Dialogue with Blockworks Research Director: From Pump.fun to Believe, memes are becoming a new way to "monetize ideas."

CN
13 hours ago

The key is whether the idea can succeed, and Memecoin can price ideas more quickly.

Organized & Compiled by: Deep Tide TechFlow

Guest: Ryan Connor, Research Director at Blockworks

Host: Jack Kubinec

Podcast Source: Lightspeed

Original Title: The Solana Token Launchpad Coming for Venture Capital | Ryan Connor

Broadcast Date: May 16, 2025

Key Points Summary

We invited Ryan Connor to discuss the launch of the Believe app. This discussion covered several hot topics, including: whether Believe can become a new model for venture capital, the meme-ification of capital markets, how to find the best product-market fit (PMF), and the issue of order grabbing on the Pump platform.

Highlights of Insights

  • The norm in the tech industry is to act ahead of regulations. The key is to get ahead and make the regulatory framework adapt to you. If you wait, opportunities may be seized by others.

  • Believe issues actual scarce units of value, such as cash flow, which is a scarce resource.

  • Believe is not a social application; it is more like a financial application and does not require daily active users like social apps do.

  • Consumer crypto has become an objective reality; it is very specific and highly financialized. Consumer crypto has arrived, and as regulations become clearer, the prospects for this field will be even more promising.

  • Pump is very successful, and I am a loyal fan of it, but not everyone is willing to participate in that intense competitive model. There is a segment of the market looking for platforms with a more curated experience. If we can combine the innovative capabilities of independent developers with users' demand for high-quality experiences, Believe could become a very promising platform.

  • Believe can help independent app developers issue their tokens, which these developers typically launch alongside their independent app projects. The supply side has potential demand, hoping to achieve some equity profit or raise small capital; while the demand side clearly has a strong desire to gain venture-like returns from these small projects.

  • The key is whether the idea can succeed, and Memecoin is a very fast and immediate way to form capital. Founders can "vibe code" an app or propose an idea, and Memecoin can price ideas more quickly.

  • Whether it is suitable to issue tokens needs to be judged on a case-by-case basis. If you decide to issue tokens, you need to clarify how to manage community expectations and develop a long-term strategy to continuously create value for the tokens.

  • A qualified issuance platform needs to be able to issue units of value, support trading, and provide price charts and grid views for evaluating the return on investment (ROI) of different tokens. However, Zora removed these features, leading to a mismatch between Zora's functional design and the actual needs of a token issuance platform, making its product logic untenable.

  • We are currently in a phase where regulations are not yet fully clarified. This puts token issuers in an awkward position—they need to remain compliant while also knowing that issuing tokens benefits products, users, and token holders; they want to give tokens more functionality but are constrained.

  • Pump is an interesting "chicken game" and has found a good product-market fit (PMF), and people will continue to play this game. Believe attracts a segment of people who are more willing to participate in "growth-hacking" style startup investments, such as investing in those ultra-small market cap tokens with real futures.

  • Projects at the planning level of Believe could indeed run out. Poor market timing, insufficient users, or lack of capital and creativity could all lead to a reduction in projects.

  • From the developer's perspective, all these applications have their own issues, but ultimately what matters is the core functionality. For Pump.fun, that core functionality is enabling users to participate in the game.

  • Novelty is not the key to success or failure. Pump.fun did not create anything new; it simply packaged the strong demand for issuing and trading arbitrary assets in the crypto space over the past decade into an easy-to-use product, reducing the barrier to entry for users, and as a result, it succeeded.

  • Consumer applications are like this; they do not create demand but optimize the realization of existing demand.

Solana's Latest Project Incubation Platform

Jack:

We want to talk about the recently popular Believe app. This is a new Solana app, a new Memecoin launch platform that provides a "pseudo-equity" model for startups.

I tweeted this morning that of the 25 new tokens launched on Solana this week, 14 were issued through the Believe platform, while Pump.fun only accounted for 7. Comparing this to the Memecoin trading situation over the past year, this is clearly a significant change. According to data from the community-developed tracking tool Believe Screener, the 24-hour trading volume on the Believe platform reached $724 million in the past few months.

