We built Distro to be, as we describe it, the world's easiest-to-use news publishing platform ever created in the history of humanity.
But we had no idea if anyone was actually using it.
We could publish stories. We could take payments. What we couldn’t do was answer the most basic question any publisher asks: is anyone actually reading this?
So we built a dashboard to track the data. Now we know: Barely anybody is actually using it.
It's a shame! Because I've been publishing these incredible columns here on Coding Journal. At least, I think they're really good. And that's what's important, right?
Well it turns out a lot of publishers do actually want to know how many people are reading their pieces, and apparently advertisers to actually care about this sort of thing as well. (Damn you, money thing!)
Designing the specs
Fortunately we have been working with a consultant, Rutvik Choksi, a former Consensys executive and startup founder who specializes in helping companies go from the proverbial zero to one – getting off the ground, finding product market fit, the kind of stuff that actually might get you some metrics.
Smartly, he realized we were going to need some sort of way of measuring our success, or lack thereof. Going in without metrics would be like opening a store with no way to count who walks through the door.
So he created a basic list of metrics that we could (and should) start tracking, as we move toward opening up the platform to more outside users.
Rutvik spent days in our Slack #metrics channel working it through with DistroBot, our in-house AI agent. What would a creator actually want to know? What counts as a “read,” exactly? Between them they turned a vague worry into a real spec, and then into a stack of tickets.
Last weekend I sat down with Claude Code to build it out.
The build-out
We started with the plumbing: a quiet system that records what happens to a story. When a headline gets served to someone, that’s “reach.” When someone opens the article, that’s a “view.” When they tap a share button, that’s a “share.” We wired in the money side too, so a creator can see what each article actually earned.
Then we built the dashboards, so a publisher sees all of it in one place instead of taking our word for it.
Then we got pickier. We added a count of distinct readers, so a hundred clicks from one person don’t look like a hundred people. And last, average read time: not how long an article is, but how long humans actually stay on it.
When it was all running, I pulled the real numbers from our first full day.
You can't unsee this
In 24 hours: 129 page views, from about 16 distinct readers, and zero shares. And the “reach”? Almost all of it was machines, not people – AI agents and crawlers pulling our headlines off our feeds.
For a platform built around the agentic press, that’s partly the point. But machines aren’t the audience an advertiser pays for.
Sixteen readers. Uhh, whuhh??? Turn that around, stack it against the population of the world, and think about how many people were crucially NOT reading my amazing Coding Journal columns.
Well, I kind of already sort of knew it of course, but now I need a good cry.
The solace is that we do have to start somewhere, and now we know exactly where “somewhere” is.
We can finally see whether anything we do moves the line. And when we sit across from a creator or a sponsor, we can tell them the truth about our audience instead of guessing at it.
And if one of my riveting Coding Journal columns manages to go viral, we'll see that too. 🤞
(HOW AI WAS USED IN THIS STORY: I asked Claude Code to write the first draft of this article – including the diagrams – by relying on the transcript of my coding session inside Visual Studio (Claude CLI) and a custom skill I developed to replicate my own writing style. I then used Distro Publisher, our official MCP server, to file the story as a draft directly from Claude CLI to my newslode. I then edited the draft in DistroVerse’s editing interface and checked all the facts before publishing. I used ChatGPT to create a story illustration.)