Review and Outlook of the Cryptocurrency Industry: Which Tracks are Worth Paying Attention to in 2023?

Author: zixi

As the year comes to a close, let's try to summarize the stories that unfolded in 2022 and take a look at what might happen in 2023. 2022 was a year that started strong but ended weak. At the beginning of the year, although Ethereum dropped to 3800, we could still see the enthusiasm for on-chain transactions and the booming NFT market. However, after the collapses of Luna and FTX, the entire market began to stagnate, and no new narrative directions emerged.
This article is divided into three parts, mainly discussing what was overly optimistic this year, what is promising for next year, and what still needs observation for the coming year.

1: What was overly optimistic this year?

1.1 NFTFI

NFTFI was actually a sector that many had high hopes for at the beginning of the year. Based on the frenzy surrounding NFTs, countless outsiders rushed into this hot sector. At the start of the year, many big players spent lavishly on BAYC (100E+); the gas war for Monkey Kingdom was so intense that the price for minting an NFT exceeded 2E. When the floor price for a single blue-chip NFT exceeded 10E, people began to wonder how such high prices could allow the general public to participate.

Figure 1: Opensea's trading DAU, transaction counts, and trading volume show a downward trend this year.

Figure 2: NFT Index

Thus, at the beginning of the year, there was a lot of enthusiasm and expectation for the NFTFI sector. People hoped that this sector would not only feature P2P, P2Pool, and other lending transactions, but also lower the barriers to entry, which became a major direction. This year, we also saw new concepts like index trading, fragmentation, crowdfunding, and buy now pay later emerge. In the first half of this year, models like partybid began to appear, and Cyan started offering NFT's BNPL.

Let’s take NFT lending, which has relatively larger market space, as an example. If we look solely at the lending volume, the overall borrowing amount has remained basically unchanged since May, with BnedDAO and NFTFI contributing more than half of the borrowing volume. However, if we observe the number of users on the platforms, the combined daily active users in this market is only around 100, indicating that the coverage is still too small. The limitations on daily active users and trading volume partly stem from the decline in the NFT market's popularity.

Here's a cold joke: why has the decline of BAYC's price in ETH terms (66%) been less than the decline in USD terms (75%)? The reason is poor liquidity; no one is willing to trade frequently to reflect the true market price, so of course, the price doesn't drop.

So why has the floor price recently dropped from 70ETH to 50ETH? It's because some holders need to cash out and can only lower the price in hopes of finding buyers, ultimately triggering liquidation points due to oracle price feeds, leading to a cascade of liquidations of large monkeys, thus starting a death spiral. It has come to the point where effective declines rely on a series of liquidations to generate sufficient trading volume, highlighting the poor liquidity.

Therefore, people were overly optimistic about the development of NFTFI this year, but Sudoswap has indeed revitalized the market.

Figure 3: NFT lending market Borrow volume

Figure 4: NFT lending market DAU

1.2 SocialFi/MusicFi and other C-end applications

Let’s not discuss whether the blockchain infrastructure is currently sufficient to support on-chain social/music products. In the past, there have been underlying public chains like deso specifically supporting social applications, but their development has been less than satisfactory, primarily because the ecosystem and user base have not been built up.

The logic of SocialFi differs from DeFi and GameFi, but it shares similarities with MusicFi, as both are based on leveraging the influence of big names/musicians to purchase related fan tokens/NFTs. The logic of SocialFi can be simplified to—big names want to monetize their social capital (gain tokens), so they need to spend time and money to prove the value of their social capital (POW proof of work). Similar to the logic of BTC/ETH, this also indicates that blockchain-based Web 3 is very suitable for the development of SocialFi.

In essence, SocialFi can be understood as big names (or individuals) using their influence to build their social capital (i.e., personal brand reputation) while also providing benefits to their fans.

However, this logic has led to significant issues for SocialFi: 1. The desire for decentralized social platforms, but in essence, decentralization is not achievable (a few celebrities can create very strong social monopolies). 2. Content mining is very difficult to define for social platforms. 3. The empowerment of platform tokens is quite important for social platforms and token distribution platforms. 4. There are very few users. Currently, SocialFi still has a long way to go for development; its market scale is small, mechanisms are relatively primitive, and there are few Web3 natives willing to use products with poor user experiences.

