The loss

Revue was a small thing that mattered to me. A clean editor, a simple list, a Tuesday rhythm. Each week was a short letter to the people who cared about the same odd corners of culture and technology that I did. When Twitter bought Revue and then quietly switched it off, the inbox went silent. The list went cold. A habit I had built over a long stretch of evenings just stopped.

The honest thing to say is that it stung. Not because the tool was special, but because the practice was. Writing a weekly letter forces you to notice things during the week. It changes how you read. It turns scrolling into looking. Losing the surface meant losing the practice, and for a while I did not know how to get the practice back without it.

The rebuild decision

The obvious move was to pick another newsletter platform. Substack, Beehiiv, Ghost, whichever was fashionable that month. The less obvious move, and the one I went with, was to ignore the platform question entirely and rebuild the editorial workflow instead. The thing I missed was not the send button. The thing I missed was the loop: notice, save, sort, choose, write, ship.

That is not a product. That is a workflow. And once you see it that way, you stop shopping for tools and start designing the loop you actually want. The question shifts from "which newsletter app should I use" to "what does my Tuesday brain need to hand off to my Friday brain so the letter gets written."

How it works now

Links land in a Slack channel during the week. Whenever something catches my eye, in a tab, on my phone, in a conversation, it goes into the channel. No tagging, no folders, no second-guessing. Just a fast inbox for things I want to look at again.

On Wednesday evening a scoring pass runs over everything in the channel and ranks the candidates. The ranking is not magic. It looks at things like source, recency, signal density, whether the topic has appeared in past issues, and a few editorial weights I keep tuning. It is opinionated software, in the way a good sub-editor is opinionated.

Then comes the part I genuinely look forward to. A small picker interface shows me fifteen ranked candidates. I spend ten minutes choosing. I drag the ones I want into the issue, drop the ones I do not, and write a short take next to each pick. The take is the bit that matters. It is where my voice goes in. The picker is not asking me to write the newsletter; it is asking me what I think.

A build agent then reads my picks and, more importantly, reads my takes, and writes the full newsletter copy in my voice. It uses the takes as the spine. It does not invent opinions; it extends the ones I have already given it. The result lands as a draft I can edit in a few minutes and send. Week ten just went out. The practice is back.

The insight that changed everything

Here is the part that took me a while to admit out loud. There is no developer in this story. There is a content, comms and marketing person who knew, in some detail, what he wanted the loop to feel like. Claude built the infrastructure. I directed the editorial logic. The scoring weights, the ranking heuristics, the prompt that turns ten short takes into a full issue: all of those are editorial decisions dressed up as code.

My non-engineering background used to feel like the ceiling. The thing I could not get past. What this project quietly proved is the opposite. The bottleneck is no longer technical skill. It is knowing what you want clearly enough to ask for it. That is a strategist's job. That is a planner's job. That is, frankly, the job most senior creative and comms people have already been doing, just without a machine on the other side ready to execute.

What it means beyond newsletters

The newsletter is a small case study, but the pattern is not small. Anywhere you have an editorial loop, somebody choosing what goes in and what stays out, somebody deciding what something means, that loop now compounds in a way it never did before. Briefing a brand, curating a feed, sequencing a campaign, shaping a point of view across a quarter: each of those is a taste problem with an execution tail. The execution tail has just got dramatically shorter.

What does not get shorter is the taste. If anything, taste gets more valuable, because the cost of acting on it has collapsed. The people who will benefit most from this shift are not the ones who learn to code. They are the ones who already know what good looks like and finally have a way to ship it at the speed of their judgement.

Close

Revue going dark felt like an ending. Rebuilding the loop turned out to be a beginning. Ten weeks in, the newsletter is back, the practice is back, and the thing I was most worried about losing, my own editorial voice, is the part the machine cannot replace and does not try to. It needs me to tell it what I think. Everything else is plumbing.

Taste is the moat. It always was. The difference now is that taste, on its own, is enough.