Solana: Ethereum Killer or Bot Playground? Diving Into the Numbers Behind the Surging Metrics

• This report seeks to distinguish fact from perception, examining whether Solana (SOL) provides sustainable, decentralized growth or if its inflated activity is hiding a shaky foundation.
• Solana’s transaction metrics are inflated by bot activity, including wash trading and high-frequency trades. Yet, artificial or not, each transaction generates fees and supports the network’s economic flow.
• Solana is progressing towards decentralization but faces challenges with software diversity and high validator costs, leading to potential network concentration points.
• The project’s success may depend on improving user experience, ensuring real users benefit from low fees and speed, while managing bot activity without compromising network stability.

The term “Ethereum killer ” emerged in 2017 to describe blockchains built to compete with Ethereum’s (ETH) capabilities.

However, in a CoinDesk interview, Solana’s co-founder Anatoly Yakovenko challenges the narrative, stating: 

“Ethereum killer is something I think that the media came up with. It’s always been, in my mind, Solana is focused on execution and transmitting information around the world and synchronizing it at the speed of light. Ethereum is focused on settlement.”

Ethereum set the standard for decentralized applications (dApps) and smart contracts, but it has faced limitations, particularly around scalability, and high transaction fees  during network congestion.

Projects branded Ethereum killers aim to solve these issues with faster, more cost-efficient networks. They promise higher transaction speeds, lower fees and sometimes unique consensus mechanisms, all intended to create a user-friendly experience that Ethereum’s congested network sometimes struggles to deliver​.

Despite initial optimism, many projects fell short of their ambitious goals, primarily due to a critical challenge: Real-world adoption.

Quite simply, users just weren’t engaging with them: Even though they might solve technical problems, most projects fail to draw and maintain the user base necessary to be considered a true “Ethereum Alternative.”

At the same time, Solana (SOL) has emerged as a standout among Ethereum competitors, achieving strong price growth and significant on-chain activity. Still, the platform has not escaped scrutiny:

• Critics call Solana a “VC coin,” arguing that institutional investors use it to attract retail investors, who will then serve as exit liquidity.
• Concerns over Solana also point to frequent network stability issues and a high degree of centralization.
• Many state that a substantial portion of its on-chain activity is bot-driven, which creates the mere illusion of a thriving blockchain network.
• Lastly, critics view Solana’s adoption profile as pretty superficial, with much of its activity centered on meme coins and non-fungible tokens (NFTs).

In this edition of CCN Reports, we will examine the criticisms of Solana to determine their validity. Through a detailed analysis of transaction volumes, developer participation and user engagement, we aim to provide a clear picture of Solana’s strengths and weaknesses.

Solana’s Popularity Has Skyrocketed—Here’s Why

Solana’s rise in popularity comes down to three core factors:

1. Efficient design as a fast, low-cost blockchain.
2. Price performance.
3. The rise of memecoins on its network.

Solana’s Architecture

Solana manages to keep transaction fees low and speeds high by using a combination of proof-of-history (PoH)  and proof-of-stake (PoS) consensus mechanisms (Figure 1).

Solana’s PoH operates by cryptographically timestamping each transaction. It allows transactions to be pre-ordered before validation, eliminating the need for nodes to reach consensus on the time sequence and enables validators to process thousands of transactions per second.

PoS ensures a reliable network security method without requiring intense computational work, thus keeping transaction fees low.

Moreover, Solana incorporates additional optimizations: The Gulf Stream protocol , which forwards transactions to validators even before a block is finalized, and Turbine , a protocol that breaks down large data blocks into smaller, more manageable packets, distributing them more effectively across the network.

How Anatoly Yakovenko’s Career Influenced Solana’s Architecture

The primary reason behind Solana’s architectural design as a fast and cheap blockchain is attributed to its CEO’s vision and past experience.

Anatoly Yakovenko’s journey traces back to his years in telecommunications. At Qualcomm, he spent more than a decade optimizing systems for mobile networks, where the primary objectives were speed and reliability. This is precisely why he had the tools and expertise to address blockchain scalability challenges in a way that prioritized efficiency.

Mobile networks depend on precise timing to handle large amounts of data without delays. Yakovenko observed that blockchains, by comparison, lacked a reliable way to establish the order of events. Existing consensus mechanisms required nodes to communicate frequently to agree on transaction orders, which slowed down the process.

