Blurring Lines

What's next in store for NFTs.

Hello!

January of 2022 was an interesting time. LooksRare launched with token rewards for NFT traders. Rewards on the platform were a multiple of the fees it used to charge at the time. So, users were incentivised to spend $2 in fees to receive $4 in tokens. As the token price rapidly declined, the interest in trading on the platform evaporated alongside it. The arbitrage had dried up, and with it, the volume.

OpenSea was back as a leader in the business until Blur launched. Blur likely learned from the LooksRare fiasco and avoided releasing the token immediately. Instead, it was released in phases. It looked at user behaviour on their platform and measured activity with a point scale.

Blur’s user acquisition strategy

Users were not incentivised to produce trading volume on Blur, which prevented wash trading. A few months after the airdrop, Blur launched a lending product named Blend. By this time, Blur had built a set of core users that stayed for the product and not for the token.

Blur did not focus on incentivising wash trades but managed to attract the bulk of the volume by collapsing the existing fee model around NFTs.

Blur announced its token in October 2022. Airdrop season 1 (the time users could make themselves eligible for the airdrop) continued till February 2023. Season 2, which was supposed to end in March and then in May, is still going on as we write this. Continued extension of season 2 means that users or airdrop farmers receive tokens after unlocks for the team and their investors, creating a conflict of interest. Some platform users received no liquidity for their tokens, while team members could sell theirs in the open market.

Until Blur launched, artists could expect to receive a percentage share of each transaction for perpetuity. This meant an artist’s income was no longer tied to the number of artworks they produced, but to the frequency with which it changed hands. By removing the royalties paid to artists, Blur made it possible for speculators to trade as frequently as they’d like. NFTs began trading like altcoins. And with it, the NFT markets changed forever.

Show Me the Money

Web3 is undergoing its business model innovation phase. This is an image of Jack Ma from 1999.

We have seen a variation of marketplaces competing for users and volume in the past. Alibaba's Taobao launched in China in 2003. eBay was already well established in China then and had over $2 billion in annual global revenue. Taking on an established player like eBay was not an easy task. At the time, eBay charged users to list products and services.

Instead of directly monetising, Taobao improved users' experiences while buying and selling and, as a result, created ventures around its core product. 

• It made listings free.

• It integrated a chatting app that allowed buyers and sellers to connect.

• It introduced Alipay, an online payment system. At the time, online payments were new in China. Alipay formed partnerships with banks, which made payments from buyers to sellers frictionless.

• It integrated ad companies like Alimama, which had a network of thousands of specialised websites that allowed sellers to reach their target audience.

As a result, rivals could barely keep up with the depth and breadth of Taobao's services. By 2010, Taobao had captured 80% of the e-commerce market in China. The takeaway is that sometimes, it is unnecessary to start monetising the core product from the get-go. Building auxiliary ventures can help the core business, and they can be monetised in due time.

Blur is following a similar playbook. Instead of going for direct monetisation, it launched NFT lending. They currently lead in the lending category. The easiest way for Blur to make money is to charge a platform fee for the trading marketplace and a haircut fee to lenders.

Here's the napkin math of how that may turn out.

For the trading side –

• The YTD trading volume on Blur is $6.4 billion. The annualised volume comes to $9.6 billion.

• The volume in 2024 could be $11.52 billion if we presume a 20% growth rate.

• The revenue would be $576 million, assuming a 0.5% platform fee.

For the lending side –

• Within four months of launch, Blur's Blend has a total borrow volume of $1.9 billion, which is ~$5.7 billion annualised.

• Assuming a 20% growth rate, the volume for borrows in 2024 can be $6.84 billion.

• Assuming a conservative 10% interest and Blur charges 10% of the interest earned by lenders, we are looking at ~$68 million in platform fees.

The total earnings for Blur in 2024 can be ~$644 million. Coincidentally, Blur is trading at $642 million FDV at the time of writing – a ~1× multiple on 1-year forward revenue. (This is not trading advice. Just an observation we made internally).

