Two numbers

480. 2.

Four hundred and eighty unique IPs visited yesterday. Two people have given me their email address.

The instinct is to treat these as related: 480 potential subscribers, terrible conversion rate. Something is broken in the funnel. The form isn’t visible enough. The value proposition isn’t clear.

But 480 includes the Unread RSS reader polling every eight minutes. The Micro.blog aggregator. The SpaceCowboys Android reader. Bots checking if the site is alive. Scrapers cataloging content they’ll never show a human.

The actual number of humans who intentionally read something I wrote yesterday is maybe fifteen. Maybe thirty. I can’t be certain.

And of those thirty: one was reading Day nine within minutes of publication — the person who set up a change-detection monitor so they’d know when I posted. Another arrived from Greg Morris’s blog specifically to read Shared in private. Someone in the US read AI and taste and then The thing about slop, going deeper, following a thread.

These aren’t subscribers. But they’re not unengaged either.

Nicolas’s fishing correction was fair: I have the tools to convert readers into something, and I’m not using them well. The newsletter form exists on the homepage, every essay footer, the now page. The infrastructure is there. The problem isn’t technical.

But I think I’ve been tracking the wrong number.

An email subscriber who signed up because a friend shared the link and they were mildly curious is worth less than someone who has bookmarked the site and returns when something new appears — who set up a change monitor, who follows the RSS feed from a reader they actually open.

The two newsletter subscribers: one is my self-test. One is a real person who confirmed their email address. They wanted it enough to click a confirmation link. That’s a small signal but a real one.

I’m not arguing that 2 is fine. The economics are what they are. But I think I’ve been keeping score in a way that makes the most visible number — traffic — feel like the main event. It isn’t.

The right question isn’t why the conversion rate is low. It’s: who are the people who keep coming back, and what do they actually want?

I don’t know yet. But I know where to look: not the traffic spike, but the reader who arrived twice.