To those who are confused during the blockchain transformation period
Some say Crypto is a Ponzi scheme, a bubble, a speculative game destined to return to zero.
Some say Web3 is a revolution, a paradigm shift, a new stage of civilization on top of technological continuation.
Two voices, a scene of narrative tearing.
Don't rush to take sides, let's talk about a more simple conclusion:
The underlying logic of business has not changed.
Whether it is Web2 from portals to Apps, or Web3 from issuing coins and telling stories to focusing on infrastructure, behind the prosperity, it is actually taking the same old road - only this time, the narrative is set on the protocol, and the capital is hidden in the code.
Looking back at the past ten years, the path of the Chinese Internet is very clear: concept-driven, financing runs ahead of user growth; subsidies pull traffic, capital drives growth; then layoffs, efficiency improvement, and profit; then platform transformation and technology restructuring. Today's Web3 is also stepping on a similar development rhythm.
Over the past year, the competition between project parties has evolved into a competition using TGE and Airdrop as ways to acquire users. No one wants to fall behind, but no one knows how long this "exchange user" competition will last.
So, I write to try to break down those seemingly chaotic narratives into several more traceable stages.
Let's take a look at how Web3 got to where it is today, and where it might be going, along the footsteps of history.
1. Review of the development stage of the Internet industry: from spreading money and expanding to industrial collaboration
I believe most people are familiar with this history:
The former Internet was a national carnival. Dozens of Apps competed to let you "get freebies" every day. One mobile phone number could eat, take a taxi, get a haircut, and get a massage, like celebrating the New Year.
Today's Internet, on the other hand, is a system engineering that has completed more than half of the journey: you know which platform to buy the cheapest things on, and which App is the most efficient in which scenario. The ecological pattern has long been finalized, and innovation is hidden in efficiency.
So I won't go into too much detail, just simply break down the four stages - reviewing these logics may help you better understand the path that Web3 is currently replicating.
1. Narrative-driven, mass innovation stage (before 2010)
That was an era when trends were defined by "nouns."
"Internet +" has become a panacea. Whether you are doing medical care, education, travel, or local life, as long as you put these three words on it, you can leverage hot money and attention. Entrepreneurs at that time were not in a hurry to make products, but first looked for tracks, created concepts, and wrote BPs. Investors were not chasing income curves, but whether they could tell a story that was "new enough, big enough, and easy to imagine."
O2O, social e-commerce, and the sharing economy. Under the rotation of rounds of nouns, project valuations have skyrocketed, and the pace of financing is dominated by the pace of narratives. The core assets are not users, not products, and not data, but a financing PPT that tells the story smoothly and matches the trend.
This is also an era where "whoever takes a stand first has a chance." Verifying the product and running through the model is the second step. First, tell the story to the forefront before you are qualified to enter the arena.
2. Money-burning expansion, traffic competition stage (2010–2018)
If the previous stage was about competing for attention with stories, this stage is about hard-robbing the market with subsidies.
From the Didi and Kuaidi taxi war to the Mobike and ofo bicycle melee, the entire industry fell into a highly consistent approach: use capital to exchange for scale, use price to exchange for habits, and use losses to exchange for entrances. Whoever can burn another round of financing will be qualified to continue to expand; whoever can get the next round of investment will be able to leave a position on the battlefield.
This is a period of putting "grabbing users" above everything else. Experience, efficiency, and product barriers are all relegated to the back row. The key is - who can be the first to become the default choice for users.
So subsidy wars intensified, and low prices became almost standard: taxi rides for less than 5 yuan, QR code bike rides for a penny, offline stores posting App QR codes, waiting for you to eat, get haircuts, and massages for free. It seems to be the popularization of services, but in reality, it is a capital-controlled battle for traffic.
It is not about who has a better product, but who can burn money more; it is not about who can solve the problem, but who can "occupy the land" faster.
In the long run, this also lays the foundation for the subsequent refined transformation - when users are bought, more effort must be spent to retain them; when growth is driven by external forces, it is destined to be difficult to self-close the loop.
3. Landing, refined operation stage (2018–2022)
When the story is told for too long, the industry will eventually return to a realistic question: "How to land after growth".
Since 2018, with the slowdown in the growth rate of mobile Internet users, the traffic dividend has gradually faded, and the cost of customer acquisition has continued to rise.
