Author: Castle Labs
Compiled by: Felix, PANews
Editor’s Note: As cryptocurrencies gradually gain acceptance in traditional sectors, they also seem to be revealing a glimpse of the evolution of the crypto era. Research firm Castle Labs argues that 2026 may serve as a turning point in cryptocurrency development, marking a shift from the speculative era of “one-click token issuance” to an investment era defined by “revenue generation and institutional on-chain adoption.” Tokens unable to generate real revenue will be phased out, while a select few high-quality protocols will dominate the future.
The start of 2026 for cryptocurrencies has been rocky. Most asset prices have declined; Bitcoin, which reached an all-time high six months ago, has been in a prolonged drawdown. A lack of recent positive news, continued outflows from ETFs, waning interest in crypto, business failures, and reduced venture capital investment have made the once vibrant "source" of opportunities in cryptocurrency appear to be drying up.
Although these are all facts with no positive aspects, we are moving toward a major shift: tokens with no connection to protocol revenue will plummet in value, and tokens with no revenue will not survive. The crypto space is transitioning from “speculation” to “investment.”
The event that accelerated this shift was the October liquidation, followed by a series of macro events, such as gold outperforming Bitcoin, leading many to wonder: Does cryptocurrency still hold investment value? Does it still possess the upward potential that initially attracted so many investors?
This article focuses on this transition and its impact on crypto assets and the underlying investment models.
From speculation to investment
Cryptocurrencies have gone through multiple phases of development, including the initial exploration period (when they were seen as geeky technology with no known use cases); extreme speculation during the ICO boom; regulatory neglect; major collapses such as Luna and FTX; and the current new era in which institutions are beginning to enter the space.
For a long time, the crypto industry has adhered to a “yield-first” model, creating a norm of speculation rather than investment. The popularity of products like pump.fun, which allow users to mint meme coins with a single click, confirms that cryptocurrency has long been a speculative bubble, with new users flocking in solely driven by the desire for quick wealth. The crypto industry’s “yield-first” approach can be divided into three categories:
- Low input, low output (meme coins)
- High input, high output (scam projects and slow arbitrage projects)
- Low investment, high return (Celebrity Coin)
On one hand, there are some simple and straightforward arbitrage methods in the market that have worked well so far and will likely remain effective, though the speed of arbitrage may slow down: memecoins. Memecoins are easy to launch; there’s no need to explain their purpose or utility to anyone, because the key to profiting from arbitrage lies in one principle: exiting before others do. Everyone trading memecoins understands this, and in some cases, losing money is their own fault—it’s simply how the market operates. On the other hand, there are projects that make excessive promises and heavily promote themselves, only to quietly disappear. Of course, there are exceptions—projects with low investment and high returns, such as celebrity coins.
Using last year’s token generation event (TGE) as an example, most can be classified as poor investments, as they resulted in significant losses for token holders by year-end. The declines were likely due to poor token economics, issuance during a valuation bubble (the primary cause), market conditions, and project sentiment, among other factors.
For a long time, crypto projects focused on building the best technology but never prioritized product-market fit (PMF), which is why we ended up with technologies nobody uses. But by 2026, things seem to be changing. As institutions move on-chain, the crypto industry’s “token-first” model appears to be fading. They want to use the infrastructure the crypto industry has built over the years, but their arrival comes with a major caveat: they don’t want to be associated with any of the tokens generated during our technology-building process; they appreciate the code and infrastructure and will use it, but this won’t positively impact the vast majority of tokens.
Recently, the New York Stock Exchange (NYSE) stated it would leverage blockchain infrastructure to support 24/7 trading. Robinhood has begun testing on an L2 built on the Arbitrum Stack to tokenize stocks and ETFs, allowing users to hold “stocks” in self-custody wallets. BUIDL from BlackRock and Benji from Franklin Templeton are both excellent RWA products. All of these enable instant settlement—a challenge that TradFi has faced for years due to trading hour limitations.
Regarding RWA, it is expected to reach the trillion-dollar level in the coming years. Private credit, public equity, and short-term tokenized U.S. debt are growing on-chain; users can now trade commodities and stocks with leverage on platforms such as Hyperliquid and Ostium, and these figures continue to rise.
