Why designing NFT sales are hard and key considerations
One of the most complex decisions that a blockchain game needs to make is how to design its NFT sales.
It’s challenging because it involves trade-offs and interconnected design decisions.
This article dives into the psychology of the buyer, how the ideal sale mechanics change over time and various factors to consider when deciding how to sell NFTs.
Developers trying to design an NFT sale:
Understanding Buyer Psychology
First off, let’s get into the mind of the buyer. Who are they and why are they purchasing?
There are broadly going to be three types of buyers of NFT sales. Buyers can belong in more than one category.
Genuine community members are those that align with your mission, project and will likely end up being players or users. They most likely follow all your socials and also engage on Discord. They likely form the core ‘community’ and engage with others who share the same interests. They have probably read your whitepaper and have asked questions to make sure they understand how it works. They typically don’t have too much money to spend (although can overlap with the other groups) and usually are only in a few projects at once. This group is going to care a lot about what you do and are interested in ‘keeping’ or ‘using’ the NFTs that they buy.
Passive, long-term investors are individuals who buy under-valued NFTs with the intention of making a profit long-term once a project is successful. They are there to find good, long-term projects to invest into, and likely will apply some sort of framework and research into their investments. Once they buy and decide your project is "the one”, they are unlikely to change their position. They will be looking carefully at how the NFTs are designed in terms of pricing & supply, as well as the future utility that they will grant. They typically engage less (being time poor), but will be watching the project carefully for a long time.
Flippers & botters are in it for the short-term money. They probably look at over 10 projects at a time and will buy / sell within a week, or sometimes within hours of a sale occurring. They do not believe in holding long-term, and believe there is more profit to be made by constantly buying low and selling high. They typically are active & engaged, as knowledge is what gives them their trading edge (also known as ‘alpha’). Many of them are going to be very technically savvy and are able to bypass your website by minting directly from the contract, using bots, or even doing things like hiring Virtual Assistants to get on whitelists.
Most projects would likely want to appeal to the first two categories, and likely that is also what their community would prefer.
But this is where we face the first dilemma.
If your project is appealing to your community members and long-term investors, it’s also going to appeal to flippers & botters. Because it means there will be demand for their purchases.
What will you do? Let’s dive into some considerations:
Have a plan to stop flippers & botters
Assuming that you care about whether flippers and botters purchase your NFTs, you should create a plan to combat this. The way to do this requires the ability to differentiate between genuine supporters and those looking to make a quick profit or ‘bots’ / ‘multi’ accounts.
The most popular solution to this right now is ‘whitelisting’, a process where only certain users who either did something previously (i.e. joined Discord before a certain date) or are about to do something (i.e. refer 10 friends to Discord) are allowed to mint NFTs. Sometimes the process is manual (i.e. hand picking contributors in Discord), and sometimes it is automatic (i.e. enter a raffle to get whitelisted).
The assumption for ‘whitelisting’ is that only genuine community members would do these actions. However, this is not always a correct assumption, and in fact flippers & botters are experts at finding ‘alpha’ to get around the system. Some examples include:
Creating multiple Discord accounts to enter any raffles
Paying a Virtual Assistant from the Philippines to post memes in your Discord everyday so they can get a whitelist spot.
Purchasing ‘verified’ Twitter accounts and trying to become an influencer for your project
Creating bots that automatically purchase NFTs as soon as the smart contract turns ‘on’
It’s going to be easier said than done to differentiate between genuine supporters and flippers, in particular if you want to do this in an automated and scalable manner.
Luckily there may be other solutions: doing things differently to the current ‘meta’ and using information asymmetry.
Figure out the current ‘meta’
There is a great post on this by Cobie which I would recommend which goes into detail into the idea of a ‘meta’.
It’s written for crypto, but the general idea is that over time, NFT projects start falling into a pattern based on what has worked previously. All it takes is one really successful project to ‘innovate’, and then others will copy to try to replicate the success. For example, Bored Ape Yacht Club was the first project to popularise a “fair” sale where every NFT had the same price, and now we see that as the standard. Later on, projects like Artblocks & Meebits popularised dutch auctions which sold out at the top of the auction price, and projects tried to replicate that.
If you want to design a great sale, you must understand the meta. Doing a dutch auction sale right after a series of ‘scam’ projects release dutch auctions could give your community an incorrect bias that your sale format is also ‘scammy’. Using a raffle system for your community might work well for the first project that does it, but not if everyone starts building bots to enter these raffles. Even using KYC could end up being exploited, something we saw with the Yuga Labs sale where thousands of KYC accounts were allegedly traded on the secondary market.