This is undoubtedly a huge success for the platform's founder, Ben Pasternak. He previously founded an alternative food startup that launched a product called Nuggs, a creative chicken nugget alternative. And now, he has launched the Believe platform.

Ryan, I think you might like Believe more than I do, so let's start with a simple question. Believe is a Memecoin launch platform, and there are many such launch platforms on the market. What makes this one special?

Ryan:

That's a good question. I often remind myself not to jump to conclusions that a startup will fail. More importantly, we need to distinguish between those "bad ideas" that are doomed to fail and those projects that could succeed if executed well.

I firmly believe Believe falls into the latter category, and I think this project can succeed.

It differs from other Memecoin launch platforms for two main reasons. First, its market strategy is very clear, targeting independent app developers and growth hackers. These individuals have historically found it difficult to profit from their projects. Believe provides them with a new monetization method through token issuance, which is a very attractive innovation for developers.

The second reason is that it has an additional filtering layer. Many past launch platforms have suffered from a lack of quality filtering, leading to poor user experiences. Believe's filtering layer is designed to meet the needs of users who want to participate in quality projects.

Pump is very successful, and I believe it will continue to be very successful. I am a loyal fan of it, but not everyone is willing to participate in that intense competitive model. There is a segment of the market looking for platforms with a more curated experience. If we can combine the innovative capabilities of independent developers with users' demand for high-quality experiences, Believe could become a very promising platform.

A New Model for Venture Capital?

Jack:

To briefly introduce, Believe operates in a way that anyone can issue tokens through it. You just need to mark a token at launch, input the token's name and code. The Believe team will filter all launched tokens, only displaying those related to interesting startups or app ideas.

I think the complexity here is that it is somewhat like equity investment, but not in the traditional sense of equity where you can invest in a company. I'm curious why projects like this should be financed with Memecoin? It seems to me that this is not a good match.

Ryan:

The interesting part here is that many independent developers cannot access traditional venture capital, and venture capital is not very interested in this group. Therefore, the appeal of this project lies in the fact that Believe can help these independent app developers issue their tokens, which they typically launch alongside their independent app projects.

So, what are independent app developers? They usually do not work at large companies or have their side projects, launching some apps each year to earn income, and may even achieve high returns. For example, Steam's data shows that most independent app developers earn less than a thousand dollars. Only the top 10% of high earners make between $150,000 and $180,000 per app, while the top 1% can reach $7 million. Therefore, from the perspective of independent developers, very few projects can obtain venture capital. As a result, they typically do not raise funds or engage in large-scale marketing but rely on precise marketing to make their apps popular.

I think this business model is acceptable. If you can create an app that generates hundreds of thousands or millions in revenue each year, even if it may not become the next Facebook, that is already a good outcome. You are unlikely to profit through equity value. If you study this seriously, you will find the future potential of this platform. This is a niche market that has been overlooked by the capital market. The supply side has potential demand, hoping to achieve some equity profit or raise small capital; while the demand side clearly has a strong desire to gain venture-like returns from these small projects. This demand has been evident in trading and the crypto market during COVID. Therefore, there is a clear demand on both the supply and demand sides, which excites me.

Jack:

I actually have many concerns about Believe, but I still want to give these projects a chance. It is much easier to be skeptical about everything, and that way you are right most of the time. But I want to ask, what is the fun in doing this?

My understanding of this is that it is a capital formation model in the "vibe coding" era. In the past, you had an idea, might raise funds for it, spend months coding, create a usable app, and then bring it to market. Interacting with users and understanding what they really want is quite difficult, so there is a lot of friction in actually creating the app. Now, if tools like Cursor become better, you can create a new app in minutes, so actual human capital, like the ability to create things, is no longer that important.

Now, the key is whether the idea can succeed, and Memecoin is a very fast and immediate way to form capital. Founders can "vibe code" an app or propose an idea, based on which they can launch a token on Believe to see if it resonates with the market. If the trading of this token goes well, then I will create the app and bring it to market to see how it performs. If the token trading for this idea does poorly, then you know perhaps you should shift your focus elsewhere. Therefore, I believe in this case, "vibe coding" becomes faster, and Memecoin can price ideas more quickly.

