EricSJ

EricSJ

推特@sjbtc9丨内容输出:美股相关、二级市场技术分析、Web3项目观点

15Following
60followers

Feed

EricSJ
EricSJ
Holding the best hand of RWA, why still can't make money?
"$5 billion on-chain, where did the profits go?" Securitize $SECZ.US, the first publicly listed company in the crypto industry specializing in tokenization, disclosed its first quarterly financial report since listing early this morning. According to the report, the scale of on-chain assets issued through its platform has nearly reached $5 billion, but quarterly revenue dropped from $19.48 million in Q1 to $14.44 million, a quarter-on-quarter decline of about 26%; adjusted EBITDA (earnings before interest, taxes, depreciation, and amortization) for Q2 also shifted from a profit of $830,000 to a loss of $5.46 million. Adjusted EBITDA can be simply understood as: temporarily excluding interest, taxes, depreciation, amortization, and some one-time expenses, how much the company's core business roughly earned. After the financial report was released, Securitize's stock price fell nearly 30% in after-hours trading (the chart below is the after-hours trend from @BITstocks_CN). In recent years, Securitize has almost been dealt the best hand in the RWA industry, with uniquely abundant institutional resources. From an industry position perspective, Securitize is an undisputed leading player in the RWA sector. It is the tokenization service provider for BlackRock's tokenized money market fund BUIDL, and also collaborates with asset management firms such as Apollo, KKR, Hamilton Lane, and VanEck; it holds broker-dealer, alternative trading system, transfer agent, investment advisor, and other licenses in the U.S.
EricSJ
EricSJ
Highly recommend everyone to pay attention to Cloudflare $NET. This company currently has a market value in the hundreds of billions online. I only realized how promising this company could be after recently building my own website. If you are like me now, rarely actively opening websites and mostly letting AI call them for you, then you will understand what I’m about to say. Because Cloudflare has embedded one of its business scenarios into the AI-to-website intermediate link. ➠ In the past: I have a need → open website → website returns data ➠ Now it is becoming: I have a need → ask AI → AI accesses website/API → fetches and organizes data → returns to me What it’s doing isn’t really creating a brand-new business. After breaking it down these days, I found that almost all its product launches revolve around this transmission path: ➠ Customer → Cloudflare → origin site/application ← Cloudflare ← external user or agent request Including future business logic, it can be said that it is evolving from "hosting the internet" to "hosting the AI-empowered internet." As of 2026 Q2, about 23% of websites worldwide already use Cloudflare as a reverse proxy (Figure 1). In other words, Cloudflare already stands at the request entry point for a significant portion of websites. Moreover, looking at data from the past seven quarters, it has preliminarily validated the growth capability of its original business segment (Figure 2). ➠ Quarterly revenue grew from $460 million to $696 million; ➠ Year-over-year growth rate increased from 27% to 36%; ➠ DBNR rebounded from 111% to 120%; ➠ Number of large customers with annual revenue over $100,000 increased from 3,497 to 4,698. This original business isn’t really the sexy kind; it’s the old story. The new story (AI requesting website access chain) is what makes me feel the space is huge. Because one person cannot continuously open a dozen websites, repeatedly call dozens of interfaces, and perform tasks around the clock at the same time. But this behavior is exactly what AI can and will do. According to Cloudflare’s own network observation, non-human traffic has exceeded 50% for the first time. Although this includes not only AI agents but also traditional crawlers, automation tools, and malicious bots, the direction is clear: machines are becoming important visitors to the internet. We also need to pay attention to the supply side that the platform itself is cutting: ➠ In 2026 Q2, Cloudflare added nearly 2 million new developers in a single quarter, while the entire year of 2025 added about 1.5 million developers. Currently, its active developers have exceeded 7.4 million (Figure 3). This means more and more machines are "passing through Cloudflare," and more and more people are "building applications on Cloudflare." More developers → more applications and agents → more computing and network requests → more calls to Cloudflare products → more usage and revenue. This is where Cloudflare closes the loop of imagination: In the past, it waited for websites to be built and then brought in traffic. In the future, it hopes applications will be born on Cloudflare. Of course, as I said, these are future matters and do not equal realized revenue yet. Plus, Cloudflare has a large free tier, and the developer platform has not disclosed independent revenue, payment ratios, or average revenue. But undoubtedly, this is a company worth tracking long-term. The original post also included opinions on whether it’s worth entering now and how to enter, which I won’t repeat here, just picking some viewpoints as a supplement.
