Token Bound Accounts
Hello!
We had the pleasure of meeting Gabby from YGG during our trip to Singapore a few weeks back. One of the many things we discussed was the evolution of Tokenbound Accounts (TBA) and why they are interesting. I did not allude much to it in our piece on identity because I figured it deserved a stand-alone piece of its own, much like we did with soulbound tokens (SBT).
Some basics in case you are hearing about it for the first time: ERC-6551 is a token standard that allows an NFT to act as a wallet. This means an NFT can own assets like stablecoins, open positions in DeFi products or other NFTs. An ERC-721 asset (NFT) is a stand-alone, isolated asset that cannot necessarily be upgraded or hold assets like a wallet does.
Note: I will use TBA for token-bound accounts and SBT for soulbound tokens for ease of reading throughout this piece. We need to find words that are easier to say..
When you buy an NFT, you buy a single asset, which may give you access to third-party services such as an event or product perks. Bored Ape Yacht Club (BAYC) NFT owners had access to a restricted Discord chat and rights to mint a handful of tokens released by Yuga Labs.
An SBT is a non-transferable asset sent to an individual's wallet, attesting to their credentials. Below, I provide a quick recap on both contexts before we dive into TBAs.
In the context of identity, these assets are currently used in two ways.
NFTs are used to attest ownership by simply checking wallet balances. Using NFTs to prove identity would mean a person was either wealthy enough to acquire it (with capital) or skilled enough to mint it early on when the NFT was released.
An SBT, on the other hand, cannot usually be acquired by capital alone. By nature, they are nontransferable1, so users have to put in effort and time to acquire them.
But what if you had a mechanism to combine SBTs, simple assets (like stablecoins) and NFTs into a single standard and gave it the ability to transact? That is ERC-6551. In this model, you convert an NFT into a wallet. A user's action can add layers of assets to the NFT. These layers can be metadata that is hosted on centralised servers or assets that are held on-chain. A user with the TBA can move single assets (like stablecoins) or transfer the TBA along with all of the assets held by it.
The term “metadata” above could be a bit confusing without context. Generally, an NFT itself does not hold any data of its own. It points to a third-party hosting service. What I mean by “hosted on centralised servers” is that an NFT’s capabilities could evolve based on what a developer wishes to be within a game. A server could check which assets are held in a wallet and customise the user’s experience accordingly.
For instance, it could unlock a certain level or enable access to a region in a metaverse. The experience itself (within a game) could evolve from time to time so long as a user owns a particular NFT. In such a model, the NFT is less about ownership of a static asset and more about the unique experiences it enables within a game.
Wait, what? Why does this matter? BAYC offers some clues.
In-Game Loyalty
In March of 2022, the price of BAYC NFTs dropped over 30% in the days that followed the airdrop for APE tokens. That is because a simple mechanism exists to price the value of an NFT like BAYC before an airdrop.
Think of it like this: The price at which a BAYC used to trade in the days before the token’s release was the implied value of an anticipated airdrop combined with the latent value a market gives for the underlying NFT itself.
Once the airdrop was claimable, markets rapidly repriced to reflect what the NFT was worth on its own. This behaviour is partly because NFTs have historically been static assets with limited use cases. Users are not incentivised to hold on to them if they don’t see an economic case behind them.
ERC-6551s offer a mechanism to drive value and loyalty to the NFTs a user owns. Currently, most NFTs are priced based on their rarity. When limited-edition collections are dropped as NFTs, users sift through them to find scarce attributes and price them higher. ERC-6551s allow users to build NFTs with differentiations that go beyond what an artist created when releasing an NFT.
To understand this analogy, we need to go into gaming. Remember when Play to Earn was all the rage and Axie Infinity’s NFTs were trending? Part of the challenge was that very little differentiated one NFT from another. There were attributes to the NFTs that made some rarer than others, and you could combine them for specific advantages within a game.
But for the vast majority, a user’s progress within a game had no impact on the NFTs they already owned. ERC-6551s and SBTs allow an alternative.
The image above tracks the evolution of a game character throughout a game.
We presume all users start with a tokenbound account. The NFT is a wallet.
Users spend time in the game to receive XP (experience points) that are given out as SBTs. A user cannot trade the XP itself. There is no liquid market for XP in the above hypothetical example. If users had to transfer their XP, they would have to transfer the NFT. The equivalent of this on the web today is selling your RuneScape account.