The issues in the MusicFi sector still exist:

1. From the user perspective, the overall number of Web 3 users is still small, and even fewer are willing to spend their limited time listening to Web 3 music, let alone spend real money on royalty-free NFTs. Essentially, relying on limited content to capture users' limited time is very challenging.

2. From the IP perspective, there is a lack of leading stars like Taylor Swift and Jay Chou; current Web 3 musicians are relatively long-tail. However, this year we have seen some leading stars, represented by Wang Feng, attempt to venture into Web 3 music.

3. From the project perspective, many projects have incomplete economic models, weak token empowerment, and the entire project resembles "I created this environment, users have this demand, so I issued a token." Music projects with strong social characteristics are similar to SocialFi. When SocialFi project tokens have poor empowerment and start content mining to initiate cold starts, the consequence is that users begin to create unlimited junk content, falling into a death spiral of mining and selling. Therefore, when music projects start Listen to Earn and Create to Earn, if tokens are not given reasonable value, the death spiral will also be the final outcome.

Figure 5: Comparison of social and music with DeFi and GameFi

1.3 Cross-chain Bridges

Last year, the market generally regarded cross-chain bridges as core infrastructure, which is quite evident, as last year's public chain landscape featured one dominant player and many strong contenders, with new public chains emerging continuously. The market had numerous public chains like BSC, Luna, Solana, Tron, Avalanche, Polygon, and Harmony. Due to trust issues with centralized official cross-chain bridges and relatively poor scalability, high-scalability third-party cross-chain bridges like Anyswap and cbridge began to thrive last year and early this year.

But what happened this year? Luna collapsed, and the Terra public chain disappeared; FTX collapsed, and Solana was affected; Aptos and Sui did not gain traction with players; on-chain TVL declined, partly due to the drop in token values and large sell-offs leading to liquidations (on the other hand, there was even an abstract story about nearly 70% of Solana's TVL being double-counted). Additionally, the inherent complexity of cross-chain contracts has made them a breeding ground for hackers.

Looking at some interesting data, a recent report from Chainalysis shows that cross-chain bridge attacks accounted for 69% of the total stolen cryptocurrency in 2022, with losses reaching $2 billion. This year, there have already been 13 cross-chain bridge attacks, including the recent Nomad cross-chain bridge attack that lost $190 million, and the Ronin cross-chain bridge attack in March that lost as much as $624 million. The coexistence of multiple chains from last year, with a competitive landscape of one dominant player and many strong contenders, has gradually turned into an oligopoly of ETH+BSC+Tron.
Cross-chain bridges are one of the most important infrastructures in public chains, but for investors, due to the presence of hackers, they may not be a very profitable business.

However, it cannot be denied that excellent cross-chain aggregators like LiFi, Chainge, and Dbridge have emerged in the market, reducing the usage and investment risks for users and investors. But (cross-chain) aggregators are a lower-barrier, highly competitive business, yet they are tools used frequently; whether they can provide investors with excess returns is worth considering.

Figure 6: $2 billion stolen in cross-chain bridges this year

Figure 7: Stargate's transaction volume rises against the trend, while trading volume gradually decreases

2: What do I see as promising for next year?

2.1 Compliance and Regulation

Before discussing compliance and regulation, let me share a little story.

In March of this year, I looked at a project that was a cross-chain bridge, which had a feature that required KYC. At the beginning of the year, I couldn't understand why an on-chain dapp needed KYC, as blockchain is an anti-regulatory world, and KYC seemed to undermine the spirit of blockchain. However, after discussing it in person, I understood why the project team needed KYC.

This is a cross-chain bridge project done by a large traditional financial market maker. They have an extremely large capital volume in traditional stock markets and other markets. Recently, they have been very optimistic about the entire crypto sector and are trying to use a portion of their capital for market making both on-chain and off-chain. Off-chain market making is already very competitive, so they are trying to explore on-chain market making space, but multi-chain leads to capital dispersion.

