The product started with a failure
In my May 18, 2021 First 1000 issue, I followed Jason Citron from Aurora Feint to OpenFeint to Discord. The useful part is the distribution loop: make the conversation easy to join, then meet the first users in the place they already gather.
Aurora Feint was one of roughly 50 games available when the iPhone App Store launched. It got good reviews, but the paid multiplayer version did not work. The source says the company made about $30,000 from the game, and Jason had roughly three weeks of runway left.
The team spun the game’s social features into OpenFeint, an “Xbox Live for mobile” idea, and used a landing page, fake screenshots and a TechCrunch story to raise a bridge round before the full product was built.
Then came Fates Forever. The game itself was not working, but its chat experience exposed a more interesting problem. Voice tools like Teamspeak required downloads, payment and shared IP addresses. Skype interrupted gameplay. The answer became Discord: a low-friction, audio-first tool that people could join through a link.
Distribution was built into the product
WebRTC let people join a conversation from a browser without downloading another app or sharing an IP address. The link was not just an onboarding shortcut. It was the thing users could send to teammates.
The first 20 customers were friends, and friends were not enough. For five months the founders kept adding features their friends requested, but usage did not follow. Then they widened the test.
A friend posted Discord in the Final Fantasy 14 subreddit shortly before a new expansion. Jason and Stan joined the voice chat themselves. Redditors could talk to the founders, go back to the community, and tell other players what happened. The source describes this as the first 200–300 customers.
For the next 1,000, the team repeated the pattern across three games with upcoming expansions. They built the integrations those communities needed and then returned to the relevant subreddits and guilds. Stan’s previous company, Guildworks, had more than 1,000 gaming-community sites hosted on it, which gave the team credibility with the exact groups they wanted to reach.
The channel decision
The Discord story gives me a simple test for a crowded product category:
- What is the smallest useful action a user can take through a link? Remove the download, account, or setup step that blocks the conversation.
- Which event makes the problem urgent? The team used game expansions because attention was already concentrated there.
- Can the founders be present where the first users gather? The live voice chat turned early acquisition into product research.
- Which narrow communities should receive product work first? Three games were more useful than “all gamers.”
- Does the community have a reason to forward the product? A voice-room link did more work than a generic launch announcement.
The takeaway
The first users were not a funnel diagram. They were people talking to each other in a place the founders could enter.
Sources and original research
Adapted from my original Discord First 1000 issue, published May 18, 2021. The historical launch, revenue, runway, community, subreddit, game and Guildworks details come from that issue; the channel checklist is my adaptation. No current Discord metrics or new community-growth study is claimed.
Updated September 20, 2026. Based on original First 1000 reporting and the sources listed above.