For about fourteen months in 2023 and 2024 I had a newsletter. It was on Buttondown, $9 a month, simple as it gets. I’d send a roundup of recent posts plus a short intro paragraph, every Friday morning. By the end I had something like 1,400 subscribers. I dropped it in May 2024 and I haven’t looked back.
Every time I mention this in conversation somebody asks why, because the conventional wisdom is that everybody should have a newsletter, especially if you write. So here is the answer in long form.
The people who actually wanted the writing already had RSS
This is the one that surprised me. When I dug into who was on the list and who actually engaged with it, the heaviest readers – the ones who replied, the ones who clicked through to multiple posts – all already had RSS feeds, and most of them had been reading the site through RSS for years. They subscribed to the newsletter because I asked them to, not because they needed it.
For them, the newsletter was just a redundant copy of stuff they’d already read in their feed reader on Tuesday. By Friday it was old news. They told me so when I asked.
The other tier of subscribers, the ones who weren’t already RSS people, mostly didn’t read it. Open rate was around 25%, and click-through was lower. So the newsletter was reaching the people who didn’t need it more than the people who did.
The weekly cadence made the posts worse
This is the bigger problem and I should have seen it coming.
When you commit to a weekly newsletter, you’re implicitly committing to having something to say every week. And I do not have something to say every week. Some weeks I have three things. Some months I have nothing. Pretending otherwise just means publishing weaker posts to fill the slot.
I noticed it about six months in. I’d sit down on a Wednesday with no particular idea, knowing I needed to ship something by Friday for the newsletter to have content, and I’d write a post. Those posts were never my best ones. They were always the kind of generic “five tips for X” filler that any blog can produce on demand, because that’s what’s easy to produce on demand.
The newsletter was selecting for a specific kind of writing. Bland, regular, structured. The opposite of what I actually like writing.
The open rate was depressing
Around 25% open rate, sometimes nudging 30%, sometimes dropping to 20%. By industry standards that’s fine. But here’s the thing: 25% means three quarters of the people who said “yes please send me your writing” are not actually reading your writing.
That’s an objectively miserable feeling. Every Friday you write a thing, you press send, and three out of four people on your list ignore it. The first time I really sat with that number I felt sort of sad about the whole exercise.
Compare that to RSS, where I have no idea how many people read each post and that’s exactly the right amount of information to have. The metric doesn’t exist, so it can’t make me sad.
Composing for email feels different from writing for the web
I didn’t expect this one. Writing an email feels like writing a letter. There’s a salutation in your head, a sign-off, a sense that someone specific is going to open it in their inbox alongside their utility bill and their dentist reminder. So you write more carefully, more conversationally, with a slightly performed warmth.
Writing a blog post feels like writing into a void, in a good way. You’re not addressing anyone. You’re just saying the thing. The voice you end up with is more honest, at least for me. Less “hey friends, hope you’ve had a good week” and more “here’s what I think.”
The newsletter format kept dragging the writing toward warmer, friendlier, less interesting. I’d write a post for the blog, then write a 200-word newsletter intro to wrap it, and the intro would always be the worst part of my week’s writing. Just sappy, smiley, performed-relatable goo. I started dreading Friday mornings.
What I do instead
The replacement isn’t really a replacement. It’s just “how indie blogs worked before newsletters got popular,” which I should never have moved away from.
I blog. When I have something to say. There is no schedule. Some weeks I publish twice. Some months I publish nothing. The site has been doing this since 2002 and the world has not ended.
I expose RSS prominently. There’s a feed link in the header, a feed link in the footer, and a small “subscribe via RSS” callout on the homepage. The feed is a clean Atom feed with full post bodies, not summaries, because if you took the trouble to subscribe you should be able to read the whole thing in your reader.
Occasionally I cross-post a piece that I think is genuinely good to one or two communities. Hacker News if it’s technical. Lobste.rs if it’s deeply technical. Sometimes a relevant subreddit, although I’m picky about that because Reddit is a different vibe. I do not cross-post everything. I cross-post the things I’d want to defend in the comments.
That’s the whole content strategy. It is doing fine.
If you’re going to keep your newsletter
Some people genuinely run newsletters well and I don’t want to talk anyone out of one that’s working. The version that I think actually justifies the work is one of these two:
- The newsletter is the primary product, not a redirect to a blog. Original writing, sent to email, that doesn’t exist anywhere else. Stratechery, Garbage Day, that kind of thing.
- It’s a paid newsletter and the money is meaningful. If subscribers are paying $5 a month, the open rate matters less, and the cadence is paid for.
If your newsletter is just a free notification service for blog posts you’ve already published, kill it. Put more effort into the RSS feed and the homepage. Write better posts. The people who really want your writing will find it.
If you’re new to thinking about this stuff, the older Tips and Tricks for Successful Web Blogging post is a decent starting point.
Buttondown, by the way, is a fine product. The decision wasn’t about the tool. It was about whether I wanted to write for email at all, and the answer turned out to be no.