Why I'm building the opposite of a "relationship manager"

The people who matter most are the ones we never put on a calendar.

You schedule dentists and standups. You don't schedule your closest friend, your sibling, the person who got you through a hard year.

So they drift — not through any fight, just through ordinary busy weeks stacking up until "we should catch up" has been true for eight months.

I went looking for something to help, and everything was a CRM. The whole category treats the people you love like a sales pipeline: "last contacted 47 days ago," streaks to keep alive, reminders that land as guilt, an AI quietly scoring which relationships are "high value." Some literally call your mom a lead. I closed every one feeling worse. Optimizing intimacy is a category error.

Closeness isn't a funnel.

So Weft is the opposite of that. It's a quiet place for the 5–25 people you actually care about — not a network, not a contact book.

You jot the small things you'd forget: her mom's surgery is on the 14th, he just moved, the book she couldn't stop talking about. You give each person a gentle rhythm, and once a day Weft softly shows who's drifted. No feed. No streaks. No AI deciding who deserves your attention.

The decisions I'm most sure about look like bad business:

- On-device. Your notes never leave your phone. No analytics, no SDKs, no cloud model reading them. The privacy label says Data Not Collected, and means it.

- No engagement loops. No streaks to break, no manufactured daily opens. By every growth metric this is the wrong way to build an app. I think it's the only honest way to build this one.

Solo, in Tokyo, native to iOS 26 — one opinionated thing made carefully.

What I'm still figuring out: how to make a deliberately calm, anti-engagement app grow, and how to charge for it without the manipulation I built it to avoid. If you've shipped something that won by doing less, I'd love to hear how.