According to QuestMobile data, as of the end of September 2022, the number of monthly active users of China's mobile Internet is close to 1.2 billion, an increase of only about 100 million compared to 2018, taking nearly four and a half years, and the growth rate has slowed significantly. At the same time, the scale of online shopping users reached 850 million in 2022, accounting for nearly 80% of the total number of netizens, and the space for user growth is trending towards saturation.
At the same time, a large number of "story-type" projects driven by financing gradually withdrew. O2O and the sharing economy are the most concentrated areas of liquidation at this stage: projects such as Jiedian, Xiaolan Bicycle, and Wukong Travel have fallen one after another, behind which is a complete set of growth models that cannot be self-consistent and lack user loyalty are eliminated by the market.
But it is precisely in this ebb that a batch of projects that truly run out have emerged. They have a common feature: they are not short-term hotness stimulated by subsidies, but they have completed the closed-loop construction of the business model through real must-have scenarios and system capabilities.
For example, Meituan gradually built a complete service chain from order placement to fulfillment and from traffic to supply in the local life track, becoming a platform-type infrastructure; Pinduoduo quickly penetrated users' minds in the sinking e-commerce market with extreme supply chain integration and operational efficiency; social networking is firmly controlled by Tencent, e-commerce is fully occupied by Ali, and games are concentrated in Tencent and NetEase.
Their common ground is not "thinking further," but running more steadily and calculating more clearly—structurally completing the closed loop from traffic to value, truly living as a sustainable product system.
At this stage, growth is no longer the only goal. Whether growth can be transformed into structural retention and value precipitation is the real watershed that determines the life and death of the project. Extensive expansion is eliminated at this stage, and what is truly left behind are the system-type projects that can build a positive feedback mechanism between efficiency, products, and operations.
This also means that the era of narrative-driven is over, and the business logic must have the ability of "self-closing loop": retain users, support the model, and run through the structure.
4. The ecology is basically finalized, and technical changes seek opportunities (from 2023 to present)
After the leading projects come out, the survival problem has been solved by most projects, and the real differentiation has just begun.
The competition between platforms is no longer a battle for users, but a competition of ecological capabilities. As leading platforms gradually close off growth paths, the industry enters a cycle of structural stability, concentrated resources, and dominant collaborative capabilities. The real moat is not necessarily a leading function, but whether the internal circulation of the system is efficient, stable, and self-consistent.
This is a stage that belongs to system-type players. The pattern is basically finalized. If new variables want to break through, they can only find gaps and technical breakpoints on the edge of the structure.
At this stage, almost all high-frequency, must-have tracks have been marked out by giants. In the past, one could still compete for a position by "launching early and burning money quickly," but now, growth must be embedded in system capabilities. The platform logic has also been upgraded: from multi-product stacking to ecological flywheels, from single-point user expansion to organizational-level collaboration.
Tencent connects WeChat, Mini Programs, and advertising systems to build an internal circulation closed loop; Ali reorganizes Taotian, Cainiao, and Dingding to horizontally connect commercial links and try to find the efficiency leverage. Growth no longer relies on new users, but on the structural compounding brought about by the self-operation of the system.
As user paths, traffic entrances, and supply chain nodes are gradually controlled by a few leading platforms, the industrial structure begins to become closed, leaving less and less space for new entrants.
But it is precisely in this structurally converging environment that ByteDance has become an anomaly.
It did not try to compete for resource positions in the existing ecosystem, but took a shortcut, starting from the underlying technology, and reconstructed the content distribution logic with recommendation algorithms. In the context of mainstream platforms still relying on social relationship chains to dispatch traffic, ByteDance built a distribution system based on user behavior, thereby establishing its own user system and business closed loop.
This is not an improvement to the existing structure, but a technological breakthrough that bypasses the existing path and reconstructs the growth structure.
The emergence of ByteDance reminds us that even if the industry pattern tends to solidify, as long as there are structural breaks or technical gaps, new players may still appear. It's just that this time, the path is narrower, the pace is faster, and the requirements are higher.
Today's Web3 is in a similar critical interval.
2. The current stage of Web3: a "parallel mirror image" of Internet evolution logic
If the rise of Web2 was the industrial reorganization completed under the impetus of mobile Internet and platform models, then the starting point of Web3 is a system reconstruction based on decentralized finance, smart contracts, and on-chain infrastructure.
The difference is that Web2 constructs a strong connection between the platform and the user; while Web3 tries to break up and distribute "ownership" and reorganize new organizational structures and incentive mechanisms on the chain.