I also want to add that the behavior of Memecoin traders can often be very crazy, and their demands may leave founders feeling overwhelmed, ultimately leading to a situation where the costs outweigh the benefits. For example, something strange happened last week when Jeffy Yu pretended to have died, seemingly to escape the environment he was in. My guess is that once you accept funding from Memecoin investors, they may make many excessive demands, and these people may not be your ideal investors. So I can't help but wonder, while this approach is interesting, is it worth paying such a high price?

Ryan:

Indeed, issuing tokens brings many complex issues. We see in some deep projects that founders not only need to manage the business, sales, development team, and marketing but also have to additionally manage a community. And the members of this community may not be the mature or patient investor group you ideally envision.

Moreover, issuing tokens may also bring some negative signals. For instance, in the traditional stock market, if a company's stock price falls, people usually interpret this as a negative evaluation of the company's prospects. But in the cryptocurrency space, this is not necessarily the case. There are often unpredictable price fluctuations, such as drastic changes caused by high leverage. Therefore, issuing tokens may add a lot of extra pressure and burden on founders.

I believe whether it is suitable to issue tokens needs to be judged on a case-by-case basis. If you decide to issue tokens, you need to clarify how to manage community expectations and develop a long-term strategy to continuously create value for the tokens.

Currently, there are indeed some confusing aspects, such as the limited documentation related to the Believe platform, which is clearly still in a very early stage, having only been launched for a few months, and the latest version has only been running for a few days or weeks. So, how will this platform develop in the future? For example, will there be a unified standard for token distribution? Will the vesting period be standardized? I noticed that the team has publicly discussed these issues on the X platform and stated that they are pushing for these standardization measures, indicating that they are on the right track.

It is worth mentioning that the Believe team provides some best practice guidance and expectation management for token issuers, which is unique in the current market. They have clearly conducted very in-depth research on this issue. It is important to remember that Believe's main clients are practitioners from the traditional Web2 field, who are not familiar with cryptocurrency. Therefore, Believe needs to help these clients adapt through guidance and education. I see some signs that they are taking this task seriously.

The Meme-ification of Capital Markets

Jack:

JellyJelly surged overnight but then quickly fell back. This morning, I saw that its price, while still far below its previous high, has rebounded in recent days with the rise of the Believe app. Part of the reason may be that the market has placed overly high expectations on it, but in reality, its liquidity is very poor, and almost no one is willing to trade it. So I wonder, when it comes to Memecoin, is any publicity good publicity? Or could this approach negatively impact the performance of its video app after launch?

Ryan:

I don't know much about the specifics of JellyJelly. However, I believe that the close connection between the current crypto capital market and the Memecoin market does indeed bring some issues. The Memecoin market is often filled with "chicken games" (a concept in game theory), market manipulation, and irrational behavior from retail investors. Therefore, some strange phenomena can occur in the market. For example, when certain tokens are listed on centralized exchanges (CEX), their prices can experience extreme volatility. Just like when Bonk was listed on Coinbase, its price once soared to 50 cents; while that may not be so outrageous, such crazy price fluctuations are indeed common because many investors tend to buy first and understand the details later.

However, the market is gradually moving towards normalization. Extreme cases like Jelly are becoming less frequent, and the phenomenon of Layer 1 (L1) tokens trading at high premiums is also decreasing, while it is no longer common to achieve valuations in the billions simply by launching Layer 2 (L2) projects. Although the market cannot fully normalize in the short term, it is indeed moving in the right direction. Despite the many topics for discussion or complaints on the X platform during this process, overall, the market is gradually normalizing, which is a good thing.

Does the Zora Model Have Real-World Significance?

Jack:

**Ryan, I have a question for you. Previously, when you participated in a podcast, you criticized the Zora model. Your point at the time was that the value of the content itself tends to zero, so issuing tokens on Zora is meaningless. Logically, the value of most *startups* or ideas could also be said to tend toward zero. So why do you have a positive outlook on Believe while being pessimistic about Zora?**

Ryan:

My pessimism about Zora mainly stems from the nature of content. Content is highly abundant and non-scarce, so the value should not belong to content creators but to aggregators, as they are responsible for curating and integrating the content.

