Simple diligence for utility tokens

EDIT 8/26/2018: This feels a little outdated. Will update some of this thinking at a later date.

In the last year, we’ve seen a huge number of initial coin offerings (ICOs). In these offerings, a team creates and sells tokens that promise some future utility. Many speculators that buy ICO tokens have enjoyed spectacular success, selling their tokens for multiples of the original price as soon as the tokens hit an exchange. Months later, some of these tokens have continued to grow in value while others have stalled or fallen below their ICO price. Beyond speculation, what’s driving this price? Over time, are these tokens worth anything?

I designed a quick and dirty diligence process to evaluate utility token crypto projects. It takes idea risk, execution risk, token necessity, and token dynamics as input and tells me whether there’s any chance the token will be successful.

This process does not attempt to make price predictions, it aims to determine the plausibility of increased usage of the token, and the relationship between increased usage and token price.

Simple diligence framework

simple diligence for utility tokens

The simple diligence framework comprises six yes/no questions and one evaluation of token dynamics. When I use it to determine whether to invest in a project, I require 100% yes answers to the yes/no questions and a positive token dynamic. If you’re more interested in assessing a concept instead of a specific project, you might de-prioritize execution risk.

Is the project strong?

The first set of questions focuses on the strength of the project itself and is split into idea risk and execution risk.

Idea risk is an evaluation of whether the concept is valid. Does it solve a valuable problem and is blockchain / decentralization needed? If the answer is no to either of these questions, there’s a substantial risk that this idea is not valued or is better pursued without blockchain.

Execution risk is whether this particular project is well equipped to be the winner in their area. Is the team good? Is the community engaged? Can they build something that they can protect from competitors? If the answers are no here, the risk is that the idea might be valid but another team is better equipped to execute the idea.

Yes answers to these questions means that usage (U) of the token beyond speculation is plausible.

Is the token designed well?

If the project seems strong, evaluate the token design.

First, is the token even needed? If it’s not needed, there’s a high risk that someone will create a clone without the token and others will prefer it because of reduced friction or lower prices.

How can you tell if a token is needed? A good starting point is to ask yourself: “if someone made a clone of this project and took out the token and replaced it with its parent token (e.g. Ethereum), would it function just as well?” If the answer is yes, the token is not needed.

If the token is needed, we reach the final and most difficult stage: determine the relationship between increased utility and token price.

A popular framework to evaluate token price is the equation of exchange: Price = PQ / V. I won’t cover this relationship in detail, but I’ve attached a list for further reading at the end of this piece. If you follow this logic, the price of a token goes up with the total value created (PQ) and down with velocity (V): how frequently the token changes hands.

This means that the price for a token used singularly for transaction fees should trend to zero. With no incentives to hold the token, velocity rises to dominate the equation of exchange.

So to assess the value of a token as utility increases, we consider the following: given a token with some number of use cases, what is the relationship between price (P) and utility (U) as utility reaches its max. We’ll represesent this relationship (the correlation between P and U) as p(P, U).

  • Positive correlation: p(P, U) > 0

  • Negative correlation: p(P, U) < 0

  • No correlation: p(P, U) = 0

Research in on token usage models is early and I’ll keep adding to this section as new insights arise, but here are some examples:

Token model How is the token used p(P,U) Examples Transaction fee Proprietary payment Negative (trends to zero) ZRX Work token Staked for permission to do work in the network Positive REP, KEEP Burn and/or mint Issued and destroyed based on conditions in the network Positive if net decrease in supply, Negative if net increase in supply MKR, NMR, MANA Discount token Discounts on services in the network Positive (but in this case, you’re better off using the discount token than speculating) BNB Governance Voting on decisions impacting the network Positive ZRX, ANT, MANA

What does this table tell us?

First, there aren’t many projects using tokens strictly for transaction fee. Second, many projects have multiple token models in place.

Let’s take ZRX as an example. Today, the primary use case for ZRX is to pay fees. But the plan is for the decentralied governance use case to dominate p(P,U) function as utility reaches max. Whether this happens remains to be seen. If it succeeds, p(P,U) is positive. If it fails, p(P,U) is negative.

The question I’m left with: given a token using multiple token models, how can you predict which model(s) will dominate p(P,U). Would love to hear others’ thoughts. I suspect we’re just too early to make predictions.


A general framework for diligencing utility tokens can be created out of two simple concepts. Is the project strong and is the token well designed? While this framework won’t generate precise valuations, it is a powerful tool for an initial assessment for any utility token project.

Further reading

Thanks to Lakshman Sankar and Spencer Noon for reading drafts of this post