Everyone is moving on-chain because blockchain can elevate finance to new heights. The dream of universal DeFi adoption is becoming reality, as institutions and every retail user operate on the same chain, enabling transparency, faster settlement, zero delays, and greater control over funds.
In this new era, applications with solid foundations will continue to thrive. Leading protocols in lending, such as Morpho and Aave, will maintain their dominance, having proven resilient and consistently innovative through even the most severe drawdowns. Additionally, protocols like Hyperliquid are emerging as one of the deepest on-chain liquidity providers, while enabling leveraged trading of public equities and commodities. As institutions expand, they require trading venues capable of accommodating their scale.
Oracles, cross-chain interoperability stacks, L2/L1 scaling, and token standards are what truly matter. Clearly, when institutions fully commit on-chain, no asset is guaranteed to deliver the best returns, but those with a strong track record will never be phased out and will be widely adopted by both institutional and retail investors.
Yield is king
Coingecko lists over 17,000 tokens.
There are approximately 5,700 protocols on DeFillama; when including only those that generated over $100,000 in revenue over the past 30 days, there are roughly 200 protocols or products, accounting for just 3.5%. The investable crypto pool is much smaller than anyone anticipated. Most tokens lack investment value.
If we analyze this data more practically by considering holder yields—the income returned to holders in any form—we find that only about 50 protocols had holder yields exceeding $100,000 over the past 30 days, representing less than 1% of the total protocols listed on DefiLlama.
The benchmark might be better raised to $1 million per month, as most tokens have trading volumes in the hundreds of millions or even billions of dollars.
Delving into the issue of low token holder income reveals its roots in the longstanding misalignment of incentives within the crypto industry and flaws in token design. A typical project involves two entities: the Labs and the DAO/token holders. The Labs represent the "team" in tokenomics—they are the original developers who raise funds by selling equity stakes and issuing tokens to investors in early stages, in exchange for capital to grow the business. Unlike equity, tokens are not legal representations of the enterprise and confer no actual rights to company profits. Investors holding equity enjoy these rights. However, when it comes to aligning product incentives with token value, token holders are typically at the mercy of the project team.
But over the past year, the situation has begun to change, with investors reducing their focus on speculative projects and placing greater emphasis on the actual profitability of protocols. This shift will elevate cryptocurrency to heights previously unattainable by a “yield-first” model.
Below are some key metrics crypto investors should consider when analyzing tokens. This article examines the top revenue-generating token protocols over the past 30 days, including Hyperliquid (HYPE), Pumpdotfun (PUMP), Tron (TRON), Sky (SKY), Jupiter (JUP), Aave (AAVE), and Aerodrome (AERO).
Price-to-Sales Ratio
The price-to-sales ratio (P/S) is calculated by dividing the protocol's market capitalization by its annualized revenue. The P/S ratio measures how much the market is willing to pay for each dollar of revenue generated. The premium reflected in this ratio indicates the extent to which users value the protocol’s future potential and growth factors.
Comparing some of the highest-revenue protocols and their tokens based on annualized revenue and price-to-sales ratio. We took revenue from the past 30 days and multiplied it by 12 to obtain annualized revenue data. The results are shown in the chart below.
We set the overvaluation threshold at a price-to-sales (P/S) ratio of 20, based on the P/S ratios of top U.S. listed stocks. Most protocols have P/S ratios close to or below this threshold, with Tron being the only one significantly higher than the others. Another threshold we considered is revenue, using an average annualized revenue of approximately $250 million. Only three protocols—Pump.fun, Hyperliquid, and Tron—exceeded this revenue threshold, collectively accounting for about 80% of the total revenue across the aforementioned protocols.
Token holder rewards
The next important factor to discuss is token holder yield. This primarily depends on the protocol’s revenue and the portion actually returned to token holders through mechanisms such as buybacks, token burns, and staking rewards. Today, token holder yield has become a popular metric that everyone is talking about, and it is even more important than actual revenue, as token value accumulates precisely through this mechanism.
Rank the protocols again based on holder returns over the past 30 days and multiply by 12 to obtain an annualized estimate. At first glance, most protocols fairly treat their holders, allocating the majority (if not all) of their income toward enhancing token value.
