Written by: Gagra Ventures

Compiled by: Xiaobai Navigation coderworld

Recently, Gagra Ventures published an article titled "The New "Value Stack", which mainly discusses the evolution of the value chain in Web3 and distributed technology businesses. It explores how decentralization and autonomous agents affect the global technology sector, and how Web3 protocols and applications form a globally accessible public good. The article also discusses how companies such as Amazon achieve commercial success by controlling different levels of the value stack, and how these principles apply to the Web3 environment. Xiaobai Navigation compiled the full article.

background

In traditional tech, it is well known that companies that target the “last mile” to the end customer often capture the most value in the product’s value chain. The last mile is also the most competitive, so for risk-averse founders, it still makes sense to extend along the value chain and produce the intermediate components or services that make up the final product.

In the distributed technology business field of Web3, since middlemen are ideally minimized, all decentralized, and service automated, it is easy to draw the following conclusions:The value chain is invertedThe earliest infrastructures to run Web3 applications were behemoths like Bitcoin and Ethereum, each with its own paradigm (such as “Internet money” and “Internet computer”) for applications to take advantage of their openness. As a result, investors realized thatThe protocols themselves may capture more value than the applications that run on them.

This thinking has led to most of the capital flowing into Web3 infrastructure, such as the so-called first layer, second layer and even third layer (or specific application chains/rollups). However, the vision behind it has evolved from being monolithic to being more modular, resulting in more infrastructure startups to fund.

Since then, we have also seen the emergence of a number of applications that have gained significant benefits by using Ethereum to reduce deployment and distribution costs, while also benefiting from issuing tokens that do not necessarily capture the value generated by the product.Token(mostly for regulatory reasons, which is temporary) have gained significant upside. Uniswap is a good example, even though they have just started charging fees. While its valuation was much lower than Ethereum at its peak, the speed of development and market launch of this application was significantly slower than competing Web2 online trading applications, while it benefited from outsourcing most of its operating costs to Ethereum validators and maintainers. This is a good business model.

So as these applications emerged, the narrative shifted for a while, with a lot of public and private market investment flowing into new super app contenders in Web3, primarily in gaming and DeFi applications (such as liquidity staking).

We have been thinking about which thinking framework can present the best Web3 business model with the most value. This gave rise to our concept of "Value Stack", which currently guides our investment decisions and portfolio composition.

The forces shaping the new value stack

We believe that two core trends will define the global technology industry. Starting with software and the Internet (as the transition between offline and onlineSafetyAs the gap closes, they will become synonymous), and we expect these trends to then trickle down to the physical world.

These trends are:

• Decentralized, publicBlockchainand similarBlockchainIt is driven by the proliferation of distributed systems (deterministic, trustless, censorship-resistant). It is first and foremost a system around any asset, including autonomous programs. We think something likeBlockchainThe technology is a “lasso” for AI, just as the legal system contained real-world violence when humans turned to agriculture. This will lead to everything (first digital assets, then physical assets) coming online. In the near future, as scalability and bandwidth bottlenecks are inevitably resolved, blockchain-like cryptographic systems will provide betterSafetysex, while keeping everyone and everything connected, so no more on-premises setup and downloadable clients;

• Autonomous agents will emerge in both the digital and physical worlds, driven by AI and robotics. Machines will account for the vast majority of Internet traffic and usage. We believe that the ultimate form of consumer applications will be a combination of highly customizable AI agent interfaces with mundane objects; while more creative and engaging applications may end up being gamified experiences for human users while still maintaining a high degree of AI components.

To reiterate: We believe thatMost business processes and our user experience will be automated, while the parameters and preferences we set for it remain within our verifiable control. Additionally, scalability limitations and complexity are current bottlenecks for decentralized analogs of the best Web2 products and will be solved in the next 3-5 years, so we assume this will be solved by the time the argument starts to fully play out.

This also means that we will end up with a very different value chain because these technologies are global in nature. Everything will be online and connected. The main conclusion here is that the Internet value stack will eventually become homogenized globally.