But the underlying driving force has not changed: from story-driven to capital-driven; from user competition to ecological flywheels, the path that Web3 has experienced is almost the same as Web2.
This is not a simple comparison, but a parallel reproduction of a path structure.
It's just that this time, it's burning token incentives; building modular protocols; and competing for TVL, active addresses, and airdrop points tables.
We can roughly divide the development of Web3 so far into four stages:
1. Concept-driven stage - Coin-driven: Story first, capital rushes in
If Web2's early stages relied on the "Internet +" story template, then Web3's opening statement is written in Ethereum's smart contracts.
In 2015, Ethereum went live, and the ERC-20 standard provided a unified interface for asset issuance, which also made "issuing coins" a basic capability that all developers can call. It did not change the essential logic of financing, but greatly reduced the technical threshold for issuance, circulation, and incentives, so that "technical narrative + contract deployment + token incentive" became the standard template for early Web3 entrepreneurship.
The outbreak at this stage comes more from the technical level - the blockchain empowers entrepreneurs in a standardized form for the first time, making asset issuance shift from a licensing system to open source.
No complete product, no mature users, as long as there is a white paper that can clearly explain the logic of the blockchain 1.0 era driven by blockchain technology, an attractive token model, and a runnable smart contract, the project can quickly complete the closed loop from "idea" to "financing."
Early innovation in Web3 is not because projects are so smart, but because the popularization of blockchain technology has brought about imagination in the blockchain 1.0 era.
Capital also quickly formed a "betting mechanism": whoever takes the new track first, who starts the market first, and who first promotes the narrative has the opportunity to obtain exponential returns.
This has spawned an "unprecedented capital efficiency": From 2017 to 2018, the ICO market experienced an unprecedented explosive growth, becoming one of the most controversial and iconic financing stages in the history of blockchain.
According to CoinDesk, in the first quarter of 2018, the total amount of ICO financing reached $6.3 billion, exceeding the total amount of financing in 2017 by 118%. Among them, Telegram's ICO raised $1.7 billion, and EOS raised $4.1 billion in one year, setting a historical record.
In the window period of "everything can be blockchain" - as long as you put on the label and build a narrative, even if the landing path is not clear, you can pre-spend the valuation imagination of the future. DeFi, NFT, Layer1, GameFi... each buzzword is a "window". Project valuations soar to hundreds of millions of dollars or even billions before the token is even circulated.
This is an opportunity to enter the capital market with a low threshold, and a relatively clear exit path has gradually formed: early positioning in the primary market, stimulating emotions through narratives and liquidity in the secondary market, and then completing the exit during the window period.
Under this mechanism, the core of pricing is not how much the project has done, but who takes the position earlier, who is better at creating emotions, and who controls the window for releasing liquidity.
It is essentially a typical characteristic of the early new paradigm of blockchain - the infrastructure has just landed, the cognitive space has not yet been filled, and the price often forms before the product itself.
The "concept dividend period" of Web3 came into being: value is defined by narrative, and exit is driven by emotion. Projects and capital seek certainty in a liquidity-driven structure.
2. Money-burning expansion stage - projects gather, and the user competition war is fully launched
Everything changes, starting with the "most expensive thank you letter in history."
In 2020, Uniswap airdropped 400 UNI tokens to early users. Each airdrop was worth about $1,200 at the time. The project party called it a "giveback," but the industry understood another word: the optimal solution for cold starts.
At first, it was just a gesture of "giving back to the community," but it unintentionally opened Pandora's box in the industry: the project party found that issuing coins can exchange for loyalty, traffic, and even an illusion of a community.
Airdrops have changed from options to standard features.
Since then, the project parties have been enlightened, and almost all new projects have taken "airdrop expectations" as the default module for cold starts. In order to show the market their prosperous ecology, they use tokens to buy user behavior, and the three-piece suit of points system, interactive tasks, and snapshot has become an option.
A large number of projects have fallen into a growth illusion of "incentive-driven rather than value-driven".
On-chain data soared, and the founders were immersed in the illusion of "success": millions of users and hundreds of thousands of daily active users before TGE; the scene cooled down instantly after TGE.
I still remember in 2024, Fusionist's on-chain DAU once exceeded 40,000, but immediately after Binance announced its listing, on-chain activity almost returned to zero.
