Building Friendcatchers — Part 1

Aug 15, 2020

I don't like relying on luck to make close friendships. It takes too much time, energy, and serendipity to find another person who cares about the same topics as me.

The data already exists to make this easy. But it's not organized in a place to make it useful. That's what I'm building now.

Friendcatchers

The name comes from Patrick McKenzie (@patio11), with a small twist in meaning. His mom told him that learning how to cook is a "friendcatcher" because it's a skill that can help him meet people — he could invite people over for dinner [source]. He applies this to a marketing strategy instead, but I'm going to apply it to meeting friends. I hope he and his mom approve.

Imagine a place where you could find other people who watch the same YouTube channels, follow the same Twitter accounts, read the same blogs, and are interested in the same niches that are meaningful to you. What we consume is a better representation of who we are and what we care about than what we can write in a bio.

Existing ways to meet people today are based on a single common interest (Meetup group, Facebook group, subreddit, etc). But the overlap of multiple interests is what creates a strong relationship.

The goal here is to build a tagging system that lets you match with other people based on the specific things you care about. Each tag is a "friendcatcher", as it's an opportunity to discover new people you will probably get along with well.

Watch the build in progress

I started building this a few weeks ago. But it's time to start building in public. I recorded the first part of a series that shows me building the application (and the progress so far). It's not edited and not fast-paced, but it's the process of building a web app.

You can watch it here, or head to YouTube.

Edit in 2022:

I made a simpler version that uses your YouTube subscriptions and/or Twitter follows instead of generic tags/topics. You can sign up for free here: https://friendcatchers.com

Edit in 2023:

Turned it off for now. It functionally worked and I still like the idea, but it needs distribution to succeed, and I don't feel like putting energy into that at this point.