In their daily market making, they need liquidity to move across different chains. If they use CEX, the large capital volume will lead to daily deposit and withdrawal limits being reached, and centralized cross-chain solutions have some trust issues (considering the FTX incident, this is indeed a long-term consideration).

If they use cross-chain bridges for liquidity, the pools of cross-chain bridges are small, leading to significant slippage and wear; if they add liquidity to the pools themselves, mixing their clean money with potentially dirty money from unknown sources, and periodically getting targeted by hackers, they will inevitably face regulatory issues—large traditional market makers are under strict regulation from the SEC, and they have no need to risk regulatory scrutiny for the small profits from current crypto market making.

So is there a way for me to continue market making in crypto while avoiding all the aforementioned issues and complying with regulations?

KYC. Through KYC, market makers can put their money into liquidity while also welcoming other KYC-verified institutions and individuals to join in providing liquidity. This reduces cross-chain wear and also meets regulatory demands, allowing larger traditional financial market makers to enter and jointly explore this yet-to-be-fully-developed on-chain market, expanding the pie together.

After the U.S. Treasury Department sanctioned Tornado Cash in June this year, I became even more convinced of the necessity of compliance and regulation. At this stage, the total market value of blockchain is not even half that of Apple; we not only need to improve infrastructure but also bring in C-end and traditional financial institutions to grow the market together.

Therefore, with the direction of regulation, projects oriented towards KYC services have begun to emerge in the market. For example, through ZK DID, users can use their information to generate ZK-Proofs off-chain via their devices, and then only need to use the Proof to verify whether they are whitelisted users. For instance, one day, if regulatory scrutiny falls on Uniswap, requiring Uni not to provide services to users from Russia, Iran, and North Korea.

One feasible approach is for Uni to collaborate with KYC service providers, where users must present their zk proof to prove they are not citizens of the aforementioned countries to use Uni. On one hand, ZK proof protects users' private information, while on the other hand, it gains regulatory recognition, making Uni's funds cleaner, more compliant, and more likely to attract traditional institutions.

2.2 Developer Tools

Let me share a case. In 2021, Alchemy benefited from the frenzy of the bull market, with a B round valuation of only $505 million in April 2021. From the disclosed data of this round, the number of customers grew 97 times in 8 months, the number of enterprises doubled in Q1 2021, and the NFT business grew over 13 times; in October 2021, it received C round investment, with a valuation of $3.5 billion. The founder stated, "The business growth rate far exceeds the valuation growth rate"; just 3 months later, in February 2022, the valuation reached $10.2 billion, with the team stating, "Since closing in October, the user base has increased by 50%." Alchemy's valuation nearly increased 20 times in one year, and the number of customers, businesses, and services for developers and the community all showed significant growth.

In 2021 and 2022, we observed an interesting phenomenon—more and more web2 developers have developed a strong interest in the crypto/web3 world. According to GitHub statistics, by the end of 2021, there were approximately 73 million web2 engineers worldwide. According to Electric Capital data, by the end of 2021, there were only 18,000 monthly active web3 engineers, with a penetration rate of less than 0.025%. From LinkedIn and OK's statistics in 2021, testing engineers and cryptography experts are among the two fastest-growing professions.

With the phenomenon of major internet companies in the U.S., China, and India starting to lay off employees, more and more web2 engineers are seeking the next opportunity. According to related rumors, "in the Bay Area alone, there are about 3,000 Chinese web2 engineers seeking new opportunities"; due to the downturn in the domestic economy, more and more engineers from major internet companies, especially those from ByteDance and Tencent, have begun their own web3 journeys. Therefore, from the data perspective, the 2 dev market is a potentially more attractive and directionally clearer incremental market than 2C.

Figure 8: Number of monthly active web3 engineers

Taking JSONRPC and API as an example, one of the markets where developers interact the most with Dapps. RPC is not originally a market with barriers; any project or individual can set up a full node to provide full-chain data for their project. However, this comes with high fixed costs and monthly operational costs, starting at $10,000 per month for project parties, and self-built nodes may also face complex operational issues like maintenance, making it not worth the effort.

