stints in a number of legacy and digital media outlets before going independent, to publish freelance pieces in fancy magazines, and build something resembling a professional reputation. It still sometimes helps to say “I’m a former columnist at the LA Times” or “I used to work at VICE…yes, yeah, before it went bankrupt” if I’m trying to land an interview or a press pass. But I think that it’s possible to build similar cred independently, on the back of doing good work, that holds up, consistently. It is now widely understood, after all, that American journalism is collapsing in on itself like a dying star, and no one but the haughtiest curators, editors and producers will turn their noses up at quality work published anywhere in the year our lord 2025.

Now this does not mean I think independent media or the rise of the newsletter industrial complex will in any way compensate for the decline of journalism more broadly. It will not and it cannot. Big tech—especially Google and Facebook—have utterly corroded the material conditions for journalism, and the AI companies are now circling like vultures to finish the job. Most crucially, good media institutions and newsrooms nurture young reporters and teach best practices in a way that no number of solo independent pubs can. I’ve had a number of really smart and talented students, junior researchers and freelancers reach out and offer to help out or even to work pro bono but I’m just stretched so thin that I can’t really manage to give anyone the time and attention they’d deserve.

But suffice to say that, for a journalist, especially with a well-honed beat—say, big tech, AI, and labor—newslettering is an option. Substack does deserve credit for normalizing this patronage model that allows me and many others to continue to do our work, on our own terms. It takes some sweat and knowhow (I’ll detail some of what I have found to be best practices below), and I have no idea how viable it would be if you didn’t already have at least some platform, experience, and reputation, but suffice to say that good work is now apt to be regarded on its merits, and I know this gets trotted out every few hours these days but I think it’s true; traditional gatekeepers do matter less than they might have even ten years ago.

And yet. As it goes with everywhere else in the media, you can spend weeks reporting a story you think is all-important, sink days into writing it, and get crickets. And then a dashed-off blog about whatever was in the news will catch the algorithmic current at just the right time and your traffic blows up and so do subscriptions. For a good long while, a blog I wrote in a couple hours about the ‘Ghiblification’ trend was one of my most-read articles; meanwhile, I spent who knows how long doing interviews, collecting on-background information, and researching a piece on the impact of AI in journalism and it found less than half the audience. Same as it ever was!

And I know, because, well,

3) The dashboard is always there

To that end, the last major question I had about doing a newsletter was essentially, “will this incentivize me to be online to an unhealthy degree, both in regards to the combing the dark recesses of the web for stories and in the watching the response to and gauging the success of every single story I write—the responsibility for which now lies with me and me alone—and will that be utterly toxic and debilitating.” So far the answer here is ‘sure but not as much as I’d feared.’

Every veteran of the blog mines of the 2010s has vivid memories of Chartbeat, the web traffic analysis software whose graphical interface would, occasionally, inject you with an adrenaline high by displaying your story as it hit over a thousand concurrent readers and rose to the top of the pack and thus indisputably demonstrate your personal worth via content generation prowess, but would more often show a sad double digit viewership ticking away into the single digits before disappearing altogether after that one last hopeful retweet wore off.

Substack’s dashboard is not quite that bad, but, because the information is no longer relevant just to a dumb blog you wrote for a wildly overvalued digital media property but *your entire livelihood* the pseudo-realtime subscriber data, traffic numbers, and gross annualized revenue projections take on a lot more gravity. It was hard not to refrain from checking it fiendishly, at first. At one point I opened my dashboard on a web browser on my phone this year, went to open a new page, and saw that the last 25 or so pages I had opened in a row were also the Substack dashboard. This was not a great moment. And I would like to be able to tell you that I disabled the setting that prompts Substack to send you an email every time every single person subscribes to your newsletter, but I am a true sicko, and so I have not. I’m fortunate to be getting enough subscribers that I no longer open all those emails but they give me another shot of adrenaline source of feedback to how a story, media hit, or the newsletter more generally, is doing.

Now, look. I wouldn’t say that the relationship I have with all this data is healthy, necessarily, as if such a thing were even possible, but I’ve ultimately found it’s not as bad as it could be. If you’re the kind of person who can log off or tune out when necessary, you’ll probably be fine. I think mostly, I passed a certain threshold where I was able to convince my lizard brain that ‘Okay, this is generally working, stop stressing the minutiae that is out of our control and is now up to the algorithms anyway.’ I’ve acclimated, is another way of putting it.