In our view, the winning applications and protocols will be global public goods that are managed and accessed through distributed networks like public blockchains. There may be jurisdiction-specific rules and restrictions for interacting with them, but most of the time, it's still the same stack.

What is the Value Stack

Now, let’s explain what we mean by the “Value Stack” with a simple example.

Amazon could be thought of as an infinite store shelf, first books, then everything else—the last thing they had to grabXiaobai Navigationkm. They captured consumers by offering and expanding their products beyond their original book offerings. Over time, Amazon transitioned from being a reselling publisher to being its own publisher, creating a platform through Kindle and later other content platforms through Amazon Prime. But what really allowed them to reduce costs and eventually become profitable was moving down the technology stack and the value stack and selling access to a backend that was idle 95% of the year except during the Christmas shopping period. This gave rise to their cloud platform, AWS. In addition, in order to improve the customer experience and develop new customers (who would otherwise use physical stores), they had to go further into the value creation chain, build an exceptionally fast delivery service, and eventually own the largest fleet of private jets and drones in the United States. They also work with manufacturers to sell some of their most popular products under their own brand to keep prices attractively low and maintain delivery speeds.

This is roughly how an e-commerce platform transforms itself into a cloud computing, infrastructure, and logistics company — by moving vertically down the value stack and owning it.. From the last mile at the top all the way to manufacturing to squeeze out all the profit margins and satisfy customers. Of course, they also rely on common infrastructure like TCP/IP on the Internet or concrete roads in the physical world, which also constitute the value chain. But to maintain the dominance of its business, Amazon needs to own the vertical stack, and it must do so gradually after first locking in the major parts that bring it real growth factors. Similar situations exist in other business giants to varying degrees.

It is a more natural progression for a company to go deep into the value chain that makes up its primary product, rather than expanding into adjacent or entirely new markets, although they still always do so, but with varying degrees of success. They always have a better chance of fully owning the vertical value stack of their business, rather than dominating horizontally (i.e. across multiple markets/jurisdictions/platforms). Owning part of the value chain creates economies of scale that insulate against competition. Expanding horizontally into new markets is riskier and less clear.

No matter how strong Amazon’s vertical value stack is in the US and other select regions, they can’t own it globally, so they either rely on partners with better distribution in a particular region, or face competition from businesses that own similar types of vertical stacks in other regions and dominate there – like Alibaba in China. They may never be exact copies of each other, for example Alibaba doesn’t own any warehouses or logistics operations, but owns other parts of its value chain (financial services through Alipay) to achieve dominance in its local market.

Therefore, the winning business model in traditional technology is to grow the value stack vertically from the top down to maximize profit margins and avoid competition. But this still does not prevent these businesses from being disrupted occasionally.

The Value Stack in Web3

We believe thatThe winning Web3 protocols and applications will become globally accessible public goods for their specific use cases, because the broadest access generally creates a power law dynamic (1 or 2 winning solutions used by the majority). These distributed products are often networks of service providers + users (e.g. Bitcoin) or just users (e.g. Uniswap). They are also inherently censorship-resistant, open to operate 24/7, and equally open to anyone to contribute to or build on top of them. Given this environment, there is no reason why users and contributors wouldn’t consider the winning solution the preferred product for a particular use case. The winners are likely to be those that capture the most mindshare, liquidity, and other network effects.

This also means that Web3 is made up of multiple applications and protocols, each of which owns a part of this decentralized global value creation chain, but no one application owns it completely. Interoperability and openness require a certain level of specialization (this depends on the field). In a world where everything is open source, you need to be the absolute best or have the biggest network effect. Furthermore, if you reduce interoperability features to force it, users will simply choose a different solution that doesn't do it because that's where the value is for them. In the end, you can of course still build your own complete value stack, but the incentives already favor building on top of existing solutions rather than creating everything on yours. Make your protocol cheap to set up and distribute it to the widest possible user base (on the blockchain)walletThe benefits of open source technology are so numerous that the more contributors it has, the stronger it becomes. This is an economic, cultural, and even social movement that is difficult to reverse.