Valuing something is more of an art than a science, and the assumptions leading to the conclusion can be wrong. Also, the conclusion here is a function of governance approval for Blur to charge fees from its users. Given that OpenSea already charges a 2.5% fee on all trades, Blur charging a 0.5% fee is not farfetched. They could charge a fifth of what OpenSea charges and remain competitive because the product suite is better.

Creator Royalties

The initial promise of NFTs was that tokenisation would help artists solve royalty issues when it came to their work. This utopian dream was shattered when OpenSea recently announced that it would sunset its operator filter that enforced creator fees. In simpler terms, OpenSea is making royalties optional.

Although OpenSea's move is recent, it acquired the leading NFT aggregator, Gem, in 2022, and Gem always supported optional royalties. Game theoretically speaking, newer marketplaces have little to no incentives to support creator royalties as it eventually adds to the fees the end user pays.

Blur’s relative dominance in volumes usually translates to a complete loss of royalties for artists behind the collections.

A comparison between how much creators earn via minting vs. royalties is a good way to understand the after-effects of all marketplaces tending towards optional royalties. Although this data is not readily available, according to Magic Eden, the median Solana project made ~6% of its revenue from royalties. For most artists, royalties make up a small component of their revenue. The mint or primary sale is where the bulk of their money is made.

For creators, the incremental cost of creating NFTs is low, and primary sales of NFTs constitute most of their revenue. Much of the NFTs traded in high volumes are usually part of a collection with tens of thousands of NFTs and not one-of-one art pieces similar to those you'd see in the real world.

The pricing power of NFTs depends on who the creator is instead of what has been created. This is primarily why collections like CryptoPunks of Bored Ape Yacht Club can still attract a premium. If an unknown artist issues an NFT today, it'd be valued more for its ability to attract a community than the quality of the art itself.

There are three key stakeholders in the NFT ecosystem today. Creators, marketplaces, and collectors or traders. Out of these, only marketplaces can enforce royalties, but that requires coordination among all the marketplaces. For them, creators are on the supply side, and traders are on the demand side. Will there be any buyers if creators disallow their NFTs from being traded on marketplaces? This is why the most likely scenario is all marketplaces resort to optional royalties.

Like everything else in crypto, NFT markets are also an outlet for speculation. It has become a niche asset class within crypto assets. When the intention to buy is to make a profit, why would you be willing to pay additionally (in royalties) to creators for every trade? Free market equilibrium is that creator fees become optional across marketplaces. Technology is not the answer if we want creators to keep earning fees from secondary trades. It needs to be a social phenomenon.

We must create a collector base that wants to honour royalties instead of coercing them into paying royalties. The evolution of online streaming and the music industry in the early 2000s offers clues about how this may occur. When streaming platforms launched, indie creators suddenly had a new distribution mode at their disposal. They were no longer at the mercy of TV networks or theatre owners to distribute their shows or movies.

According to Vox, there used to be 80 shows a year across TV networks, and there are now 500+ shows a year across streaming platforms. The number of episodes per show was higher for TV, so the shows ran for extended periods. Commercials during a show were a significant source of revenue. 

Streaming platforms removed commercials as consumers were paying for access. The subscription revenue picked up the slack in revenue. Naturally, the nature of shows changed to adjust to the new model – there are now more shows, and the number of episodes is much smaller (eight to ten episodes per season). For producers, the economics didn't change much. They usually sell rights to a streaming service for a certain period instead of relying on advertising revenue. But for writers, fewer episodes means less money and the need to find more shows. 

On TV networks, writers get paid every time an episode airs. With streaming platforms, writers get paid a lump sum amount regardless of how many impressions the episode creates. While this paints a gloomy picture, the new model allows more writers. With 500+ shows, the number of writers needed is higher than before. However, the work and pay per writer have gone down.

