This website is a fun little experiment we built to explore the Clubhouse social graph.
It was made with 💕 by Travis Fischer , Tim Saval , and Tokyo.




We honestly just thought it'd be a fun side project. With the social audio revolution taking off and Clubhouse becoming Silicon Valley's newest unicorn, we were really curious to see what we could learn by analyzing this shiny new social network.

How does it work?

Over the past few months, we’ve been crawling the Clubhouse website and API for publicly available user data via our open source crawler.
The Clubhouse API has pretty strict rate limits, so this was done incrementally using multiple accounts during March/April 2021.
User data that we've collected is restricted to publicly available data, including:
  • Name
  • Username
  • Profile photo
  • Bio
  • Number of followers
  • Number of following
  • List of followers (sample)
  • Invited by profile
  • Invited by club
  • List of following (sample)
  • Linked twitter handle
  • Linked instagram handle
All of this user data is aggregated and stored into a secure neo4j graph database, which allows us to perform some really interesting graph analysis... If you're interested in analyzing this data yourself, please get in touch.
Note that the data powering this site is not related to recent reports of a Clubhouse data breach.

How is this website built?

The core graph visualizations are built using React, d3, and an awesome wrapper library called react-force-graph.
The loading animation is built using Lottie and Framer Motion.
The particle animation is built using react-particle-animation.
The docs are powered by Notion and react-notion-x.
The webapp is built using Next.js and is hosted by Vercel.

How much user data is behind this site?

As of 4/12/2021, we currently have:
  • 2.7M clubhouse users
  • 15M follower relationships (User A follows User B)
  • 1.2M invited by relationships (User A was invited to Clubhouse by User B)

Does this data come from a Clubhouse data breach?

No. There have been some recent reports of a Clubhouse data breach with ~1.3M user's public profile data posted online via data dump(s).
We want to be very clear that the data we're using to power this site did not come from these data dump(s), but rather from our own open source crawler over the past 2 months.
Being able to aggregate this much user data together (whether it's publicly available or not), however, is certainly a security breach of a similar type to the ones recently reported by news outlets.
Here is Clubhouse's official response:

What's next?

No idea. We're really excited to share these visualizations with the community and are just hoping that we don't get sued by Clubhouse... 🤫

How can I get in touch?

If you're interested in access to the social graph data, please use this form.
For all other inquiries, please reach out to Travis or Tim directly.