What I didn’t expect was how all this data would make things feel personal. Every time someone signs up as a paid subscriber—these email notifications I still read—I have a goofy habit that I should probably not air in public but what the hell, this is a blog, and I say ‘thanks Nelson451’ or whoever it is, actually, aloud. It’s a thing I’m still and think always will be thankful for, and this has become some kind of a ritual for honoring it, I guess.

Alternatively, of course, when someone bails and cancels the subscription, you can’t help but take it a personally. This is obviously a preposterous reaction to have. People have plenty of reasons for cancelling a newsletter subscription (and this does not apply to those who take the time to note that they’re cancelling due to financial hardship, I could not empathize more and thank you for your support while you could) but to those who list ‘Time’ or ‘N/A’ or the dreaded ‘CONTENT’ as a reason for leaving, you can’t help but feel like you blew it. Was my writing too boring? Were there too many typos? Too much AI stuff? Not enough AI stuff? (I did have an investment director at the Saudi Arabian Foreign Sovereign Wealth Fund cancel a paid subscription because I criticized OpenAI taking investment from said fund, which he said was racist, but I have to say I don’t feel too bad about that one.)

This will pass in time I’m certain, but it’s an unexpected wrinkle in the Newslettering Life to be sure. All these digitally mediated relationships with patrons who are in a sense quite literally paying your salary—I’m freer than I’ve ever been to write what I want, but you do get all these signals, some clearer than others, that, say, you’ll get more traction if you stick to AI industry criticism, and talk less about the tech oligarchy, or LA’s helicopter surveillance blight, or whatever. It’s real, and it’s strange; you don’t have so much of a direct relationship to a lot of your audience, aside from those who write notes, leave tips, or comment, but a sort of fluctuating composite, and if you wanted to, you could look at the stats and try to serve them accordingly.

I don’t really have any interest in doing that, and so far I haven’t had to. But still, you can see where a writer interested in maximizing new subscriptions and reach could follow the incentives in that data, which in turn would dictate your writing a certain way. Which makes sense! Because regardless of what you do, if you’re writing a newsletter, well, you have to run a business.

How to newsletter for fun and for profit

I never wanted to have to think about the act of writing and reporting as a business, but if you’re running a smallish newsletter and you are not independently wealthy and you are hoping to make it into a sustainable and viable means of paying rent, well, it turns out you have to do exactly that.

This has been the single biggest learning curve of BITM: The Newsletter. Not just having to constantly factor in what might inspire folks like you, reading this story, to sign up and support the project, but what to offer in return, and how to manage it—Business Stuff. You have to scan for problems with Stripe, ensure Substack’s estimates for your gross annual pay are in fact correct, and deal with requests for refunds, accusations from random people’s banks that their cards have been fraudulently charged, and so on. There’s new overhead to manage, like hiring a company to set up the aforementioned S-corp and another to set up payroll, and for paying, when I can, freelancers, editors, and artists.

I’m going to try running some of my first ads in coming weeks, so that will be good, but I have a feeling my ceiling here is likely going to be pretty limiting—there are not a ton of companies I feel comfortable giving ad space to, and there are no doubt many more companies who do not want to advertise on such a critical publication. But for all the reasons above, I can use the ad revenue, and I think it’s an experiment worth trying.

All of this is to say I’m not great at the business side, though I’m figuring it out, and it’s a serious time consumer. I am happy to share what I’ve learned thus far, however. Right now, my business model is basically one part NPR-style aspiring to the better angels of people’s nature, and repeatedly saying ‘hey this requires a ton of work can you pitch in’, and to do so in occasionally creative and occasionally intrusive ways, and another part, yes paywalling stories.

1) You really have to paywall

You just do. In fact, I’m probably going to be doing more of it. Sorry in advance.

I took a quick peek at my backend stats, and at the posts that got more paid signups than others, and only two things are clear: You really only get more signups if you either 1) have an unusually popular hit story that draws lots of readers and thus more potential converts, or, because you can’t bank on that happening all that often, 2) put up a paywall.