As a result, RPC services will continue to centralize, which explains the rapid expansion of node providers like Alchemy and Infura; this is essentially a market with a very high Matthew effect. Additionally, there are quality node providers in the market like quicknode and Infstone. RPC is a highly competitive sector. Furthermore, based on RPC, service providers can leverage their strong engineering capabilities to offer cloud databases (datacloud) and APIs.

In terms of providing cloud databases, decentralized/centralized multi-chain real-time databases have emerged this year, allowing developers to avoid spending significant costs and time parsing on-chain data; they can simply use SQL to query data and focus on product development. Moreover, developers have numerous unique API needs, which significantly impact product development timelines.

If a product has a multi-chain real-time database and developers can independently call data to combine APIs, it will meet the long-tail needs of developers and greatly reduce the time many developers spend reinventing the wheel. In the second half of this year, such products began to raise funds, with new Infra competitors—overseas Space and Time, Goldensky, and domestic Chainbase—competing with established Infra players like Alchemy, Infura, Instone, and The Graph.

Additionally, a batch of data analysis tools has emerged in the market, including versatile tools like Dune and Footprint, as well as more specialized data analysis tools focused on DeFi, NFTs, GameFi, public chains, financing data, and on-chain black and white lists, such as DeFiLlama and TokenTerminal. Their monetization methods are primarily through SaaS subscriptions or API calls, both of which face commercialization challenges to varying degrees.

Specialized databases can connect their cleaned databases in niche areas to more general databases like Space and Time and Chainbase. This way, general databases can enhance their data, while specialized databases can increase their monetization channels, effectively turning specialized databases into an API Marketplace that connects with more developers and ecosystems, followed by revenue sharing.

3: What else should we observe next year?

3.1 DeFi

Influenced by FTX and Babel, CeFi has once again become a target of criticism. Let’s first give an example to illustrate the problems of asset management-type CeFi that manipulates black boxes. CeFi attracts users primarily through APY. If two large institutions, X and Y, are in the market, and X offers a 5% APY to attract users while Y offers a 6% APY, if X and Y have similar branding and backers, most users will choose to deposit with Y due to the higher APY.

However, X will have to increase its APY to 6.5% or 7% to attract customers due to market competition and income issues. What seems like a modest increase in APY is actually harmless during a bull market, as the strong trading, quant, and structured products within CeFi institutions can yield returns far exceeding 5%-7%.

The problem arises when market conditions change drastically, and CeFi institutions also leverage their positions; during events like 3/12, 5/19, and Luna, traders may not have time to adjust their positions before being swept away by market crashes. This has led to the recent downfall of giants like FTX and Babel. If users knew where the high APY came from, would the customer acquisition problem be resolved?

As a result, DeFi has returned to the public eye recently. We see that the ratio of DEX/CEX spot trading volume has stabilized at 15%, down about 40% from the early bull market, as we are now in a deep bear market with overall on-chain activity being poor. However, from the on-chain trading volume, 61% comes from Uni, 27% from Curve, and 9% from Dodo.

From the perspective of derivative trading price feeds coming from DEX and the liquidity of derivative trading coming from DEX, DEX now resembles the infrastructure of the entire DeFi ecosystem. Although the DEX model is continuously innovating, the undeniable fact is that Uni's market share has remained stable at around 70% this year. This may stem from the reliance on user/chain market makers/trading bot trading paths and Uni's depth advantage.

Benefiting from the negative reputation of CeFi this year, will politically correct DeFi find a new direction next year? This still needs observation.

Figure 10: Comparison of DEX/CEX spot volumes

Figure 11: DEX Market share

Figure 12: Weekly DEX Market Share

3.2 Enterprise Service Sector

The enterprise service sector is also an interesting direction. There are more and more blockchain startups now; in the early stages, managing issues with Excel was not a big problem, but as the number of personnel increases and income and expenditure grow, refined management becomes increasingly important. Unlike traditional companies, many crypto startups conduct their income and expenditure on-chain, leading to traditional SaaS tools not being well adapted. In the North American market, blockchain enterprise service companies have emerged, such as the payment protocol Zebac (which raised $28 million in March 2022), developer management platform Convex (which raised $26 million in April 2022), and financial management tool Meow (which raised $22 million in July 2022).