More importantly, Zora is essentially a token issuance platform (token launchpad). However, to cater to some user groups in the Ethereum community—who typically focus on the conflict between "big companies and small companies" and the monetization of creators—Zora removed key functions of a token issuance platform. A qualified issuance platform needs to be able to issue units of value, support trading, and provide price charts and grid views for evaluating the return on investment (ROI) of different tokens. However, Zora removed these functions in favor of emphasizing some visual features that, while aesthetically pleasing, are of no help in assessing financial returns. This led to a mismatch between Zora's functional design and the actual needs of a token issuance platform, making its product logic untenable.

Believe, on the other hand, is completely different. It issues actual scarce units of value, such as cash flow, which is a scarce resource. Looking ahead, we can imagine it will launch some form of revenue-sharing mechanism or rights for token holders. Most importantly, the Believe app provides price charts and other tools that allow users to assess the financial characteristics of the tokens. This is a true token issuance platform, equipped with the functions needed for trading and evaluating potential financial outcomes. Therefore, Believe is fundamentally different from Zora.

Believe is very different because it actually represents scarce units of value. Cash flow is scarce. If you look to the future, you can imagine some form of income sharing or rights relationship with token holders. And most importantly, I can view price charts on the Believe app to assess the relative financial characteristics of the tokens in that app. Therefore, Believe is an issuance platform equipped with the functions needed for trading and evaluating potential financial outcomes, which is the distinction between it and Zora.

Zora can fix these issues at any time if they wish. They can openly acknowledge that they are a token issuance platform without needing to masquerade as a social media tool or a creator monetization platform. They are fully capable of doing this, but so far, I have not seen them take action.

Opportunities and Challenges of Regulatory Arbitrage

Jack:

I want to add that Memecoin is a significant opportunity. For example, when visiting the Pump website, I don't like its content; most Memecoins seem ridiculous. But in the Believe app, all tokens are curated, making them seem more meaningful. When you click on a token, you find some interesting startup projects. I saw one project that is an AI entity that can access your social media history, which made me wonder why Grok couldn't do that. Maybe it's due to privacy issues, but I always want Grok to give suggestions based on my tweets, and it can't.

However, I also want to voice a concern for users in the crypto community: is this model somewhat unethical?

From a higher perspective, founders launch Memecoins to attract investors to invest in these tokens, while investors do not gain actual equity rights. We all know that the value of Memecoins will ultimately tend toward zero; the key lies in the timing of entry and exit. While Believe provides a counter-sabotage mechanism, I am not sure how effective it is, and insiders may still find ways to trade. I checked the creator documentation of Believe, which mentioned that if founders do not promise returns, ownership, or financial rights, and the tokens are clearly just for entertainment, they may be in a legal safe zone. This indeed circumvents legal risks, but it also seems to encourage ordinary investors to participate without giving them actual rights, while founders may profit significantly from it.

Ryan:

I think this needs to be analyzed on a case-by-case basis. If the capital market is free and open, and someone issues an arbitrary unit of value, I believe the market will price it reasonably.

But we are currently in a phase where regulations are not yet fully clarified. Although we can see a trend toward clearer regulations, they have not yet been fully implemented. Although legislators and regulatory bodies have sent signals, we do not yet have a clear legal framework. This puts token issuers in an awkward position—they need to remain compliant while also knowing that issuing tokens benefits products, users, and token holders. They want to give tokens more functionality but are constrained.

So what should they do? Wait for the day when regulations are clear? But the norm in the tech industry is to act ahead of regulations. Think about it: if Uber or Airbnb had fully complied with the rules from the start, these companies might not even exist. The key is to get ahead and make the regulatory framework adapt to you. If you wait, opportunities may be seized by others. For example, if Believe does not seize this opportunity to become a token issuance platform for developers, it may be replaced by others.

Of course, application developers who choose to wait may also fall behind. Therefore, I think this heavily depends on the specific situation. Clearly, each case needs to be evaluated individually, as there will indeed be people trying to exploit this open market for fraudulent activities. But I see that Believe provides curation tools, and the future of clearer regulations is no longer far off, which excites me. I am willing to give most issuers some trust.