This is also the crux of the writers' union strike in Hollywood. There is no technological solution to this stalemate. Ultimately, there will likely be a social consensus where all the stakeholders in the creative process are somewhat equitably compensated.  We saw a variation of this with music in the early 2000s too.

The music industry went from physical albums or records (pay per album) to iTunes (pay per song) to streaming (pay for the streaming service). Physical albums took different forms, like vinyl records and CDs. But, the revenue model remained the same. According to the BBC, 

About 13% goes to the artists, while 30% goes to the label, with a 17% cut going to the government in the form of VAT (applied at 20% and therefore 1/6 of the purchase price). About 17% goes to the retailer, while the rest goes to manufacturers (9%), distributors (8%) and the spend on administering copyright (6%). 

When iTunes launched, instead of buying the album CD for $18, listeners could purchase individual songs for 99 cents. Records offloaded the entire album on you, although you liked only one song. But Apple’s iTunes allowed you to purchase a single track. Records were the full buffet, and iTunes was the à la carte. The Beatles did not sell their albums on iTunes for the first seven years in protest.

Streaming platforms evolved the business models behind music from being a one-time consumption purchase to an attention economy model. The time spent by the audience on streams mattered more than the number of times a track was purchased.

Around 2008, Spotify entered the market with music streaming services. Before this, the number of times you listened to the song after purchasing it did not impact artists commercially. It changed with streaming. Platforms like Spotify pay artists on a per-impression basis. The typical rate is $0.006 per impression. The table above shows how these three models affect artists’ incomes depending on the number of times a track is played. Artists had to evolve with each of these iterations.

What’s Next for NFTs

Getting lost in the weeds of royalties is easy, but the point is to try to find where we are headed. It would be criminal to mention NFTs and not bring up Solana, where a lot of the innovations for the primitives are happening.

Affordable transactions on Solana have allowed marketplaces like Tensor to flourish. Seemingly a copy of Blur at launch, Tensor has shown that 'cheap and fast' can create a distinct product. Tensor's market-making mode allows traders to place orders that are not possible on Ethereum-based marketplaces.

Market-making orders allow users to sell NFTs incrementally when the price increases and buy when the price decreases. Users can choose what the incremental percentages on either side are. In essence, it provides liquidity for NFTs the way exchanges have it for tokens. This is possible only because users are not spending much for on-chain transactions on Solana.

This change in the fee model is seen with the mint costs on Solana. Compressed NFTs on Solana reduce the cost to mint NFTs by over 99%, which could unlock a new way of using NFTs. Familiarity breeds loyalty they say. Cheap minting costs allow brands to experiment with NFTs constantly and reach their target audiences.

We will see more low-cost, one-time consumption good NFTs. When the cost of minting an NFT is as low as sending an email, brands would be incentivised to issue as many, as frequently as possible. NFTs would become less about trading or speculation and more about the use cases they can enable.

We are already seeing this at Mirror. Our last article was minted over 11,000 times on the platform for a cost of next to nothing. Readers have begun discovering our content through the on-chain footprint of people interacting with the NFTs we released there. In such a model, content discovery occurs less through algorithms and more through what happens on-chain.

The brick phones of the 1980s share a lot in common with Bored Apes from 2021.

The last generation of NFTs looks very similar to the brick phones of the 1980s—High-status, low-utility, and built for the elite. Much like mobile devices, as the cost to mint, trade, and transfer these primitives collapse due to emergent L2s, we will see everyone having a variation of NFTs in their wallet. It would likely not cost tens of thousands of dollars and serve entirely different use cases.

This transition has already happened on the tech layer. You can send millions of NFTs for a couple hundred dollars today. It just happens so that brands don’t know what to do with these primitives in a UI that sparks joy for the consumer. It is only a function of time before NFTs become as common as advertisements on the internet.

We are working around how NFTs translate to better content discovery engines, but more on that in the coming week.

Signing out to touch grass,
Saurabh