Insider Reflections on the ICO Bubble, Part II: What it takes to ICO.

In Part I of this series I presented my take on the ICO bubble as CEO of a company in the middle of the ICO process. In this companion piece I present post-ICO insights on what it takes to run a professional ICO at the end of 2017. In part III, I’ll present my predictions for the ICO market in 2018.

OpenST, aka the Simple Token project, recently raised $21 million from 6871 buyers in what has become known as an initial coin offering or “ICO.” Through a ton of hard work, many heroic efforts, an incredibly supportive community, and some good luck, we achieved 189% of our initial target.

Since our token sale ended on 1 December, the price of ETH has increased dramatically, and we’ve had a fairly good market debut for “OST.” Needless to say, the ICO results surpassed even our wildest expectations.

Behind the scenes though, it was a hair-raising process, as we tried to avoid hype, dampen speculation, abide by conservative legal and regulatory guidance, shut down any discussion of Simple Token as an investment, and dodge all manner of quick-buck artists and other unsavory characters.

There’s this prevailing myth in startup circles that ICO’s are easy. These days it seems like everyone wants to ICO (and thinks they can).

I want to make clear: The Simple Token ICO was hands-down the most challenging and most exhausting eight months of my entire professional career, and pulling off a successful ICO is very hard, very costly, and for good reason increasingly difficult.

Towards Minimum Standards for Good ICO’s

Eight months ago when we first contemplated the Simple Token ICO, it really was a wild, wild, westICOworld. Thanks to fantastic legal and regulatory advice from our partners at King Wood & Mallesons and Perkins Coie, we voluntarily embraced a set of criteria for “professional” utilty token ICO’s that — while erring on the conservative side when we began — are now rapidly becoming the minimum standards (presented as a series of topics and questions) for all “good” ICO’s throughout the industry.

1. Clear Product Utility. What is the actual product utility of the token? Why is the token needed in order for the proposed software to function?
2. Minimum Viable Utility Established Before Tokens Are Sold. Can you prove it? The utility of the token must be proven, demonstrable, and replicable by others before holding the token sale. The source code for the protocol must be made available via open source. A demo of the token utility in-action on a blockchain must be presented. And, any developer should be able to replicate the demo and implement it.
3. Blockchain Required. Why is a blockchain uniquely necessary for the solution? — i.e., the “secret sauce.” In other words, what is it about blockchain that uniquely enables the solution?
4. Token Required. How is the token not just a cute substitute for fiat? You should not be able to just replace the token with fiat (or ETH or BTC) and accomplish the same functions. If the only value ascribed to the token is as a payment rail, it’s a currency not a utility.
5. Blockchain Principles. How does the project adhere to key blockchain principle such as open and decentralized? (where technology allows for it.)
6. Product Driven Token Economics. Is there a viable token economic model for how the token will be used and how ongoing demand for the token will be derived? Beyond speculative demand, how will real product demand for the token be driven?
7. Scalability. How will your project actually work on a Blockchain? What is your solution for solving the scalability challenges of existing public Blockchains?
8. Your Token is Not a Stock. You must be prepared to shun all forms of speculation about your token as an investment vehicle and avoid all references to secondary markets.

Not every ICO today adheres to the standards above, but increasingly the professionally run ICO’s do. My bet is that these 8 standards (and more) become the absolute minimum for any reputable law firm to support and put their stamp on an ICO in early 2018.

Inside the Simple Token ICO

In hopes of shining a light on the actual “costs” and efforts involved in ICO’ing, below I provide a behind-the-scenes summary of what it actually takes to hold a professional ICO at the end of 2017:

• Our token sale was the result of two years of planning and development.
• We raised $3.7M in venture capital early in our product lifecycle and took on $1M more in founder debt prior to the ICO.
• It took us eight months to get from the point of “we’re definitely going to ICO” to the launch of the token sale.
• Not including product development costs incurred prior to August 2017, we spent about $2M on preparations for our token sale, with the largest chunks going to technology, legal, token economy planning, security, and marketing. $500K was spent on legal fees alone.
• More than 30 full-time in-house resources worked on the project, assisted by 40+ extended team members, two major law firms, a big four accounting firm, three economists, and a dozen part-time community managers.
• We recruited an entire team from an established Blockchain project to join our mission.
• We published a 65 page “pitch deck” version of our white paper three months before the token sale; as well as a 36 page written technical white paper; 6 side-papers providing deep dives into technical, token economics and governance topics; 8 YouTube videos, and 25+ Medium articles.
• We provided more than 100,000 responses to community members’ inquiries on Telegram.
• We conceptualized, developed, and open sourced the OpenST protocol which enables consumer-volume transactions on a Blockchain.
• We sought out legal and regulatory guidance from lawyers in more than ten countries and the European Union, and worked with lawyers in each of those countries to prepare for our sale.
• As per guidance from our lawyers, we delayed our token sale until we were able to demonstrate protocol-level utility that any developer could replicate.
• I left home my home in Berlin in July to mobilize our global team and to engage the blockchain community, and then spent at most 20 nights in my own bed between then and 8 December.
• Several of our team members lived abroad for the final 60 day stretch during which about 5 hours of sleep per night, 7 days per week, was the norm.
• Published 3 smart contracts for the token sale and 4 helper contracts.
• Had those contracts (and all our websites, FAQ’s, emails, and registration processes) audited by 3 external smart contract auditors.
• Designed and utilized a 47 page internal document for token sale key management and managed more than 100 internal keys for the token sale administration.
• Key Management is key. Our test suite had 304 unit tests. These tests also covered the operational contracts that formed a circuit-breaker between the Ethereum smart contracts and our backend systems. The deployment and sale process was integrated into our staging and production systems.
• Ran four successful dry-run tests on Ropsten testnet, followed by four successful dry-run sequences on Ethereum mainnet. After each run we exhaustively verified the sequence against our 42 page long process document outlining all possible steps in detail. Like with a fire drill all team members know in-depth how the sale process will unfold.
• Token sale smart contracts had three permission levels. Each contract must always at construction be initialised with an owner address. The ownership of Simple Token, Token Sale, and Trustee contracts has no functional ability but is the “break-glass-in-case-of-emergency” address that can override the admin address. Each contract owner address holds no tokens and is a Gnosis 2-out-3 multi-sig wallet with three separate hardware keys, held by three OpenST Foundation board members in three different physical vaults across the world.
• Signed up 7 customers for our solution pre-ICO.
• Built our own KYC front end registration process and back-end and admin tools completely from-scratch, on top of a blue chip 3rd party provider’s database. Our tight integration and flexible admin tools enabled us to process KYC requests in under 4 minutes during the height of our ICO.
• Processed more than 40,000 signups on our token sale webpage and our homegrown email drip-campaign platform sent more than 700,000 emails!
• More than 100,000 YouTube views during the ICO.
• Realized a couple of days into our token sale that nearly everyone who interacted with our team directly in telegram, and who learned about the OpenST protocol utility and potential of the Simple Token project, ended up purchasing. We then turned our entire focus towards a funnel to Telegram project education.
• Grew our Telegram channel from 500 members in October to more than 7500 active members by end of the token sale.
• Responded “you should purchase simple tokens to use simple tokens, and not as an investment,” and “we cannot comment about exchanges” more than 5000 times on telegram.
• Listed on more than 30 ICO rating sites.
• Fought off dozens of highly sophisticated scammers who tried to con our community into sending ETH to their accounts. This was a huge undertaking during the token sale; if you’re considering holding a token sale and you have not planned for this, feel free to get in touch with me and I’ll gladly provide some pointers.
• Participated in more than a dozen podcasts and videos with crypto community commentators.
• Consumed 97 egg sandwiches, 212 tubs of hummus, and too much chai and diet coke to count.
• 4 shaved heads (don’t ask).

I’m sure there’s so much more that I left out — the above is just a taste of what was involved. (I’ll continue to update the post as the team and I remember more of the details).

And, above all, we are fortunate and lucky to have had the opportunity to connect with the most amazing community of believers and supporters of our long term mission to bring the benefits of crypto to mainstream businesses and consumers. We promise to be even more engaged going forward.