EricSJ
EricSJ
I built a website and just understood that Cloudflare is already worth hundreds of billions of dollars
$Cloudflare(NET)$ This company has already surpassed $100 billion, but it is not a stock frequently discussed in the market. This company came into my view not because it rose over 20% in the past month, nor because any recent financial report data was particularly outstanding. Rather, it was when I was building my website recently—from construction, operation to GEO design—I hardly left the Cloudflare platform. It was at this time that I first fully understood what Cloudflare actually does. After learning about it during this period, I believe its business and future market space will be completely reshaped with the support of AI. Many people's most intuitive perception of Cloudflare is still as a CDN + cybersecurity company, but now and in the future, it has truly entered the middle layer of AI and website interaction. The original business still has a large market, but it is not very attractive; the market space can be estimated. However, the latter is growing exponentially, and the future peak is nowhere in sight. This content is my personal research record, and I also hope that through this perspective, it can help you critically understand Cloudflare's current and future business and strategic mainline. 1. First, let's talk about the existing business. What Cloudflare is doing now, if you are not very familiar with website building, and this article explains based on products, it might be abstract, just like the list shown in the picture below: So we can simply imagine a scenario like this:
EricSJ
EricSJ
I built a website and just understood that Cloudflare is already worth hundreds of billions of dollars
$Cloudflare(NET)$ This company has already surpassed $100 billion, but it is not a stock frequently discussed in the market. This company came into my view not because it rose over 20% in the past month, nor because any recent financial report data was particularly outstanding. Rather, it was when I was building my website recently—from construction, operation to GEO design—I hardly left the Cloudflare platform. It was at this time that I first fully understood what Cloudflare actually does. After learning about it during this period, I believe its business and future market space will be completely reshaped with the support of AI. Many people's most intuitive perception of Cloudflare is still as a CDN + cybersecurity company, but now and in the future, it has truly entered the middle layer of AI and website interaction. The original business still has a large market, but it is not very attractive; the market space can be estimated. However, the latter is growing exponentially, and the future peak is nowhere in sight. This content is my personal research record, and I also hope that through this perspective, it can help you critically understand Cloudflare's current and future business and strategic mainline. 1. First, let's talk about the existing business. What Cloudflare is doing now, if you are not very familiar with website building, and this article explains based on products, it might be abstract, just like the list shown in the picture below: So we can simply imagine a scenario like this:
EricSJ
EricSJ
For those who trade on their own, having a fixed trading and analysis cycle is extremely important. For example, if you are doing short-term trading and use the daily chart to analyze the overall trend direction, then stick to the daily chart instead of switching between 4H and 1H charts. The timeframes influence each other; if you see a bullish trend on the higher timeframe, the lower timeframe might show a bearish signal. You should enter trades following the major trend and hold positions against the minor trend~ #交易之声:你的经验值得被听到
EricSJ
EricSJ
$SOL is temporarily facing resistance in the 15-minute timeframe. For intraday trades, try to catch the retracement to see if it can break below this trendline and reverse direction accordingly. If lucky, the position can be closed before sleep tonight, with a stop loss set directly at the previous high of 76.66. If the bet is placed, count it as settled; if not, the stop loss can be pushed to hold until the next day
EricSJ
EricSJ
Nearly $400 billion left by Buffett has started to be spent #伯克希尔结束净卖出,重启大额配置 What is this telling the market? This might be the most important Berkshire Hathaway financial report to seriously look at in the past three years $BRKB Because this data shows Berkshire has become a net buyer of stocks again, ending the previous 14 consecutive quarters of net stock selling In recent years, Berkshire has been selling in various ways without repurchasing Meanwhile, cash kept piling up, eventually reaching nearly $400 billion The latest report shows that in Q2, Berkshire bought about $23.5 billion in stocks and sold about $3.7 billion, with a net stock purchase of nearly $19.8 billion in a single quarter What I find most worth mentioning is the buyback action ➠ In the first half of the year, Berkshire repurchased about $4.8 billion of its own stock, with the vast majority occurring in Q2 ($4.5 billion) Here, some background is needed Starting in 2026, Greg Abel will officially take over as Berkshire CEO, while Buffett will continue as chairman; Berkshire's current buyback policy has clearly placed capital allocation decisions in the CEO's hands, requiring consultation