I have sent a short newsletter for about four years. It is not a growth engine and it has never been my job. It is the most reliably rewarding thing I do on the internet, and nearly everything I got right came from getting it wrong first.
One reason to open
Every issue has to be openable from the subject line alone. Not clever, specific: the one thing inside, said plainly. "Why our migrations kept failing at two hundred rows" beats "Issue 42: lessons learned" every time, because the second one asks the reader to trust me and the first one tells them what they get.
If I cannot write that subject line, the issue is not finished. It is a collection of things.
Short is not a style choice
Mine run between three hundred and six hundred words. That is not minimalism, it is arithmetic: somebody reading in a queue has about two minutes, and an issue that needs six will be saved for later, which is where email goes to die.
Long writing belongs on the blog. The newsletter is a knock on the door.
Write to one person
Every issue is written to somebody specific, usually a person who asked me something that week. This is the most useful trick I know. Writing to a person produces the register of a message. Writing to an audience produces the register of a press release.
You can hear the difference immediately. "You will probably hit this on the second deploy" is a sentence you cannot write while imagining a crowd.
Give something away every time
Every issue contains one thing the reader can use without me: a snippet, a checklist, a number, the name of a tool. Nothing that ends with an invitation to find out more somewhere else.
The consequence of that rule is trust, and trust is the whole asset. People stay subscribed to something that has never once wasted their time.
The unsubscribe link is the most honest metric on the internet. Respect it and it stays quiet.
Consistency over frequency
Every second Wednesday, for four years. Not weekly, which I could not sustain, and not "whenever I have something", which decays into never.
The cadence matters more than the interval, because it becomes an expectation rather than an interruption. Missing one issue costs almost nothing. Missing four in a row means starting again with people who have forgotten why they signed up.
What I stopped doing
Treating open rates like a grade. Adding a header image. Recapping the previous issue. A monthly links roundup that took two hours to assemble and was read for forty seconds.
What is left is a subject line, a few hundred words, and one useful thing. That is a format I can still write on a bad week, which is the only real test of a publishing habit.
Written by
Ari Bennett
Software engineer and writer
I build web software for a living and write about the parts that are hard to search for: the trade-offs, the dead ends, and the small habits that make a codebase pleasant to come back to a year later. A decade in, I still keep a paper notebook next to the keyboard.
More about the writer