I am not denying the airdrop thing. The essence of an airdrop is to buy user behavior, which is an effective way to attract new users under the premise of not consuming financing funds during cold starts. But its marginal effect is rapidly diminishing. A large number of projects have fallen into a formulaic cycle of airdrop user acquisition, and after user acquisition, whether your business scenario and product capabilities can have retention capabilities is the real return of value and the only correct solution for the project party to survive. (Note: Projects that survive by manipulating the secondary market with funds are not within the scope of this discussion)
In the end, bribing users to buy is not the core of growth. Without a commercial foundation built on even the scenario, the airdrop ultimately consumes the interests of the project party or the user. When the commercial model is not a closed loop, the token becomes the only reason for users to act. And once the TGE is completed and the rewards are terminated, users will naturally turn around and leave.
3. Commercial verification stage - real scenarios, narrative verification
I often advise project parties to think clearly about one thing before spreading money:
What problem are you solving for which scenario? Who are the most critical contributors? After TGE, will this scenario still be valid, and will anyone really stay and use it?
Many project parties told me that they can quickly complete user growth through token incentives. I will ask, "What then?"
Usually at this time, the project party will be silent for a while, smile, "Oh..."
And then, there was nothing after that.
If you just want to exchange a wave of interactions by "issuing incentives," then you might as well directly issue Meme. At least everyone knows this is an emotional game, and there is no need to bear the expectation of staying.
Finally, everyone began to look back: what kind of structure did these traffic, interactions, and coins spread out lead to? At the end of spreading money, I turned out to be a clown 🤡.
So the keywords at this stage have become: use scenarios, user needs, and product structure. Only by relying on real scenarios and clear structures can we find a growth path that belongs to us.
To be honest, I personally don't like Kaito's business logic - it is more like an extreme form of "bribery culture." Behind it is a high degree of utilization of incentive mechanisms, and it can even be said to be a repackaging of the relationship between platform and content.
But it is undeniable that Kaito succeeded. It is an actual commercial scenario. The expectation before TGE became the accelerator for the project to occupy the market, and the music continued after TGE. Because Kaito provides a business logic that allows KOLs to expose projects, the wool comes from the pigs, and key people still remain on the Kaito platform itself.
Although many KOLs may be clear that this logic will eventually backfire on themselves, in a market of structural opportunism, "strategic compliance" has become the most rational choice.
At the same time, I am also very pleased to see that more and more projects are starting to build around real scenarios, whether it is trading, DeFi, or basic capabilities such as identity systems.
Those teams that choose the right direction at the right time and polish real products are gradually taking root and sprouting and building their own industrialization path through the positive cycle capabilities of vertical scenarios - from use to retention, and from retention to monetization.
The most typical example is exchange-type products: they transform high-frequency needs into structural traffic, and then complete the closed loop through asset, wallet, and ecological linkage, and walk out of the "structural evolution line" in Web3 projects.
4. Structural precipitation period - platform finalization, variable contraction
The real positive cycle business scenario is the entry ticket for the project to obtain the industry's right to speak.
For example, Binance starts with transactions, gradually connects liquidity, asset issuance, on-chain expansion, and traffic entrances, forming a full-process scheduling system from off-chain to on-chain; Solana uses light assets to ignite and underlying performance to undertake, precipitating a feedback structure of community, developers, and tool systems.
This is a cycle in which the industry shifts from project experimentation to structural precipitation - no longer competing for speed, but starting to compete for the completeness of the system.
But this does not mean that new projects have lost the opportunity to break through. The projects that can truly run out are not the loudest or the most widely spread narratives, but those that can structurally "fill in the gaps" or "reconstruct" the model.
Remember ByteDance in the era of mobile Internet?
I believe that in the post-blockchain era, a new cycle driven by AI is coming. There will definitely be projects like ByteDance that can quickly run through the structure under the right incision with AI and complete industrial breakthroughs and self-closing loops.
The platformization stage of Web2 left behind giants and flywheels, and also left behind gap breakers like ByteDance; the structural period of Web3 may also give birth to the next variable project that uses the correct structure to "kill from the edge".
If it is basic construction, it should be the basic construction created for the native AI era to promote the development of technical products in this era, just like the mission of Ethereum in the blockchain 1.0 era mentioned above;
If it is DAPP, then it must be an application that uses AI to break the original user threshold (the web3 user threshold is too high) and break the original business order.
If someone asks me, how will the future of web3 develop?
I would say: "Just like the Internet of Everything, its true potential lies in the post-blockchain era, reconstructing usage paths, reducing collaboration barriers, and spawning a batch of products and systems that can truly run."