On reaching certain levels of XP, a user can mint additional NFTs that go into the user's wallet. A game could interpret these NFTs as upgrades to the character. For instance, owning a wand or a hat could give an edge in terms of the attack or defence capabilities of the character.
Remember that these NFTs representing upgrades could be transferred to an external wallet by the TBA. Unless a user has the necessary XP, it could be discounted by the game's server when upgrading the character.
As the game evolves and players receive more XP, they could unlock additional NFTs representing upgrades. These NFTs could be minted directly from the game (for free) or traded in a free market.
What does this solve for? Structurally, it reduces the amount of speculation with a game's assets. Since the reward is XP and cannot be transferred or traded as quickly as tokens, it should create a base of loyal users more interested in the game than the asset rewards they receive.
Additionally, it keeps users retained longer, as selling their assets is the only way to cash in. There's the alternative of selling stand-alone NFTs that translate to upgrades within the game. But the market for those assets would be small as only users with sufficient XP would trade their assets with one another to combine NFTs and unlock specific abilities within a game. (The reason is a game would not recognise a TBA that does not have the necessary XP simply owning NFT upgrades. The XP would be a mandatory check on the server side).
TBAs are interesting for guilds because they allow stand-alone characters (or gamers) to form decentralised autonomous organisations (DAOs). In the current guild model, there are very few mechanisms to vet the actual capabilities of a gamer. If TBAs were used, members forming a guild (or a DAO) to coordinate and play games together could vet the XP, past transactions and assets owned by a gamer. It makes coordination between gamers easier when coordinating and taking on hard levels within a game or competing.
In games like RDR2, every upgrade to a character’s dressing is purchased with cold hard dollars. The belt, gun, hat, necklace and earrings could each be an SBT in a hypothetical world where Rockstar games embrace NFTs. A player could use the same TBA in GTA 5 to have a completely different set of perks, given that the server-side interpretation of a TBA could vary from game to game.
Why would Rockstar Games - the publishing studio behind both games, do this? Part of the reasoning is that it allows third-party developers to create value for their userbase by creating unique experiences for players with a TBA from Red Dead Redemption. A simpler reasoning is that such a model where a gamer can receive perks on multiple games for purchases in any one could create more brand loyalty and user retention.
The challenge is that such a model would not work in games where everybody starts from scratch. Would it be great to build a Red Dead Redemption or GTA 5 character with such upgrades? Definitely. But a model like the one above would fall apart in a game like DOTA where every player starts at the same level.
It is important to note that token-bound accounts are a new design philosophy. If we are to look at it from the lens of 2019 NFT trades, they may not make much sense. In the past, NFTs were optimised to switch hands frequently because it meant the developer received royalties on each trade. In 2022, Blur disrupted that model entirely by reducing royalties paid by traders.
All these sound interesting on paper, but one has reasons to wonder: Where has such a standard seen any traction?
Co-creating Influencers
Naturally, not everyone plays games or needs to upgrade an in-game character constantly. There is a larger game we all engage in: social networks. They are uniquely suited for such a token standard, as we tend to 'own' our social profiles for long periods. Web3 native social products like Mirror or Lens allow users to collect posts by their favourite creators.
They also allow multiple clients to surface content to users. Lens's latest update (v2) allows the creation of token-bound accounts in their protocol. It means that an NFT (like a crypto-kitty or a BAYC character) could have its own Lens account, which – in turn – owns its list of assets. Why is this relevant at all?
I believe it gives a new infrastructure for how we think of intellectual property rights and asset ownership in Web3.
For instance, a creator who has built a brand around a non-existent, AI-based character could sell it to a third party willing to pay for it. This happens in the Web2 world already. Pages with millions of followers on Twitter or Instagram are sold quite often. The only difference is that with a token-bound account, this process would be as simple as transferring an NFT to a third party. The recipient would, in turn, have access to the follower base and assets owned by the NFT in question.
Does it seem far-fetched? Yes. There aren’t enough users on Web3 native social networks to warrant such a use case. But technological trends elsewhere point to the need for this. Consider the case of Lil Miquela. In 2018, she was one of the Time’s most influential people online. She has collaborated with brands like Calvin Klein and Prada. That's justified when you consider she has over 2.7 million followers on Instagram. The kicker is she's AI-generated.
