Usually, a percentage of the tokens is sold to ICO participants and a percentage kept for the company’s needs. The token distribution and allocation of the token is usually a chapter in the future company whitepaper. A pie chart displays how and to whom tokens will be allocated. But how much tokens are allocated (amount) and what are they used for? how much token should I spend for advisor? is 15% of all tokens too much for founder? How many company use reward pool and what is the best size?

I’m trying to answer all these questions at https://ico.tokens-economy.com/distribution/ You can discover how much token are given for pre-sale, main sales, or reserved for particular needs across a bit less than 900 ICO!

After analyzing 896 ICO, up to 24 main categories used to describe token distribution have been identified:

advisors, airdrop, bonus, bounty, burned, community, company, crowdsale, ecommerce, foundation, founder, investors, legal, lockup, marketing, operations, pool, premined, presale, referrals, research, reserves, team

The tedious work was to get the data and map categories (people used a huge amount of synonyms: up to 1936 unique words/sentences, including typos) down to 24 categories!

Some examples:

  • crowd-sale: is, sales, crowd sale, crowd-sale, free sale, ICO round, main-ico, coinsale, coin sale, ICO token, public, …
  • bounty: ICO bounty, bug bounty, gift, bounties
  • frozen: frozen, lock-up, vesting, lockup
  • and the list goes on….

After that, graphing all these values was easy thanks to #google charts API 

I will update the data regularly, so keep visiting this page in the future.

How it was done

  • Data are stored in Google Sheet, 2190 ICO, read from Whitepapers using PDFBox.
  • A category parser read and match token distribution categories (> 1936 unique words/sentences) and their respective values
  • A category reducer reduce the number of categories to a more manageable number by mapping similar category together. E.g. Early Bird investors -> preico
  • A category analyzer can query these data using multiple category selector strategies.

What’s next?

I will improve the category reducer over time to catch more and more synonyms and increase the coverage of ICO taken into account by the category analyzer.

I plan to export the rules used in the category reducer and display them beside each pi chart soon.