Beyond Verticalization of NFT Marketplaces

Why is the UX of OpenSea and Magic Eden generally the same (discovery on homepage, basic set of linear filters on the left side of collection pages, same search presentation, etc.)?

Or, framed another way, why doesn’t the front page of OpenSea or Magic Eden have an activity feed (like Fractal or Hawku), or a livestream (like Twitch), or the results of a recent game tournament that heavily leverages NFTs?

The answer is that history is path dependent. OpenSea was founded in 2018, when there were only a handful of NFT collections such as CryptoPunks. CryptoPunks was a collection of 10,000 pixelated NFTs with some simple and filter-able attributes. Bored Ape Yacht Club (BAYC) was clearly heavily inspired by CryptoPunks: 10,000 algorithmically generated images, with simple and filterable attributes. These two collections are, for practical purposes, identical. Most of the PFP collections that have followed—Pudgy Penguins, Moonbirds, DeGods, Solana MonkeyBusiness (SMB), Azukis, etc.—all map to the same structural model and historical paradigm. Beyond the art itself, there are no meaningful differences in the design of the collections themselves (obviously they have very different communities).

It’s clear that communities are mapping to the technical specifications of marketplaces, which is not how the world should work. Marketplaces should map to the unique characteristics of communities, not the other way around. The design space for online communities is vast, considering the ability to leverage and configure novel elements across DeFi, NFTs, DAOs, social tokens, and more.

There’s a fairly commonly held view that NFT marketplaces will verticalize. This naturally makes sense because it’s actually weird to have a domain space exchange (.e.g., .eth or .sol) with the same UX as a PFP exchange (e.g., BAYC or DeGods). Marketplaces should be optimized around the type of content they host.

But this line of thinking doesn’t end there. This reasoning can go much farther. NFT marketplaces are not just going to verticalize. At a minimum, they are going to:

1. Become owned by the communities which they serve
2. Integrate many non-commerce functions (e.g., live streaming, chat, forums, etc.)

Community-Owned NFT Exchanges

Magic Eden and OpenSea are throwing off cash (see here and here). Where is that money coming from?

The NFT communities themselves, in the form of 2-2.5% transaction fee per trade.

I’m not going to argue that trading fees will go to zero. They won’t; rather, I will argue that trading fees will not go to a 3rd-party exchange separate from the NFT community, but rather back to the community itself. This shift is a sign of community maturity. When the community—and its brand—is strong enough to stand on its own it will capture the opportunity to further fund its treasury and increase brand loyalty through features and engagement that are custom-built for the community.

BAYC is already doing this: 2.5% of each trade goes to Yuga Labs (in addition to the 2.5% that OpenSea charges). There is no reason to believe that BAYC holders are particularly loyal to OpenSea. On the contrary, I would expect that if Yuga Labs launched a BAYC exchange where 100% of fees flow back to the ApeCoin-governed community treasury, the vast majority of BAYC and MAYC volume would quickly move there simply as a function of brand loyalty and a desire to grow the community treasury.

The problem is that 3rd-party NFT exchanges are parasitic to NFT communities. They don’t really provide value back to the communities (other than some value around Launchpad discovery, which only provides value at one point in time, and not on an ongoing basis), and they explicitly extract value. The marketplaces that survive have to provide real value—whether it’s letting players dig into insights, analysis and data, providing a channel for prospective players to discover new games, or creating a community that like-minded players can join. NFT marketplaces have to provide more than being just a financial intermediary.

If my theory is right, why hasn’t this happened yet? I posit it’s because there is a meaningful amount of work that goes into hosting an NFT exchange, and that work hasn’t been entirely commoditized, yet. Metaplex and Grape Network (both are portfolio investments), along with Reservoir, and others are building out all of the required features to commoditize general-purpose exchanges fully (e.g., indexing, querying, search, curation, etc).

For example, Grape Network just launched the first DAO-owned NFT exchange. It’s still early, but it’s coming together quickly.

Once that infrastructure is generally available and communities can easily host their own exchanges, a few interesting things will likely happen:

1. Fake and fraudulent PFP volumes will collapse to near zero. Obviously the UI of the exchange at bayc.com (as a theoretical example) will only index and serve authentic BAYC NFTs, and not fakes. As long as the user is on the right domain, it should become nearly impossible to purchase fake NFTs.
2. The community will want to embed the exchange everywhere that the community congregates. For now, this primarily means social apps: Discord, Twitter, Reddit, etc. Obviously, the web2 social apps don’t properly integrate NFTs as well as they theoretically could to support NFT communities. I expect this will create an opportunity for crypto-native versions of these apps to flourish, such as Satellite (crypto-native Discord) and Farcaster and Orbis (crypto-native Twitter). Users should be able to bid/offer natively in these social spaces.

NFT exchanges need to become API-first products. There will be a single shared liquidity pool that is accessible across many front ends. And they will be community-first products as well.

Building Exchange-Enabled Community Hangout Spaces

People hang out in all kinds of online spaces associated with various communities. Some examples:

1. Fortnite lobbies
2. Discord servers
3. Subreddits
4. Twitter
5. Telegram

These experiences are almost entirely disjointed. People communicate on one channel, often just to re-organize in another. There is almost no shared state among these community instances.

There are two common places that tie these hangout spaces together: games themselves, and NFT marketplaces. People chat about the games mostly outside of the game, and then go into the games to play. Similarly, they learn about what NFT communities are doing on social channels, then go to NFT marketplaces to complete a transaction.

Communities are unlikely to ever congregate in one place. At a minimum, there will always be a Twitter-style global stream that is not community specific. And so communities will want to embed their marketplaces in a Twitter-like product, in addition to Reddit-like and Discord-like products. It’s therefore paramount that the NFT marketplace that powers these communities is API-first. It should be a marketplace that’s intended to live within other user interfaces from chat to livestream to dedicated two-dimensional HTML.

Given that communities will naturally congregate across a variety of social channels, what does that mean for the primary domain (e.g., bayc.com)? A few things: bayc.com becomes 1) an aggregator of community hangout spaces, and 2) the flagship UI for that community’s marketplace.

You can imagine that when you go to bayc.com you should see:

1. Trailers for new BAYC-branded movies
2. A live stream from a prominent community member in the BAYC metaverse (Including in-stream tipping in the native currency of the community)
3. Highlights from recent community events, IRL BAYC graffiti, etc.(You can imagine that below some pictures of existing graffiti, there is a call to action that says “Help spread the word in your city” with some links and commerce functions, links to graffiti discords, etc.)

Commerce should be natively embedded throughout the site as users browse. The content itself will likely become dynamic based on the users on-chain credentials and current assets (e.g., first time users see movie trailers above the fold, whereas seasoned BAYC veterans see live streams from their favorite community members who are live in the BAYC metaverse).

Of all the major consumer Web2 products that exist today, the closest to what I’m describing is probably Twitch. However, Twitch is 1) insufficiently community-centric and 2) obviously not designed to be extensible to support the diversity and nuance described above.

The more that communities grow, the more complex this management becomes. This creates natural challenges and opportunities, and it becomes even more complex when multiplied across myriad communities. This is why NFT communities will move away from generic NFT marketplaces.

NFTs, like everything else in crypto, are fundamentally about sovereignty. Their marketplaces will be no different.

Thanks to Stephen Hess, Charlie Graham, Matt Shapiro, John Robert Reed, and Shayon Sengupta for providing feedback on this essay.