CABS ICO: Learning From Mistakes

Evgeny Xata
8 min readOct 20, 2017

--

Initial Coin Offerings (ICOs) have taken the financial world by storm in 2017. For several simple reasons they have proven themselves to be very efficient financing tools, which — at least judging by the numbers — would in some ways appear to be superior tools to their predecessors.

Without going into too much detail, the most obvious reason is because ICOs offer the promise of instant liquidity, so investors need not need to worry about having their funds locked up for years. Whereas previously a start-up investor would have to be prepared to wait years for a return on their investment, in the token economy typically no more than a month of patience is required. This results in a vastly expanded pool of potential investors.

But is there such a thing as a “perfect format” for an ICO? For example, can an ICO provide both certainty of participation as well as a clear valuation to all participants? Can it provide them with a smart-contract based token while avoiding the risk of a network overload of the kind experienced in multiple previous ICOs such as with Bancor?

As Ethereum creator Vitalik Buterin pointed out back in June of this year, there probably is no perfect format for ICOs which covers all bases and addresses all the conflicting goals. Nonetheless, that doesn’t mean that we can’t learn from previous experiences to design a better one.

So what is an ICO?

There is some discussion about what exactly constitutes an ICO. Originally it was a one-time thing: the initial (i.e. first) offering of the token for sale, such as the Ethereum ICO, where a fixed number of ether tokens was sold for prices ranging from 2,000 to 1,337 per BTC. This concept has become a little muddied of late, with pre-ICOs, private placements, closed rounds and so on.

Basically, the goal is to provide various avenues to match supply with increasing levels of demand over the course of the initial launch process. So how can this be done in a way which accomplishes that, while providing a good user experience for investors? To explore these questions, let’s first look at the individual parts which have evolved over the past 3 years since the Ethereum offering.

Token issuance

Tokens can be issued via their own native blockchain (e.g. Skycoin), as Ethereum ERC20 tokens, as ERC23 tokens (an improved version of the ERC20 concept), or on the Waves or Metaverse blockchains, to name just a few.

Some of these blockchain platforms require programming skills to create — e.g. Ethereum — whereas others such as Metaverse offer blockchain as a service. Additional functionality such as more complex smart contracts (e.g. those offered by the ETH and ETC blockchains) or cross-token digital identity (e.g. offered by Metaverse), may also be part of the offering.

Pricing and timing

There are essentially three classic approaches to pricing:

  1. Accept investments in multiple currencies but price in only one. So investors may send, for example, BTC, but ultimately the tokens are priced in Ethereum (ETH). BTC amounts invested are converted to ETH at the current market rate.
  2. Accept investments in multiple currencies with different prices for each, for example, by offering 100 tokens for one ETH and 1,300 for one BTC.
  3. Accept investment in one currency only, e.g. Ethereum. This is often done in combination with a smart contract, which in theory is supposed to return the purchased tokens to the original ETH address as ERC20 tokens. I say “in theory” because things don’t always work out as planned.

All of the above may use fixed or variable prices, and may or may not offer volume discounts. The total number of tokens issued may or may not be fixed.

Each phase of the sale of tokens (“closed round”, “pre-ICO”, “ICO”) typically has an established start date and start time.There are different methods to establishing the “end date and time”’. A recent theme is that, once a certain amount of tokens have been sold, the phase is over.

Closed round

A closed round is for insiders and usually comes with a minimum investment cap. This is for those backers that have known about the project before the PR campaigns even begin and isn’t always disclosed to the public.

Pre-ICO

The pre-ICO has come to fulfil two functions. The first is to test the system since the last thing you would want is something to go wrong on the day and never have your project materialise. The testing could be done by members of the team, of course, but the more people involved the better. Secondly, not only does the pre-ICO generate investment funds from those willing to test the water, but it also feeds the hype and fear of missing out, in addition to boosting the project’s visibility.

At the pre-ICO the price is typically fixed. Moreover, those that have researched and followed the project are rewarded for their interest with a lower price than the one offered at the ICO.

Can it be done better?

We would argue that it could, and to demonstrate this, let’s look at an example from a previous ICO.

The Bancor ICO employed the third model mentioned above: all funding was to be in Ethereum only. While Bancor was little known two months prior to its ICO, its clear value proposition, obvious use case and reserve fund concept made it appear to be a “sure winner” throughout the crypto investment community. Initially, Bancor guaranteed that investors would have a minimum of a one hour window to invest. And while there was a general investment cap, it did not apply to the initial hour. All investors wanting to obtain Bancor tokens would be able to get theirs provided they sent their funds within that initial hour.