Those writers who get 10% of their readers to go paid, I bet they paywall a lot more than I do. I’ve been doing it really infrequently, and usually at the end of a full article above some bonus curation. But every time I put up a real paywall, I get a dozen subscribers at least, sometimes more. And I get it! Paywalling works on me; I just re-subscribed to the New York Review of Books and finally ponied up for the Verge’s paid tier to read paywalled articles. We need the prompting and the prodding, the hand to be forced. I wish I could keep everything outside a paywall, and will always keep reportage, AI Killed My Job, and so on free for all to read—but paywalling is one of the few reliable tools a newsletter writer has to try to raise subscriptions, pending a truly massive readership.

2) Volume matters

You should post a lot, it turns out. Volume is clearly rewarded. If you want to grow, you’ve got to feed the algorithm. There are two pieces to this, and both are straightforward. First, it’s simple business guy math (and remember, I am a small business owner, I am an expert in this stuff): The more you post, the more opportunities you’re creating for readers to discover your work and/or upgrade to a paid subscription. But second, Substack is a platform. In fact, it’s a platform first and foremost. At its most basic foundation it’s an email list, sure, but everything the company is doing—the Notes app, live chats, video and podcast functionality—is designed to encourage creators and users alike to spend more time on a platform. And every digital platform that algorithmically serves content since the dawn of social media rewards volume. The more you post and engage, in other words, the more likely the Substack app is to pick up one of your posts and share it in stranger’s feeds and on Weekly Reads, and for your Notes to get more traction.

It’s not all the platform, either. It’s just as much that you’re writing to an audience of readers who, like everyone, have been conditioned to platform logic, too. Many of us are primed to expect new stories, voluminously, and if we don’t get them, some of us will cancel their subscriptions. This is built right into the platform design: One of just a few reasons a cancelling paid subscriber is offered, along with “Price”, “Time,” and “N/A” is “Low Volume”—ie, “you’re not publishing enough.” Brutal.

The veteran newsletterer Max Read, whose yearly state-of-the-newsletter updates inspired this very post, has written eloquently and at length about how most of us who do this work would rather consider ourselves Writers engaging in a noble craft, not content creators turning out product, and it can be demoralizing to know that you will be materially rewarded for surrendering to the dictates of “posting regularly” but that’s the truth here.

3) Cross-recommending works

One of the biggest things that keeps me on Substack (for now) is its recommendation engine and network effects. If Substack’s stats are to be believed, then I’ve gotten thousands of subscribers from recs alone. It’s nice when big newsletters recommend you—thanks Parker Molloy, Max Read, Taylor Lorenz, and Hamilton Nolan—and each can send hundreds of new readers your way. Even better is if you’re one of just a few pubs recommended by a newer publication when it blows up. I remember when Matt Pearce went on Chris Hayes a while back (to talk about, fittingly, the collapse of journalism) and he had BITM recommended on his Substack—the signups poured in all night. Cheers Matt! (And a shoutout to all of you who arrived here that way.)

The Substack Question

Which brings us to the Substack Question—that is, why am I on Substack at all when there are Nazi publications on the platform?

This piece is already gargantuan, and I’ve been meaning to get into this in a longer post of its own for a while, but for now, suffice to say: I think Substack’s content moderation policy is bad on a number of levels—look also at the mealy-mouthed plagiarism policy that tells users “do not publish” plagiarized content but also does not technically disallow it—and dictated largely by a desire to do as little of it as possible, probably for cost reduction purposes. If you can say ‘we don’t allow porn’ as Substack does, you are already moderating speech. It should be easy enough to say ‘we don’t tolerate hate speech’ and take down Nazi publications, even if that becomes somewhat challenging in practice, as every social media platform that has ever existed has discovered. I do think the extent of the problem is probably overstated; I’ve never encountered any Nazi content in the wild here, via weekly recommendations or on Notes. And the Nazi stuff here pales in influence compared to what’s rampant on X, for instance. But it’s definitely there—you can find some vile stuff without looking too hard.