with the chairman In other words, after consultation, they will only execute buybacks if the buyback price is below a conservatively estimated intrinsic value So at least one message from Q2 is that Berkshire's top management believes the price is undervalued and worth buying The interesting part is that Berkshire itself is a huge asset portfolio Buying back Berkshire stock is essentially no different in nature from buying Apple or related portfolio holdings Because Berkshire is not a single-business company but a basket of capital allocation composed of insurance float, operating businesses, and stock investments This is more direct than any interview statement like "I am confident about the upcoming market" _____________________________________ However, Q2 net stock purchases only account for about 5% of the broad cash at the end of the period So it is still too early to conclude that the bearish trend is over based on just this quarter We can only say that in recent years, this cash has been more like a card placed on the table Berkshire chose not to play the card, but in Q2, part of this card was finally played But the card is far from fully played At least playing the card is a good sign; we can observe more, and I won’t elaborate on more financial data Just throwing out this one viewpoint
EricSJ
EricSJ
The fee scale of Q2 Pump fun$PUMP is about $212 million. Breaking down this part: In the Bonding Curve phase, the actual fee captured by the Pump Fun protocol is $62.06 million, which corresponds to a fee rate of 0.95%. That means for every $1 million in transaction volume, Pump Fun can earn about $9,500 in fee revenue. At this stage, when users buy or sell Meme, Pump Fun receives a platform share from the transaction fees. However, from the quarterly trend, this income peaked in 2025 Q3 and then entered a downward phase overall. Although there was a brief rebound in 2026 Q1, it fell again in Q2, and the overall trend remains downward. Pump Fun has basically said goodbye to the rapid growth phase brought by the early Solana Meme boom. More precisely, the Bonding Curve business heavily depends on the issuance of new Memes and early trading enthusiasm. Therefore, when attention to Memes on Solana declines, or some funds and users are diverted to other ecosystems, this part of the income is often the first to be affected.
EricSJ
EricSJ
The biggest source of fees for Pump Fun is no longer "posting Meme"
Pump Fun Q2 still maintains quite strong profitability, but after several quarters of verification, its revenue structure has undergone a significant transformation. $PUMP is like how many people think Pepsi is just about selling cola, but in reality, its other beverage and snack businesses already account for more than half of total revenue. Pump Fun also experiences this kind of counter-market consensus situation. Initially relying on Launchpad fees, its revenue structure gradually shifted from capturing income solely from the issuance phase to expanding across the entire token lifecycle. This article will analyze its on-chain revenue situation in Q2 and share insights. First, a simple numerical conclusion before diving into details: From Q2 data, although the overall fee scale has declined compared to the peak period, it still exceeds $200 million, and the largest fee source is not the well-known Bonding Curve, but rather PumpSwap. This is also one of the least known aspects of Pump. 1. Fee scale (revenue item) In Q2, Pump Fun's fee scale was about $212 million. Breaking this down: In the Bonding Curve phase, the actual fees captured by the Pump Fun protocol were $62.06 million, representing a fee rate of 0.95%; that is, for every $1 million in transaction volume, Pump Fun can earn about $9,500 in fee income. At this stage, users buying
EricSJ
EricSJ
A little-known fact: Pump fun's DEX is also the largest DEX on Solana This means that the Pump Fun protocol is both the largest Launchpad and the largest DEX on Solana, occupying the leading position in two tracks with one project Currently, this situation does not exist on the mainstream chains in the market Moreover, the largest source of fees for Pump fun actually comes from PumpSwap, not the LaunchPad part of the business In the chart in Figure 2, you can actually divide them into three parts to look at its main revenue structure This is the chart data I have organized; if you look at the original Defillama data, it would be even more chaotic: (1) Bonding Curve fee scale = 62.06 million + 10.86 million + 5.22 million, totaling about 78.14 million USD This is the main business for which Pump fun is known in the market, responsible for the token lifecycle's launch phase From the protocol capture perspective, this stage is still currently Pump Fun's largest source of cash (2) Pump swap fee scale = 23.40 million + 25.64 million + 74.87 million, totaling about 124 million USD This is the part of the data most easily overlooked by the market because the past market perception of Pump Fun was mostly limited to "a platform that issues Memes" But in fact, after a Meme or token completes the Bonding Curve phase on Pump fun, it enters PumpSwap for continued trading This stage corresponds to the second phase of the token lifecycle (3) Other business = 9.05 million + 1.06 million, totaling about 10.11 million USD This part is currently small in scale, totaling only about ten million, belonging to ancillary tools, so