Derek Thompson spent 17 years at The Atlantic. He was one of their most recognizable voices, the kind of byline that drives subscriptions. In June 2025, he left for Substack. He stayed on as a contributing writer, but his primary audience (the people who opened his newsletter every morning) followed him out the door.
He's not an outlier. Casey Newton left The Verge. Mehdi Hasan left MSNBC. Jeremy Scahill and Ryan Grim left The Intercept. Taylor Lorenz left The Washington Post. Anne Helen Petersen left BuzzFeed. Every one of them chose a platform with no editors, no fact-checkers, no institutional backing, and a 10% cut of everything they earn.
The question every publisher should be asking isn't "how do we stop this." It's "why does a startup with a simple text editor keep beating us?"
It was never about the money
The assumption in most newsrooms is that writers leave for Substack because the money is better. And for the top 50 (the ones making over a million dollars a year) it is. But there are 17,000 writers earning money on Substack, and the median income is roughly $16,000. That's not a salary. That's a side project.
Writers don't leave newspapers for Substack's economics. They leave for the experience.
On Substack, a writer opens a blank page, writes something, hits publish, and it lands in their readers' inboxes. That's it. No CMS workflow. No editorial queue. No request to resize the hero image. No waiting for someone in production to manually rebuild their piece in an email tool. The gap between finishing a piece and a reader seeing it is measured in seconds.
At a newspaper, that gap is measured in departments.
What Substack actually got right
Substack didn't invent newsletters. It didn't invent subscriptions. It didn't even build particularly sophisticated technology. What it built was a feeling: this is yours.
Your words, your readers, your list. You can leave whenever you want and take everything with you. The interface is simple enough that you never think about it. The analytics are basic but clear. The payment system works without you touching it. There's nothing between you and the person reading your work.
That's the product. Not the features, but the absence of friction.
And it exposed something newspapers never confronted: their own newsletter authoring experience is terrible. Not the journalism. Not the editorial standards. The tools. The workflow. The twelve steps between a writer finishing a piece and that piece arriving in a subscriber's inbox.
But Substack isn't the answer either
Here's what's happening on Substack right now that most newsrooms aren't tracking.
In late 2023, The Atlantic reported that the platform was hosting scores of white-supremacist and explicitly Nazi newsletters, at least 16 with swastika imagery. 247 writers signed an open letter. Substack's co-founder responded: "We don't like Nazis either," and did almost nothing. They removed five newsletters. Those five had about 100 readers combined.
By October 2025, some of Substack's most prominent writers (Anne Helen Petersen, Lyz Lenz, Virginia Sole-Smith) left for Patreon. Not because of money. Because of moderation, email deliverability problems, and the fact that Substack's support team was essentially nonexistent.
Casey Newton, who built Platformer into one of the most respected tech newsletters in the industry, moved to Ghost in January 2024 for the same reasons. He wanted to own his infrastructure, not rent someone else's reputation.
Meanwhile, Substack takes 10% of every subscription dollar, on top of Stripe's fees. For a writer earning $500,000 a year, that's $50,000 going to a company that provides a text editor and a send button. Writers with real audiences are doing the math and leaving.
The platform that promised writers freedom is becoming the thing they wanted freedom from.
The thing newspapers have that Substack never will
Here's what a newspaper offers that no independent newsletter platform can replicate: institutional credibility at scale.
The New York Times's "The Morning" newsletter has over 5.5 million daily readers and has crossed one billion total opens. It converts 42% of newsletter subscribers to paid plans within six months. That's not because of Substack-style growth hacks. It's because the institution is the product. Readers trust it.
The Washington Post grew newsletter traffic to their site by 129% in a single year. The Atlantic hired nine newsletter writers in 2021, giving them base pay plus bonuses tied to subscriber conversions, combining institutional support with individual incentive.
Fact-checking. Legal support. A brand that opens doors. Distribution to millions. An existing subscriber base. Editing that makes the work better, not slower. These are competitive advantages that no solo creator on Substack can match, and that most newspapers are currently squandering on newsletters with broken authoring tools.
The real problem is that newspapers never built a great newsletter experience
Puck News understood this. They built a company around personality-driven newsletters where journalists get equity, revenue-based bonuses, and editorial autonomy within a shared infrastructure. Semafor did something similar. These aren't traditional newspapers; they're new media companies designed from the ground up for newsletter-era journalism.
But legacy publishers? Most of them are still asking their best writers to draft a newsletter in Google Docs, hand it to a production team, and wait for someone to paste it into Mailchimp. The writer doesn't see what the reader will see. They don't control the layout. They don't pick the audience. They don't know how it performed until someone pulls a report three days later.
Compare that to Substack: write, preview, send, see results. In real time. With your name on it.
The gap isn't editorial. It's experiential. Newspapers have better editing, better reporting, better credibility, and a bigger audience. But they have a worse authoring experience than a free tool that launched in 2017.
That's what needs to change.
What "better than Substack" actually looks like
A newspaper that wants to keep its best writers (and attract new ones) needs to offer three things Substack can't.
First, a real authoring workspace inside the institution. Not a Google Doc that gets reformatted downstream. A place where the writer shapes the edition, writes the intro, adds newsletter-exclusive content, previews on mobile, and sees exactly what the reader will see before it goes out. The writer should feel ownership of the newsletter without having to own the infrastructure.
Second, the institutional advantages that make the work better. Editing. Fact-checking. Brand trust. Legal protection. Audience scale. Revenue from sponsorships that a solo creator could never sell. These aren't legacy overhead; they're the reason a newspaper newsletter is more valuable than a Substack with the same content.
Third, transparency and speed. Writers should know how their newsletter performed (opens, clicks, growth) the same day it sends. Not buried in an ESP dashboard. Not aggregated across campaigns. Per edition. In real time. The feedback loop that makes Substack addictive for writers should exist inside the newsroom, too.
Get these three things right, and the value proposition flips. Writers don't have to choose between creative control and institutional support. They get both. And the newspaper keeps its best people, along with the audience those people bring.
The window is open
Substack's content moderation crisis, its 10% fee, and its lack of infrastructure for serious publishers are creating an opening. Writers who went independent are discovering that independence comes with loneliness, burnout, and a platform that doesn't pick up the phone when your emails stop delivering.
But this window won't stay open if newspapers keep offering a newsletter experience that feels like filing a story through a fax machine. The tools have to match the talent. The authoring experience has to match the editorial ambition.
Your best writers aren't leaving because they don't value the institution. They're leaving because the institution hasn't valued their experience as newsletter authors.
Fix that, and you don't just keep your writers. You build the newsletter channel that readers (and advertisers) actually want.
Embrin gives publishers a newsletter authoring workspace that sits on top of their CMS. Writers shape each edition, preview it, and send it to the right audience in about 90 seconds. No copy-paste. No production queue. No reason to leave. Book a demo with your own content.