Jack:

But the risk of doing so seems to be transferred to the users. Founders might be able to add some commitments, such as providing future equity access for holders of Memecoin. For example, Trump's Memecoin offered some securities-like features, such as a dinner opportunity for the first 220 holders. However, it seems that Trump himself may not face legal issues because of this. So, Believe might be fine too, but you are still transferring the legal risks to the users. They hold these tokens without any actual rights or benefits, yet they seem to accept this. Even with platforms like Zora repeatedly emphasizing that the tokens will have no value, there are still people willing to trade them.

Ryan:

If you visit Trump's token website, you'll find that they explicitly state these tokens are "purely collectibles." Many token issuers circumvent legal issues by providing some form of "threshold access," designing it as a membership card rather than a security. As long as the tokens do not involve future cash flow rights, there is still legal room for defense.

But ultimately, I wouldn't blame the founders and the platform. The problem lies in the lack of a clear legal framework over the past few years. In fact, regulatory clarity should have been achieved eight years ago, but it hasn't happened.

Jack:

When I browse Believe, I notice it has two tags: "New Projects" and "Featured Projects." As you mentioned, Believe's curation of Memecoins is indeed more advantageous than Pump, enhancing the user experience. But if that's the case, your core users might quickly exhaust the tokens they can invest in. And while the "vibe coding" era allows for the rapid creation of many mini-apps or projects, it still doesn't compare to Pump's simple model of "upload a JPEG and then issue a token." This model once drove Pump's rapid growth.

Does Believe have an upper limit? For example, will users feel that there are no more interesting ideas to invest in after three days and then return to Pump?

Ryan:

I don't think so. Every platform has its own positioning for early users. Pump is clearly an interesting "chicken game" and has found a good product-market fit (PMF), so people will continue to play this game. Believe attracts a segment of people who are more willing to invest in "growth hacking" style startup projects, such as investing in those with ultra-small market caps but real futures. Of course, there will be some overlap between these two markets, as many crypto users aim for 10x or even 1000x returns.

I believe these market caps should remain relatively low. For instance, a $20 million market cap is already a very good result for a reasonably valued project. This is an ideal state. Some tokens we currently see with market caps in the hundreds of millions or even billions are clearly too high for the niche market that Believe is targeting. Therefore, any downward adjustment of Believe's current token market cap, in my opinion, is healthy, as it aligns more with financial reality.

As for whether the projects curated by Believe will run out, I think it is indeed possible. For example, poor market timing, insufficient users, or a lack of capital and creativity could all lead to a decrease in projects. If Believe's popularity declines due to a lack of users, capital, or creativity, I wouldn't be surprised. But I welcome any adjustments to Believe's current token market cap, as they do seem to exceed actual levels. The market cap of these tokens should be closer to that of small startups rather than reaching higher valuations.

Has the Believe app found its product-market fit?

Jack:

I think most social media apps have the concept of "infinite scroll," algorithms that continuously push content to users, seemingly without end. But Believe has an endpoint, and I'm wondering if it might become something we forget about a week after discussing it, like Zora.

Ryan:

Believe is not a social application; it is more like a financial application. Therefore, it does not need daily active users like social applications do. I think the ideal situation is for users to be active weekly or monthly, rather than daily. If daily users are not the target, then there is no need for an "infinite scroll" design. This is also why many financial applications do not have this feature, as users should not be investing every day. In these very niche markets, doing so makes no sense.

Analysis of the Buying Frenzy on Pump.fun

Jack:

I'm not entirely clear on the specific operational mechanics, but I've heard that Believe is collaborating with Meteora to introduce a bonding curve with an anti-bot mechanism. In simple terms, before the token reaches a certain market cap, trading incurs higher fees, and once the market cap exceeds this threshold, the fees decrease.

This design addresses a core issue of Pump.fun: the vast majority of tokens are immediately bought up after release, even completed within the same block. If you're just an ordinary user looking to find a 1000x return opportunity by browsing Memecoins, you might miss out as soon as the token is released. If raising costs can prevent large-scale buying frenzies, that would indeed be a nice improvement. However, it seems no one is really discussing this issue. Do you think users actually care?