Expanding the metaphor of the value stack as composed of technology slices, from a vertical perspective, the slice of value captured by a specific application or protocol may appear smaller/thinner (limited vertical value capture) for the reasons mentioned above. However, the nature of the global market created by Web3 allows these businesses to serve more users than regional winners such as Amazon or Alibaba. Therefore, from a horizontal perspective, Web3 may have greater potential value than similar products in Web2.

We see Web3 winners growing horizontally in their respective categories, not vertically. The global value stack is a single homogenous set of layers stacked vertically into a large technology stack, with each protocol and application being a single slice of that layered body.

What does the winner in the “storage” space look like in such a scenario? It’s the protocol that hosts data for most of the world’s internet, i.e. in China and the US and everywhere else, which means it takes the storage portion of AWS, Alibaba Cloud, Azure, Tencent Cloud, Google Cloud, Hetzner, and all the other cloud and current on-premises storage businesses. Now multiply that by the ever-expanding storage needs of a digitized world where AI generates and consumes orders of magnitude more data than we do today, and you get a rough idea of the scale that such a layer of the value stack could occupy.

The key question is, given the open source nature of these technologies, how do you ensure you win in network effects while also capturing the majority of the value they create? We strongly believe we are at the beginning of figuring out the optimal economics for such systems. Experimentation with digital assets representing access/ownership of these platforms is just beginning to break through. TheseToken快速实现流动性的路径既是礼物也是诅咒，因为它激励人们尽快推出这些Token，并且只根据早期迭代的试错进行微小改进。然而，对于价格持续下跌的代币的产品来说，很难逆转方向。这就是为什么这些代币经济学的失败可能会给即使是非常成功的产品的建设者带来沉重的代价。

We strongly believe that the big era of token design is coming, as well as a testbed for the most complex and cutting-edge economic theories. We have heard from economists working with our portfolio companies that most people’s career paths in traditional economics are boring because they have little real-world output. In contrast, in token economics, they can implement and continuously adjust these closed-loop network economies.

How we put our framework into practice

As an investor, you want the winner or at least the second largest protocol/application serving their respective stack. We believe that protocols further down the value chain (i.e. the very back end of the internet) will be more like “winner-take-all” commodities. Whereas the application layer may have multiple simultaneous winners serving the needs of different user groups but still being ubiquitous in their respective market segments.

Furthermore, we don’t know which layer of the stack will be the winners capturing the most value, and we expect demand for each service/protocol will be cyclical, but more profitable the higher you go in the stack (think of necessities vs. non-necessities in an economic cycle), including middleware (e.g., the rise of gaming is more beneficial to certain middleware than the rise of productivity software). So, you want to own as many winners in each category as possible, and across the stack, to get defensiveness in your portfolio.

Additionally, while we don’t know how valuable each category might be, by heuristic one might pick the larger categories first. We prefer projects in areas that appear to be less competitive but that make sense both from a first principles and founder talent perspective.

The key to ensuring you win the race against the winners or strong competitors is to ensure they have the defensibility to build a moat around them. We believe that in a world of interconnectedness, this means firstly network effects and secondly strong value capture mechanisms. Therefore, as an investor, we primarily work with portfolio companies to improve their composability with other portfolio companies and the broader space, while also designing the most appropriate value capture mechanisms.

One thing to note about network effects is that over the past 20 years of Internet growth, we’ve learnedTechnology is a popularity contest, and it’s not always the best solutions that get the most adoptionHowever, we encourage portfolio teams to have the freedom to not compromise when the technology itself is not consumer-facing. It will become even more important in a world where integration decisions are driven by AI, which removes human bias and social thinking when choosing the best technology. This adds a whole new layer of questions to think about, but that is not the question to consider now.

The article comes from the Internet:Web3’s new value chain: Successful projects have global market value and win a wider range of users

Related recommendations: Web3 + AI A panoramic review of the track: with more than 130 projects, what other undiscovered treasures are there?

Web3+ AI There are at least 140 concept projects, 84 of which have issued tokens, mainly focusing on infrastructure, NFT game metaverse, data, computing, chatbots, etc. Written by: Peng SUN, Foresight News Graphics by: aididiao.eth, You Sheng, Foresight Ne…