This is just one aspect, reflecting that token buybacks are underway and, if continued at a similar pace, could add millions of dollars in value to the token. To better understand this value accumulation, we also compare these tokens’ performance relative to the period following the October liquidation event, to more clearly illustrate the impact of token appreciation activities.
In the chart above, there are some outliers, such as TRON, HYPE, and especially SKY, which has shown relatively strong performance. Among these three tokens, TRON has exhibited low volatility and a relatively sideways trend, while HYPE diverged from the movement of other tokens in late January.
This indicates that buybacks alone are insufficient to boost token value; other factors—such as broader market declines, token unlock schedules and cliff releases, market sentiment around the sector narrative, and overall protocol sentiment—also play a role. All of these factors will be discussed in subsequent sections. Before that, let’s compare the two highest-yielding protocols and their token performances: Pumpd.fun and Hyperliquid. As shown in the chart below, when both tokens have positive buyback programs, HYPE performs better (HYPE’s annualized holder income is approximately $660 million, while PUMP is around $380 million), due to stronger overall market sentiment for the protocol and market pricing based on future supply shocks and unlock events.
Token economic model design and oversupply
In the crypto space, tokenomics aims to help projects raise funds from investors, incentivize users, sometimes conduct community fundraising, and allocate token supplies to the project team. There are no strict rules for designing tokenomics—each project handles it according to its own needs. This aspect is critical because it determines not only the near-term supply pressure of the token, but also how value accumulates, the mechanisms that offset selling pressure, and how well the token aligns with the interests of its holders.
The chart below shows the unlock schedules for a series of tokens with fixed supplies. While most tokens will eventually be fully unlocked, the unlock speeds vary significantly: PUMP unlocks the fastest, while HYPE unlocks the slowest. Generally, slower unlock schedules are preferable, as they reduce the risk of sudden supply shocks and the resulting market pressure from mass selling. For tokens like AAVE and SKY, the majority of supply has already been unlocked; for JUP, the long-term unlock schedule is discretionary rather than deterministic and is managed by the DAO.
It is important to note that unlocked tokens can be further categorized into investor unlocks, team unlocks, and community unlocks. Community unlocks may be used for staking rewards, incentives, and airdrops. This requires token-by-token analysis and plays a crucial role in understanding the token’s seller dynamics.
Lindy effect
The longer something has existed, the more likely it is to continue existing.
This is the essence of the Lindy effect, which applies to nearly all businesses, including on-chain operations—innovation is the key factor, as companies that fail to innovate cannot survive in the long term.
Last year, the cumulative revenue from crypto protocols was approximately $16 billion, with revenue highly concentrated among a few top protocols. The top ten protocols accounted for 80% of net revenue, with the top three making up 64%, and Tether alone accounting for 44%.
In addition, not all protocols have issued tokens; for example, Circle, the second-highest-revenue protocol after Tether, has its stock listed on the New York Stock Exchange under the ticker CRCL. Meanwhile, Tether has not issued a token. Even among the top ten protocols, only three have issued tokens, indicating that token issuance is not always the optimal strategy and depends on the protocol’s design.
Returning to the Lindy effect, in most cryptocurrency categories, the top two protocols hold the largest market share and dominate the space. This is even more pronounced in the stablecoin category, where Tether (USDT) and Circle (USDC) account for 84% of the entire market, followed by other participants such as Sky (USDS) and Ethena (USDe). In other areas, this pattern may be less obvious but is still evident—for example, in lending, the top two protocols by TVL (Aave and Morpho) hold 64% of the market share. The same pattern manifests across multiple categories, including prediction markets, yield, liquid staking, and restaking.
The Lindy effect is significant also due to the frequent hacker attacks on protocols in the crypto industry. This year alone, over $130 million has been lost from smart contracts, and over time, total losses have reached tens of billions of dollars. As a result, it is becoming increasingly difficult to entrust funds to any new protocol, as you cannot predict when it might be hacked. Therefore, the length of time a contract has been operational and the protocol’s existence are crucial, because the system has withstood the test of time and never failed. Even in cases where the system does not function as expected—such as recently with Aave’s CAPO oracle reporting an error—users are still refunded, because the protocol’s treasury can absorb the cost. Moreover, the longer a system has been in existence, the more it proves its resilience during market downturns. Top protocols continue to function as expected during market downturns, strongly indicating that everyone should adopt these battle-tested systems.
