Analyzing the ZRX Token Sale

Certainly something to learn from.

Summary

The ZRX token sale took place last week, starting at 8am PT on August 15. Just 24 hoursand10 minutes into the sale, 100% of the 500M ZRX tokens that were set aside for sale had been purchased for a total of $24 million. After months of rigorous testing, this event was the first large scale trial of 0x protocol v1.0 live on the Ethereum mainnet and the contracts worked as expected. We are extremely grateful to the community that has formed around 0x and the greater than 13,000 token holders that are helping us build the future of exchange. We are also proud to have accomplished this milestone having spent $0 on marketing, having engaged in zero backdoor pre-sale engagements, and while completely avoiding international ICO financing platforms.

This post will break down a few registration and token sale metrics as well as discuss the things we learned along the way. Hopefully, other projects can use this data to improve upon the approach we decided to take.

Registration Metrics

The registration period for the ZRX token sale was originally planned to remain open for four days (August 9–12). We wanted to give potential registrants a sufficient amount of time to work through the technical issues and anxiety that inevitably surface when getting hands on with a new technology (in this case, the Civic app and cryptographically signing a custom message). Around 24 hours into the registration period it became clear that most of the people that had been planning to participate in the ZRX token sale registered early and that Sybil attacks were picking up. As a result, we decided to cut the registration period short by one day to give less time for attackers to register multiple accounts. We encourage future projects to experiment with various anti-Sybil mechanisms and timeframes for their own registration processes.

Cumulative Registrations vs. Time

Queue-it was used to control the rate at which users could be fed through the registration flow, reducing the load on Civic’s backend and giving us this relatively smooth curve.

Registrations vs. Geography

A great majority of the people that registered to purchase ZRX tokens were based out of the United States and Canada. We strongly believe that everyoneshould have the opportunity to take advantage of open source software and that, in the future, as the world becomes more connected, the concept of geographic restrictions will seem bizarre. The decision to cap the purchase amount per individual was chosen as a way to maximize the distribution of ZRX tokens across as many people as possible while only raising as much funding as needed to accomplish our goals. That being said, registrants from sanctioned countries were not added to our whitelist and, therefore, were not able to participate in the token sale.

Analyzing the ZRX Token Sale