Ryan:

I think in these ultra-low market cap situations, users don't care much. As a user who has also observed many Memecoin user behaviors, my feeling is that the amounts everyone invests are generally small. This is more like a behavior similar to sports betting, especially on Pump.fun, where users are pursuing high returns akin to multiple bets. Therefore, paying extra fees is not important to them. Those participating in this game are usually mentally prepared to lose that money. So, these fees do not significantly impact their decision-making. However, this mechanism itself is still quite interesting. I believe that Believe does need a stronger "anti-buying frenzy mechanism," as these tokens may involve future equity attributes, and being bought up could lead to significant price fluctuations, which is detrimental to market health. On Pump.fun, the buying frenzy doesn't seem to have such a negative impact; in fact, it can be said to remain effective under this model.

Jack:

I think buying frenzies will still occur. Even if the fees are slightly higher, if a Silicon Valley growth hacker launches a token, someone could easily set up a script to complete the purchase the moment the token is released. This situation is likely to continue. While this mechanism may make buying frenzies less "random" than on Pump.fun, it is still feasible in the release of Memecoins. If you know that the creator you are targeting is likely to be recommended by the Believe platform and gain higher visibility and market performance, then even paying higher fees, targeting them can still be profitable. So I'm not sure if this mechanism can truly eliminate targeting behavior. Of course, as you mentioned, this mechanism may be necessary, but ultimately it is more about increasing the discussion heat on social media rather than significantly changing user behavior. Many tokens are indeed targeted or internally traded, which personally disappoints me regarding the use of these tokens. But the market's goal remains to find the next 1000x return opportunity before others do.

Ryan:

From a developer's perspective, all these applications have their own issues, even those that perform well, such as bugs, unethical behavior, and account abuse. But ultimately, what really matters is the core functionality. For Pump.fun, that core functionality is allowing users to participate in this game, as it is both fun and potentially financially rewarding. I believe developers will try to address these issues and find ways to mitigate them over time. This will be a cat-and-mouse game, but ultimately these problems will not hinder Pump.fun's success. Just as Mev (Maximum Extractable Value) does not hinder the success of blockchain or Nasdaq. These issues do exist, but in comparison, the benefits are clearly greater.

Summary and Future Outlook

Jack:

I want to make another point: Believe reminds me of Kickstarter. While we feel that Believe is novel in our discussions, it is more like a combination of Kickstarter and Memecoin. Its concept is to provide crowdfunding support for those who cannot access venture capital or other fundraising channels. This model has existed in Web2, but strangely it hasn't really entered Web3. And going back to the securities law issue we discussed earlier, Kickstarter typically promises some returns to supporters, such as T-shirts or funded board games, and they seem to operate quite well.

What is the essential difference between Kickstarter and Believe? If Kickstarter can operate legally under securities law, why can't Believe?

Ryan:

I don't know much about Kickstarter's specific model, but I suspect it may have some KYC (Know Your Customer) or accredited investor screening processes. Additionally, I think there may be an exemption in the law that allows for fundraising without following the usual securities law processes if the amount raised is low enough and the number of investors is limited. However, I'm not a legal expert; this is just my guess, but these could be the main differences between it and Believe.

As for novelty, I don't think it matters. For example, Facebook's name originally came from the university's Facebook directory; Snapchat's Stories feature was later copied by Instagram and renamed Reels, ultimately achieving success. So, novelty is not the key to success or failure. Pump.fun is a great example. It didn't create anything new; it simply observed the strong demand for the issuance and trading of arbitrary assets in the crypto space over the past decade and packaged those demands into an easy-to-use product. They just lowered the usage threshold for users, and as a result, they succeeded. Many consumer applications are like this; they do not create demand but optimize the realization of existing demand. So, while Believe may not be novel, it has the potential to stand out through some optimizations in details.

Jack:

Has the recent hype around Believe changed your view on the consumer crypto space? Can you summarize what this means for the field?

Ryan:

I believe consumer crypto has become an objective reality; it is very specific and highly financialized. I don't see it as a decentralized Instagram or Twitter, but rather as some interesting consumer applications adjacent to gambling, with unique innovative capabilities. Pump.fun is a typical example; since its launch, it has generated $700 million in revenue. More and more founders are focusing on this highly financialized experience. So consumer crypto has arrived, and as regulations gradually clarify, the prospects for this field will be even more exciting.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

Bitget:注册返10%, 送$100
Ad
Share To
APP

X

Telegram

Facebook

Reddit

CopyLink