I won't elaborate here _____________________________________ There is another point in Figure 2 that needs to be expanded: In Defillama's original statistical scope, the total revenue created by a protocol is fully included in the revenue items So there are some repeated items in the cost section, which can easily cause confusion, so you have to look at the profit part, which is the money the protocol actually keeps. I did not expand on the details in the original chart Of the approximately 212 million USD in fees in Q2, Pump fun ultimately retained about 91.55 million USD in gross profit, corresponding to about a 43% gross margin ➠ PumpSwap's profit ultimately captured by the protocol was 23.40 million USD, while Bonding Curve retained 62.06 million USD For these two business segments, the fee scale is larger for the former than the latter, but the value ultimately captured by the protocol is significantly higher for the latter Comparing the change curves of these two business segments over the past few quarters (Figure 3), Bonding Curve clearly depends on market sentiment and has a strong positive correlation with Solana's on-chain activity. Looking at the detailed data change points, this part is very sensitive In contrast, PumpSwap's growth curve is smoother, which further indicates that Pump Fun's revenue structure has changed ➠ In the past, the platform mainly captured value during the token issuance phase, but now the trading phase is becoming a new income supplement The issuance phase has higher protocol capture capability, thus contributing higher profit margins Swap business requires distributing more value to liquidity providers and creators, reducing protocol profit margins, but it covers the longer-term trading demand after token issuance Finally, a brief mention: from the daily K-line chart, $PUMP is in a pullback phase within an uptrend, and the current price around 0.0022 has not yet reached a support level That is all~
EricSJ
EricSJ
The biggest source of fees for Pump Fun is no longer "posting Meme"
Pump Fun Q2 still maintains quite strong profitability, but after several quarters of verification, its revenue structure has undergone a significant transformation. $PUMP is like how many people think Pepsi is just about selling cola, but in reality, its other beverage and snack businesses already account for more than half of total revenue. Pump Fun also experiences this kind of counter-market consensus situation. Initially relying on Launchpad fees, its revenue structure gradually shifted from capturing income solely from the issuance phase to expanding across the entire token lifecycle. This article will analyze its on-chain revenue situation in Q2 and share insights. First, a simple numerical conclusion before diving into details: From Q2 data, although the overall fee scale has declined compared to the peak period, it still exceeds $200 million, and the largest fee source is not the well-known Bonding Curve, but rather PumpSwap. This is also one of the least known aspects of Pump. 1. Fee scale (revenue item) In Q2, Pump Fun's fee scale was about $212 million. Breaking this down: In the Bonding Curve phase, the actual fees captured by the Pump Fun protocol were $62.06 million, representing a fee rate of 0.95%; that is, for every $1 million in transaction volume, Pump Fun can earn about $9,500 in fee income. At this stage, users buying
EricSJ
EricSJ
Although liquidity is poor on weekends, currently $BTC is forming a triangle continuation pattern on the chart. In my view, being at the upper boundary position, it's completely feasible to try a day-trading directional move by shorting. Stop loss placement is also convenient because it's the weekend; the 65,000 level is hard to break through easily, so just place it right above the 65066 integer level, or more conservatively at 65120. Why place it so close? The reason is still the weekend—if it can spike up to this level on the weekend, then just admit defeat. #交易之声:你的经验值得被听到
EricSJ
EricSJ
The biggest source of fees for Pump Fun is no longer "posting Meme"
Pump Fun Q2 still maintains quite strong profitability, but after several quarters of verification, its revenue structure has undergone a significant transformation. $PUMP is like how many people think Pepsi is just about selling cola, but in reality, its other beverage and snack businesses already account for more than half of total revenue. Pump Fun also experiences this kind of counter-market consensus situation. Initially relying on Launchpad fees, its revenue structure gradually shifted from capturing income solely from the issuance phase to expanding across the entire token lifecycle. This article will analyze its on-chain revenue situation in Q2 and share insights. First, a simple numerical conclusion before diving into details: From Q2 data, although the overall fee scale has declined compared to the peak period, it still exceeds $200 million, and the largest fee source is not the well-known Bonding Curve, but rather PumpSwap. This is also one of the least known aspects of Pump. 1. Fee scale (revenue item) In Q2, Pump Fun's fee scale was about $212 million. Breaking this down: In the Bonding Curve phase, the actual fees captured by the Pump Fun protocol were $62.06 million, representing a fee rate of 0.95%; that is, for every $1 million in transaction volume, Pump Fun can earn about $9,500 in fee income. At this stage, users buying