Learn the meta.
This is why Guild of Guardians was one of the first projects at the time to do both a stealth drop and whitelisting for retroactive behaviour. We wanted to get NFTs into the hands of as many of our community members as possible while minimising the number of pure flippers.
Game theory & asymmetric information
This is an area which is less explored (probably because it requires more complicated engineering work) but is one where you turn the sale itself into a game.
It can be great for the actual sale. Pak does a great job of this as part of his ‘performance art’ persona, where in almost all of his past NFT sales the buyer does not have complete information. He reveals and changes the rules as the sale continues, meaning players then have the choice of predicting and applying ‘strategy’ to their purchases.
Having asymmetric information in a sale process can also be a great deterrent to both botters and flippers, because they do not have time to prepare. For example, deciding to whitelist community members based on a retrospective, unannounced formula could be effective in capturing genuine supporters.
It’s all a balancing act
At the end of the day everything is a trade-off and a balancing act. Deciding to do a retrospective, unannounced whitelisting system for example might be good to eliminate ‘fake’ community members, but could result in genuine community members also missing out.
Some of the key trade-offs that games in particular need to make are:
Game economies: Balancing the sale of NFTs with their actual supply and utility in the game. Selling more NFTs might mean more revenue, but could come at the expense of the economic sustainability of the game
Engineering execution: Creating complex and unique sale mechanisms might have a better outcome, but take more time to develop and test. In addition, deciding how to deal with scaling, high server volumes, DDOS attacks & gas fees can be a big challenge for many projects. NFT sales are notorious for having launch issues, and there is a trade-off
Segment preferences: The sale mechanics themselves will result in different types of participation (i.e. some sales might preference wider distribution, others might enable really big investors to purchase a large amount). Deciding which mechanics to go with (and accepting that not everyone will end up being happy) is the harsh reality of most NFT sales. Even something as basic as the time of launch will end up giving advantages to users in certain geographies
Rolling it all up in practice
Ultimately all of the above are just considerations. They don’t actually tell you at all how to design a sale, just some of the things to think about.
A simple approach for applying these considerations would be to look at the goal, identify the mechanisms available to you, and then deciding on the sale mechanics afterwards.
Simple example framework below:
1/ Goals
Each NFT sale is designed to achieve certain objectives, and projects need to decide which one they are optimising for. Some examples include:
Maximising revenue vs leaving money on the table
Minimising ‘gas’ wars and thus fees paid by buyers
Maximising distribution to get NFTs in the hands of as many different buyers as possible
2/ Mechanisms
Different frameworks exist to define the choices in an NFT sale, but broadly speaking some of the most important decisions for a project will be:
Pricing: Fixed pricing? Variable? Auction (English or Dutch)?
Supply: Fixed? Variable based on another factor?
Information: Fully transparent? Asymmetric (i.e. new ‘info’ revealed each day)?
Timing: How long does the sale last? When does it start (hint: which timezone will you give an advantage to)?
Other: Is the reveal instant or delayed? How does randomness work? Are there first-come-first-serve advantages? Are there burning or redemption mechanics?
3/ Decision
This last part is the hardest. But you have to make the call and live with the consequences.
Conclusion
Designing NFT sales are tough.
For Guild of Guardians in particular, our preference is to always give back to community members, which means we want to make it appealing to be an early backer of the game (rather than just us making as much money as possible).
This makes our lives harder because it means we attract all sorts of buyers, and need to factor this into our mechanism design decisions.
However, designing effective sale and distribution mechanisms is also what we believe will result in the long-term success of the game.
Additional Recommended Reading:
This article definitely doesn’t cover everything to know about running or designing a NFT sale. Some recommendations here:
Vitalik explains the challenge with fixed pricing and proposes Proof of Personhood as a solution (although as we’ve seen recently with Yuga Labs Otherside landsale, even proof of personhood via officially KYC addresses can get exploited and traded)
Paradigm wrote a guide on NFT launches, which dives into things like distribution and metadata reveal which aren’t covered in this article
Note: Personal thoughts only. If you enjoyed this please subscribe. Follow me on Twitter here: www.twitter.com/xdereklau (DM’s open)
I just finished reading your posts, and they provided some great insights! I look forward to watch Guild of Guardians grow and succeed.