On the other hand, innovation is equally important, as market leaders continuously innovate and improve their products. For example, Morpho is bringing numerous institutions into on-chain finance through its vault architecture, enabling them to customize vaults to best meet their needs. Aave will also achieve this with its upcoming v4 upgrade, introducing the Spokes feature. Additionally, Aave allows institutions to borrow against tokenized RWA through its Horizon instance.
The next wave of cryptocurrency will be driven by institutions and "delegated finance"; protocols best positioned for both of these directions will experience the greatest growth.
Crypto doomsdayism
In Citrini's article "The Global Smart Crisis of 2028," they write:
The best way to continuously save users money—especially as trading begins between agents—is to eliminate fees. In machine-to-machine transactions, credit card processing fees of 2%-3% are clearly a target.
Agents began seeking payment methods faster and cheaper than credit cards. Most agents ultimately chose to use stablecoins via Solana or Ethereum L2 layers, enabling nearly instant settlement with transaction costs as low as a few cents.
This marks the beginning of our next chapter, one that moves beyond institutional adoption of cryptocurrency and focuses on agent-based finance and the broader application of AI agents to blockchain technology. This shift is already underway, with many protocols integrating AI agents to streamline user workflows and eliminate long-standing UX bottlenecks in crypto products. All these efforts fall under the category that emerged by the end of 2024: the convergence of DeFi and AI (DeFAI). While it has gained traction, it has also, like other aspects of crypto, become a “yield-first” narrative—but it nonetheless highlights how integrating more AI can significantly enhance the crypto experience.
By June 2028, the majority of crypto trades are executed by agents with no human involvement. Agents seek optimal returns based on users’ risk preferences. For non-crypto-native agents, blockchain is regarded as the ideal platform for executing most transactions due to its low cost, high efficiency, and verifiability. Over time, block space becomes cheaper, drastically reducing transaction costs. Cryptocurrencies are no longer complex—you simply give an AI agent a prompt and some funds, and it earns the best possible return for you. Cryptocurrency and blockchain have finally gone mainstream and are widely adopted. To improve overall capital efficiency, agents move capital away from protocols generating low yields or underutilized liquidity pools, concentrating it in a few venues that deliver the best returns. Most public chains and protocols are effectively obsolete due to lack of usage. The value of the tokens you invested in has dropped to its lowest point since your initial investment; you begin to regret not exiting back in 2026. Only a few tokens have risen in value—those that genuinely generate income and continuously accumulate value through revenue streams. Capital withdrawn from all other tokens flows into a small number of tokens with real performance and utility. Although the total market cap of the crypto market has increased compared to March 2026, most tokens have not benefited from institutional adoption or the growth of agent-driven finance. The dream of crypto technology has been realized—it is now widely used by the masses—but the evolution of tokens has diverged dramatically from what many expected.
It is March 2026; regardless of whether you believe the above scenario will come to pass, protocols with positive cash flow will sustain long-term growth, and their tokens will thrive.
Conclusion
For years, crypto protocols have focused on technical issues while neglecting product-market fit, which has been the largest risk investors have overlooked—but the market has finally recognized this. For years, the prices of most tokens have continued to decline, with all-time highs long behind them, making it clearer than ever that change is coming. The rise of certain tokens in 2026 reflects the importance of revenue data and token-first strategies, as investors begin shifting from speculation to investment.
Bad actors in the crypto space always profit from the "yield-first" narrative, while most participants walk away with loss-making portfolios, becoming liquidity exits—a highly unhealthy dynamic. With the influx of institutions, this awareness has deepened, as they are less interested in our assets and more focused on the battle-tested infrastructure we have built over the years.
As we continue to advance with institutional and AI-powered crypto infrastructure, this trend may become even stronger, as more investors seek “hard metrics” that convince them to buy tokens or stocks.
Related reading: Talking with a rug-pull expert: From getting rich quick to being left behind—do ordinary players still have a chance to strike it rich?