The Autobid Contract

Token Distribution 2.0

5 min readDec 4, 2017

Update: The Autobid contract has expired.

It’s time to face reality: the traditional ICO model isn’t perfect.

Even though the standard ICO approach can be very useful for fundraising (which is great because it can help teams bootstrap new projects in crypto), and it can help with distributing projects’ tokens to interested communities of users or investors (allowing these networks to start off on the right foot), there are some unfortunate drawbacks and shortcomings to the approach, too.

Many tokens that were distributed via ICOs end up declining in value after the initial distribution takes place, leaving some users disappointed and upset with the market performance of their purchase. The teams behind the token sales sometimes don’t live up to expectations (in some cases even disappearing after funds are raised, leaving their poor communities “holding the bag”). Even in the cases where the teams do start delivering on the promises they made while hyping their respective projects, users often find themselves frustrated with the overall platform experience, left with the unenviable decision of whether to wait and hope for things to improve, or to find an exchange that allows them to get rid of their tokens and get whatever they can out of them.

There has to be a better way…

After much thought and discussion, we at Delphi Systems have come up with a new token distribution and fundraising model, which we believe to represent a leap forward from the traditional approach, and which we hope to see more projects adopting as the space evolves. We are hoping that this solution (or derivatives thereof) will be used more and more often in the “Second Wave” or “Next Generation” of token-based projects.

The Autobid Contract

We call our solution the Automatic Bidirectional Distribution contract, or Autobid for short. We have a working implementation available on GitHub, but a description of the contract in plain English will probably go a long way in terms of communicating the benefits it offers.

An Autobid contract must specify a predetermined token conversion rate (in our case, the setting will be 10,000 PHI tokens per ETH), a contract expiration time (in our case, 6 months after deployment: July 31, 2018), and an admin account (which receives access to whatever is left in the contract after it expires). After the contract is deployed, it is funded with a particular token balance. The contract remains active until either the expiration date is reached or it runs out of tokens.

While the contract is active, anyone can send ETH into it and it will automatically send them tokens in return (at the exchange rate originally specified). As an example, with an exchange rate of 10000 tokens per ETH, a user who sends the contract 0.1 ETH will immediately receive 1000 tokens from the contract. So far, this is pretty standard functionality for fixed-rate token distribution contracts.

The distinguishing functionality which separates the Autobid contract from other preexisting models is that it is bidirectional, meaning that while it is active, any users who have received tokens already can redeem those tokens for ETH at the exact same exchange rate. This gives users the ability to “try before they buy” or to change their minds if they are disappointed with the overall platform experience after acquiring their initial stake. Other than negligible gas-costs, users can freely and frictionlessly convert back and forth between tokens and ETH without worrying about locking themselves in or committing in ways that they might otherwise come to regret.

The Autobid model shifts the risk away from the users, and onto the team building the project instead. If the team doesn’t deliver what they promised, users can simply redeem their ETH and move on. At the expiration date, the leftover tokens are given back to the administrator/team (so if the team was honestly just a little bit behind schedule, they can simply set up a new Autobid contract and “start round 2” by funding it with those same tokens).


From users’ and participants’ perspectives, the Autobid model is superior to the older ICO models in a number of ways. Participants are guaranteed a fixed price per token, and can purchase their stake at an entirely predictable rate. Even better than that, users who wind up second-guessing their decision to participate can risklessly back out, and convert their tokens back to the original ETH they used to buy with. In other words, the Autobid contract gives users a money-back guarantee, at least for as long as it is active (we recommend launching such contracts with a multiple-month-long duration, to give users ample time to experiment with the platform and token, but it is entirely up to the team deploying the contract what the exact setting is configured to be).

There is an important consideration that must be properly handled: the Autobid contract must either be provided with 100% of the total token supply or the remaining tokens outside of it must be locked up for the duration that it is active (otherwise the controller of these tokens could unfairly claim ETH from the contract that they shouldn’t have access to). But as long as external token sources are taken care of, the contract cannot be gamed, and participants are not at any risk by testing out the tokens firsthand. This provides yet another advantage to the Autobid model in cases where the tokens serve as a signaling or platform-voting tool; while the distribution contract is active, users can acquire tokens and then use them to signal in favor of (or against) different options or routes that the project or team suggests. This opens up a new form of team ⇄ community dialogue, allowing users to meaningfully influence or steer the direction of the project as it gets off the ground (and before the vision has fully solidified). In the case of Delphi, our compound token (with the sPHI signaling component) takes full advantage of this benefit, and we’re excited to let the community vocalize their preferences in a meaningful new way. We look forward to tapping into the wisdom of the crowd and giving users a way to speak up about things they feel strongly about without risking their voices being drowned out by those around them.

It’s about time for the ICO model to get an upgrade. We think that the Autobid model represents exactly that.