As generative AI makes it easier for creators to develop virtual influencers, the desire to collaborate, co-own and trade digital brands will become increasingly relevant. Web3 native primitives allow multiple individuals to own the art or data sets that go into training such a model. They could even stream the payments (from brands or audience bases) to multiple contributing wallets.
It may seem impossible because brands take years to build and form trust. The Lindy effect of a brand (like Gucci) may not be replicated soon. On the flip side, though, a community-owned influencer that engages in co-creation may be able to scale an audience base far faster than traditional influencers or brands. As the tooling around DAOs, AI, and on-chain IP come together, we will see brands disrupted like influencers disrupted traditional media.
How does any of this relate to the token standard I just mentioned? If Miquela were an NFT and had SBTs for each brand interaction or on-chain proof of her collaborations, the NFT would be worth way more than the generic ones we have in the market today. If we are to see thousands of virtual influencers online in the future, wouldn't it make sense that they should be tradable? That would be far-fetched but within possibilities, considering we already trade human keys on FriendTech.
A future where multiple creators fine-tune a virtual influencer to scale and sell to a third-party buyer seems dystopian. Perhaps we should step back and look at what is possible here and now. A different approach to scaling ERC-6551 as a standard would be to look at a user's Web2 data to identify patterns that may make the person valuable.
For instance, although multiple products (like Audius) exist to help connect emerging artists with an audience base, none have meaningfully scaled. This is because the on-chain social graphs of Web3 native music product users are not built by observing a user's ability to identify and endorse an emerging artist.
One way to solve this problem would be to scan a user's Spotify playlist history and issue SBTs in proportion to how early they identified with an emerging artist. Weightage could be given based on how early and frequently users listened to an emergent artist who has since blown up. Asset Money seems to be taking a similar approach with their product launch.
If NFTs that are token-bound accounts are issued to such users, and reward badges (like the ones you receive on Audible or Apple Watch) are given as SBTs, you could – in theory – create a curated, on-chain graph of music enthusiasts who are incredibly adept at embracing and endorsing new musicians.
A major artist like Taylor Swift may have no reason to use such a product. But somebody up and coming could better target the most active listeners using tools like this. Naturally, artists are not data scientists developing SBTs from your Spotify data—a third-party platform would have to do this for them.
Platforms like these would tap into Web2 native data to create on-chain graphs of user bases with specific characteristics. These users could then be incentivised with discounted tickets, exclusive merch and so on to help build an audience base. One of my favourite instances of an artist building loyalty is J. Cole doing concerts for $1 during the early days of his career.
Such products open up highly profitable niches that do not exist on the internet today.
Long-Term Games
ERC-6551 is a standard and an early stage one at that. Developers and creators decide what's built on top of it. The standard is intriguing because it focuses on building reputation and credentials instead of acquiring and selling an asset. Using it for games, music, or creator brands would be more about gradually developing value around its NFT core than public speculation.
A simple way to consider this is through the lens of venture capital. When a venture capitalist (VC) invests in a firm, he has, in essence, a SBT of sorts. An investor's track record measures conviction and skill in detecting early-stage ventures. The VC's access to better deal flow at lower valuations increases depending on how many successful bets they make. This process is how reputation is built in the off-chain world. It takes time, but reputation aids in compounding value in the long run and unlocks new levels much like in a game.
At its core, ERC-6551 enables this long-term reputation-building for digital assets. The digital asset can be a media brand, a user's behavioural patterns or an in-game character. Gradually adding characteristics to an asset can be crucial in building consumer retention and loyalty in Web3. Is it as rewarding as seeing a token pump? Absolutely not. But not everybody wants to flip assets and deal with volatility. TBAs are one of many tools that empower users who want to hold an asset and build a reputation around it.
One of the things I did not explore sufficiently in this piece is the role the token standard would have in finance. For instance, bills of lading are increasingly going digital, and there is a strong case for using ERC 6551 there. But we’ll dig into that another time.
If the theme interests you, I highly suggest hanging out in the TokenBound Telegram group or digging through the documentation on their website. I also recommend tracking adoption through SeaLaunch’s dashboard here.
Going to play Red Dead Redemption 2,
Joel John
The caveat here is that an SBT held in a multi-sig wallet is actually transferrable, as users can simply rotate out the public keys used on them.