While the direction is interesting, is this a large market? Assuming a monthly fee of $1,000, if there are 1,000 early blockchain companies using the product after some time, the revenue would only be $1,000,000 per month. The maximum annual revenue would be $20 million, and using the PS standard of 15-20 for Nasdaq SaaS companies, that would only be a ceiling of $400 million. Not to mention how long it would take to BD 1,000 companies and whether there could be $1,000 monthly revenue per company. Convex raised $26 million in one go, with a valuation of at least $130 million to $260 million, so the returns for investors may still be limited. The ceiling of the enterprise service sector still needs observation.

3.3 Traffic Sector

The traffic sector is a well-worn topic, usually referring to sectors like gaming, wallets, music, and social media where traffic enters.
In the gaming sector, many major companies (Funplus, Photon Games, Ubisoft, etc.) have begun to explore Web3 gaming, but whether players are willing to pay remains to be seen.

Regarding social media, we have seen Tencent, Meta, Twitter, and others boldly exploring Web3 social, with some new Web3 social products like debox and some social search protocols like lens, rss3, and mask. We look forward to Web3 having its own social platform in the next bull market, one that can attract enough newcomers into the circle and gradually convert users.

For wallets, this year has seen the emergence of "new concepts" represented by AA and ERC4337. But upon closer inspection, social recovery and gas subsidies—aren't these just reheating old ideas? Wallets like Argent and Loopring had already implemented similar capabilities to the current 4337 AA contract wallets years ago, yet they still haven't taken off. Essentially, the need for gas fees to create wallets, along with the bugs and complications in the social recovery logic, and the need for BD with other DApps, make development very challenging. However, everyone believes in the fact that wallets serve as traffic entry points, so I still want to observe what kind of wallets can occupy a certain market space in the recovery of the market next year and the year after.

3.4 Layer 2

Layer 2 was a fantastic direction in 2022. This year, we saw type 3 zkevm Scroll striving to transition to type 2.5 or type 2; type 4 zksync gradually shifting to type 3 zkevm and launching zksync 2.0; and type 1 zkevm Taiko's testnet also went live. However, in reality, a truly usable zkevm is still a bit far from us. Previously, during a lecture by Vitalik at NTU, I asked him about the real distance to a true type 1 zkevm.

If we want to achieve a true type 1 zkevm, it may take another 10 years. During this time, we not only need to overcome algorithmic shortcomings and optimize circuit algorithms significantly, but we also need to rely on stabilized circuit algorithms to design ASIC miners to accelerate proof generation. Currently, generating a proof for a type 1 zkevm takes about an hour, which, while not affecting normal use, does impact user experience when it comes to cross-chain operations and withdrawals.

In summary, with the gradual consensus on modular public chains, the positioning of L2 is becoming increasingly important. Based on Starkware, we can also iterate to develop app chains for L3 and L4. Although the theory is beautiful and Ethereum's market share is gradually increasing, the technical implementation of L2 still requires time, and ecosystem development also needs time.

Figure 13: Different types of zkevm

Finally, here’s a calendar of the crypto world for this year, lamenting the hardships of people's lives.

Figure 14: Crypto world calendar for 2022

Due to space constraints, I won't elaborate further. In fact, there are many possibilities, such as new sectors represented by knowledge graphs—used to trace entities rather than addresses, which is a new idea; on-chain advertising, incentivizing users to feed data to ML/AI to generate models, and finally using algorithms to accurately allocate ads to users (Web3 bytes); with the explosion of Web3 gaming and music, whether Web3 can provide cloud rendering through idle machines from retail investors; and whether new decentralized wireless can change the traditional operator model.
However, many stories this year have revolved around reheating old ideas, and there haven't been many new concepts in the past six months. I hope 2023 will bring new narrative directions.