I struggle with this. Do I follow many other writers and effect a consumer boycott of sorts and move to smaller platforms where reach may be limited but the owners are more ethical? Or do we writers who care about this stuff stay here, and try to push back so that the largest and most influential newsletter platform might be forced to do better while we have some leverage here? I’ve talked to Ghost about migrating, and I’m also thinking a lot about the latter. We need more durable solutions than hopping from one platform to another and hoping management of the next tech platform will meet certain ethical standards. Independent journalists and creators everywhere need power. To that end, there are some interesting ideas percolating; more on them soon. I also have a profile on Ko-Fi for all those who want an alternative way to support this work. And I still have hopes for a project I was involved in to build an alt recommendation engine to replicate Substack’s network effects, though it’s stalled out since the world caught on fire.

The future of Blood in the Machine

So that’s where I’m at. It’s been a thoroughly dark and uniquely destabilizing year, and building this newsletter and community, doing this work, has been both a serious challenge and one of the only bright spots. I’m encouraged but anxious about BITM’s financial viability, enthusiastic about its capacity to move the needle as an independent publication, and acclimating to living at the mercy of the platform. I’m blown away that the newsletter has grown the way it has, and at the thoughtfulness and rigor in your comments and feedback. I would do it all again, in other words, and I’ll take the risk of building this thing over a legacy gig any day. At least for now. Ask me again in a year.

Finally, I’m thrilled that you’re sending me word of your luddite events and AI protest actions, and that after I share them, you’re telling me a lot of folks who turned up say they heard about it on BITM. I’m happy to be here writing, breaking critical AI news, and building a resource for workers and users to learn about big tech’s myriad adverse impacts, to organize, and to resist.

The Luddite Renaissance is in full swing Brian Merchant · Sep 21 Read full story

And I have a solid idea of what I have to do more of going forward, to get this thing to a truly sustainable place and beyond. I have to post more frequently, and will have to paywall at least one story a week, at least until I hit a sustainable level of support. I’m going to experiment with running some ads from select nonprofits and companies. And I have a lot of ideas for what I’d like to do more of coming down the pike. I’ll finish the AI Killed My Job series. I plan on doing more regular live video talks with workers, authors, technologists, and thinkers, and finally port that into a podcast. I want to feature more voices here, and run more guest posts or freelance articles from people with unique experiences and angles on AI, big tech, and labor. It honestly all depends on what kind of resources I have to work with.

What I can accomplish at BITM is ultimately—and here, finally, comes the pitch—up to how many readers wind up chipping in with a paid subscription. There’s room for BITM to grow even faster, even bigger, and to shine a light on Silicon Valley and the tech oligarchy with even more vigor. With enough support I could start regularly paying for freelance contributions from AI-critical artists and writers. I could hire an editor to help kick BITM into the next gear, and beyond that, I can think about building out an actual bloody team to amp up the critical tech, culture, and labor coverage. I currently have more ideas and ledes and interviews and projects waiting in the wings than I can possibly tackle with just me alone.

If any of that sounds like something you’d like to support, and you’re able to, do consider pitching in, or upgrading your subscription. I’m told that for a lot of folks, newsletter subscriptions are a tax write-off, especially if they’re picked up through a work account or academic department. To that end, if you’re at a foundation or philanthropy that supports a mission of tech accountability, BITM could benefit from your backing; there’s much ground to cover. Don’t hesitate to get in touch.

Once again, I’m sincerely thankful to all of you who make this work possible, and who have dedicated some of your hard-earned cash to the mission of Blood in the Machine. Please keep sharing your ideas—what do you want to see more or less of? More original reporting? More roundups of critical tech action? Fewer posts on OpenAI? A podcast, more live chats? Let me know! It’s a weird but very good feeling to know that this work exists thanks to your backing, that it has at least earned a month or a year’s worth of your support. You’re covering my rent, my electricity bills. You’re covering my car payments, and my gas money so I can drive down and interview workers protesting automation on a picket line. You’re helping to pay for my son’s piano lessons, and helping to ensure the next generation of humans can create real music and not just AI-generated pap. You’re enabling me to continue to be a pain in the ass of the tech billionaires, AI boosters, and the Trump administration that loves them. You’re making this critical work possible, at a critical time.

For that, thanks again. And hammers up.