Unfortunately things didn’t quite work out as planned. Demand was overwhelming and the Ethereum network collapsed within the first 60 seconds of the ICO. Not only could most users not get confirmation of their payments, even worse, they had no way of knowing if they should attempt to resend them or not. Did their payments fail? Or succeed? If they resent them, would they be sending double the amount or not? Some people received their ERC20s fairly quickly, but most did not. All in all, quite frustrating.

In light of the confusion and network failure, the Bancor team opted to extend the minimum duration from one hour to three. While I’m sure many of the frustrated would-be participants appreciated this gesture, it did not resolve the underlying issue, and the Ethereum network did not recover until after the close of the three hour window. The result was that most investors did not get the tokens they bid for.

Be that as it may, Bancor did sell over $100 million USD worth of tokens, considerably exceeding their original target. In fact, the ICO was so successful in monetary terms that Bancor tokens were essentially too plentiful. Going back to Vitalik’s analysis, we can conclude that the ICO’s investors fell victim to the certainty of valuation problem. Due to the lack of an investment cap, they essentially didn’t know what they were bidding on. So, while the Bancor value proposition retains its validity, the return on investment to the initial investors has obviously been much slower to materialise.

Thus, despite having a solid project, Bancor’s ICO offers us some lessons. So how can we do better?

The CABS securitisation token solution

The CABS (CyberTrust Asset-Backed Securitisation) token will be launched in October of this year. Each token will allow its holder to convert one un-securitised BTC (or Bitcoin Cash (BCH) or Ethereum (ETH)) into one securitised BTC (or BCH, or ETH) . Since securitised BTC are currently worth about twice an un-securitised BTC, the implied potential value of one CABS token is approximately 2 BTC (the process of “wrapping” the BTC with CyberTrust’s securitisation platform & process creates a more valuable CABS token). If an investor pays 0.1 BTC for one CABS token, the upside potential is 1.9 BTC. Those interested can look in the white paper for further detail.

So how can an ICO offer both certainty of participation and certainty of valuation? Here’s one approach to accomplishing this:

First, avoid overwhelming the network at the outset of the ICO. How? The answer to this is actually quite obvious: allow potential investors to deposit funds before the onset of the ICO. This ensures certainty of participation.

Second, allow participants to use those funds to bid on a fixed number of available tokens. This facilitates certainty of valuation. Investors do not risk their investments being diluted by an unknown total number of tokens. If the ICO is oversubscribed then each investor will still receive tokens for the project but in a quantity proportional to the oversubscription. When the ICO opens it will remain open for a minimum of 3 hours, so there is no need to rush to place orders within the first few minutes.

How will this work out on the ground? There will be a total of 1 million CABS tokens available for purchase. 160,000 will be made available in four mini closed rounds in the next few weeks with fairly high minimum purchase amounts: 5,000, 1,000, 300 and 50 tokens respectively for each round. In order to make these more equitable, it will be possible to request to participate via the website.

140,000 more tokens (another 14%) will be made available in the pre-ICO round. This will be carried out in more or less the same form as the ICO, with a minimum three hour window, though at a somewhat discounted price. This will be followed by the ICO itself, during which the remaining 700,000 tokens will be sold. Some volume discounts will be available.

Dealing with oversubscription

While enthusiasm is good for project momentum, we also want to target a broad distribution. Not only do we want everyone participating to get a share, but we also want to include would-be investors who don’t hear about the project until after the ICO launch. In other words, to ensure a broad distribution we want to avoid selling ALL of the tokens in the first 3 hours.

To accomplish this, we have set a cap of 400,000 tokens. If participants bid for, say, 800,000 tokens in the initial 3 hours, then each participant will get 50% of the tokens which he or she bids on. In this scenario, two rounds of 150,000 tokens each will be made available in the two following weeks. Unfilled portions of bids will be returned to bidders and may be withdrawn if desired, or used in the following round.

This approach avoids network congestion issues, while minimising the disappointment factor, thereby heightening the investor user experience.

Creating a better user experience

In at least one extreme case the ICO sold out in less than one minute (e.g. the BAT ICO which sold out in 24 seconds). Obviously, this was not an ideal user experience for most participants. But it did make the challenge clear. The ICO phenomenon has changed the world of financing indelibly. The deposit and bid format offers several crucial improvements over previous formats, but as Vitalik noted, no one solution has it all.

Join our community:

Telegram: http://t.me/cybertrustbank
Slack: https://goo.gl/Eoq4Rb
Facebook: https://www.facebook.com/cybertrustbank
Twitter: https://twitter.com/cybertrustbank
GitHub: https://github.com/cybertrustbank
Reddit: https://www.reddit.com/r/CyberTrust/
Steemit: https://steemit.com/@cybertrust

You can also visit https://www.cybertrust.io to get a copy of the white paper and learn more about CyberTrust.

--

--

Evgeny Xata

6th wave crypto economics central banking blockchain, AI, IoT, and digital assets!