You probably can’t make more than $1 million a year on Substack. But Matthew Yglesias does.

Matthew Yglesias co-founded Vox.com, but left four years ago to start his own Substack. It’s turned out be a very profitable decision.

Taylor Lorenz has worked at some of the US’s biggest and most prestigious news outlets: The Atlantic, The New York Times, and, most recently, the Washington Post.

Now, she’s off to work on her own via User Mag, a new Substack publication she’s launching this week.

A few years ago, leaving a Very Established Media Outlet to write your own newsletter might have seemed like career and financial suicide. But now Lorenz, who has carved out a real specialty covering internet culture, has a path to follow: Establish yourself and your brand at a big publication, start your own Substack, and profit.

Like Matthew Yglesias has done.

Four years ago, Yglesias left Vox (where he and I were co-workers) to start Slow Boring, a Substack publication that markets itself as a place to find “pragmatic takes on politics and public policy.” (Sample headline: “Harris is right on the merits about fracking.”) The Substack was an immediate success, and within a few months, Yglesias had 9,800 subscribers, paying him around $80 a year.

I caught up with him this week on my Channels podcast to get his thoughts about next month’s election, the state of polling, and what he thinks of his Vox cofounder Ezra Klein’s recent ascent to kind-of-Democratic king-maker. But I also wanted to know how his business is going.

Very well, it turns out. Yglesias says he now has nearly 18,000 subscribers, which means he is likely grossing at least $1.4 million a year.

He doesn’t get to keep all of that money. Substack takes 10% of his gross and he has other costs, like credit card processing fees. But Yglesias and his family get to keep the vast majority of what Slow Boring brings in. Besides a full-time editorial assistant, Yglesias’ only full-time employees are himself and manager Kate Crawford. Who happens to be his wife.

This most certainly isn’t a path available to the average journalist. But it remains an attractive one for a certain group of high-profile writers who think they have an audience that’s willing to pay them directly — and the ability and interest to run their own small business.

You can hear the entirety of our interview over at Channels; the following excerpts are edited for clarity.

Who is your audience? Who do you think you’re writing for?

It’s a lot of people who are really, really invested in politics and policy — much, much more so than the average person — and who are probably less left-wing than a lot of what you see in the media. But also very animated about Trump and pretty strongly partisan at this point.

Are they professional political people? People who are into this as a hobby?

It’s a mix. I’ve got a good number of political professionals — I check [my subscriber emails] to see who are my House and Senate staffers, who’s in the executive branch, who’s on the campaigns, who are the kind of people who are significant donors or operating in that world.

But it’s mostly regular people. I did a meet-up with subscribers in Los Angeles, and it was a lot of lawyers and entertainment industry people — but people who care a lot about politics.

So I did a thing: I talked to a donor advisor who normally works for very wealthy clients and worked with him on creating a list of key House races to donate money to — winnable races where the Republican is outspending the Democrat. And we got over $200,000 in small donations out of that post, which says to me that there’s a large audience of normal people who just care a lot more about politics than most people. So when you give them that kind of information, they actually go do something.

When you left Vox, what were you expecting out of Slow Boring, financially?

I signed a deal with Substack they were doing at that time to mitigate my downside risk where they guaranteed me $200,000 [in exchange for taking the majority of the revenue Slow Boring generated for a year]. And that very quickly turned into a deal that was unfavorable to me because it was much more successful than that.

You were losing money on the upside.

Which is fine. A completely fair deal. I did it because my downside risk was mitigated. Now, I’m totally on my own. I get the money.

But it just goes to show: This succeeded beyond my expectations.

And I think we’ve seen that with a lot of different people’s Substacks at this point. Obviously, it’s not easy to succeed on this platform. But a bunch of people who have a range of different perspectives are doing very, very well on Substack.

There’s an incredible niche-ification of media writ large. And this is kind of an example of that — that by unbundling, a certain number of us have been able to unlock a lot of economic value. And now some people like Bari Weiss are re-bundling. (Weiss, who started a Substack after leaving The New York Times, is now running a media company that investors reportedly value at more than $100 million.)

What was it like to go from Vox.com, which you co-founded and which eventually built up a sizable staff, to such a small operation? What was most challenging for you to figure out?

I really don’t manage it.

It’s my Substack, so I’m in charge. But really, Kate [Crawford] is in charge, and I’m more comfortable with that. Kate and I differ a lot on questions about things, like how and when we should schedule things. Because that’s not really my forte in life.

The biggest challenge, starting out, was just deciding what the rules of the game were going to be.

I wrote a blog starting when I was a junior in college, to just bang stuff out when I could. And I did that for several years. So I kind of had the idea that I was going to go back and do that again. But I don’t think that really works in this day and age because the space is too crowded. There’s just too much content.

If you just want to see some disorganized thoughts about the world, those are incredibly accessible nowadays. If you want to get somebody to pay for something, you need to deliver a greater level of professionalism.

When I first started doing the Substack, I really did revert to pieces not having proper ledes: “So I was drinking some LaCroix, and I was thinking about how I used to pronounce it, like it’s French and …” and now [I think] “No, this is an article. It has a lede. It needs a good headline. I want people to open the email.”

Are you spending time figuring out what kind of headlines and posts do best? “This policy post did poorly; let’s give them more of this policy post instead”?

We’re trying to understand what grows the list and what converts free people to paid. Retention, unfortunately, is challenging to quantify.

Said every subscription business ever.

Right. This is the problem. At the scale that I’m at, if I could sign a pact with the devil where I never grow, but I’m guaranteed against shrinking, I would sign that. Right?

So the thing I’m most interested in is: What will make people who are on month 11 of their subscription feel like they got their money’s worth? But that’s the one where the analytics give me the least.

Join the club. Every subscription business is bedeviled by churn. Netflix does an amazingly good job at it. It’s one of the reasons Netflix is soaring.

My mother worked at Newsweek when I was a kid. And everything about that business was different from contemporary media — except they cared a lot about subscriber churn and they didn’t really know what was going on.

Is this your last job? Do you think this is what you will do until you stop working?

I hope not. I hope more things come along in the future.

You wouldn’t want to keep this as an ongoing thing?

I’ve only ever done digital political writing in my life, so I expect that I will keep doing that. But I’m waiting for some business genius or editorial visionary to be like, “Matt, like, you’ve got to come work with me and these other people and it’s going to be amazing. And by bundling with more stuff, we’re going to have way more revenue and it’s going to be incredible.” I think about this all the time, but I haven’t seen my way clear to it. Maybe somebody else will.