In 2017, Yakovenko proposed Proof of History (PoH), a method that timestamps transactions using a cryptographic clock. This approach reduced the need for constant communication between nodes, allowing the system to process transactions more quickly and at a lower cost.

His experience optimizing hardware also shaped Solana’s design. At Qualcomm, Yakovenko worked with multi-core processors and GPUs to maximize performance through parallel processing. Solana applies similar principles by using its Sealevel runtime to execute non-overlapping transactions simultaneously.

Other innovations, like Gulf Stream and Turbine, further enhance the network’s speed.

Most importantly, Yakovenko’s focus on practicality may be one of the main reasons Solana achieved low fees and high scalability, as he prioritized reducing costs and designing a system capable of handling high transaction volumes efficiently.

SOL Price Performance

Price performance plays a central role in driving attention. Past cycles show that assets with strong price performance have attracted influxes of new users. As the value of a cryptocurrency rises, activity on its network tends to increase accordingly.

Solana is no exception. Its price surges from January 2021 to the end of October 2024 led to spikes in daily active addresses (see chart below).

Solana Memecoins

Memecoins gave the final push that propelled Solana to popularity.

First came Samoyedcoin (SAMO) in May 2021, followed by Bonk (BONK) in January 2023. In December 2023, the arrival of Popcat (POPCAT) and Dogwifhat (WIF) sparked new waves of retail speculation.

By January 2024, Pump.fun emerged, allowing users to trade tokens and create their own memecoins. Now anyone could launch a token, creating a self-perpetuating cycle of new coins, trading activity and hype.

For many, memecoins offered a simple, exciting entry point into Solana’s ecosystem. The price spikes of established tokens like BONK drew in retail interest, with Pump.fun giving users the impression they could recreate the same explosive potential with a token of their own.

Yakovenko addressed  the surge of memecoins on the Unchained Podcast:

“What people want to do is trade memecoins, they want to trade digital assets, they want to have fun. A lot of the stuff is basically just activity that’s fun. I see it as no different than like loot boxes and iOS.

The serious apps that people think about, that Apple is indispensable for, is Uber or whatever, because that’s the magic. I have a logistics pipeline created in a moment’s notice from any point around the world to me anywhere I am. But Apple takes like what, a 2% cut of that fee or something if people use Apple Pay?

The joke that I make is the number of rational markets is countable, but the number of irrational markets is uncountable. It’s just inevitable.

Also, I’d like to remind myself that when Steve Jobs shipped iOS and the App Store, he had to look at the top-10 list and see the fart apps in this beautiful thing that he built.

I don’t think that the weird stuff is going to go away. I think people are just too early. My guess is that, like, you know, within five to ten years, if every business in the world has some service that can manage their wallet and create assets and consume assets, they are going to start with this experiment and expect all their customers to have wallets and stuff like that.

They will just become part of their day-to-day business activity. And then we’ll see more mid-level, like I’m getting taco NFTs from Taco Bell and I’m converting them. It’s loyalty points but with a little bit more flair. I think the space will mature and get bigger simply as crypto gets bigger.”

Solana’s popularity also benefited from major partnerships with PayPal, Visa and Google Cloud , but the primary drivers of its retail appeal remain price performance, efficient design and the memecoin surge.

Have Solana Crashes Become a Thing of the Past?

Critics point out that Solana’s blockchain frequently crashes, calling it unreliable and unstable. This notion may not be entirely fair .

The reliability issues that fuel this narrative stem from 2022, which saw 27 incidents totaling 108 hours of downtime (chart below).

The next year, Solana experienced two outages, totaling 19 hours, and by 2024, the project had one incident, which lasted five hours.

An outage is more than just a temporary offline period. It undermines user trust and disrupts economic activity, impacting everything from DeFi trades to NFT sales, limiting revenue streams for projects and reducing incentives for developers and validators.

Yakovenko acknowledged these complexities.

“Building a complex system is hard, because you just don’t know all the ways it’s going to be used, all the ways it’s going to fail,” he said on the Unchained podcast.

In a CoinDesk interview , Yakovenko offered a different perspective on the pressures Solana faces, describing the memecoin craze as a “blessing.”

Reliability Is a Challenge for Any Network Under Pressure

A network’s true resilience becomes evident only under substantial load. Blockchain networks, like all high-demand systems, face reliability challenges as they scale, since each significant increase in demand reveals new vulnerabilities.

When traffic surges, even the most robust infrastructure can face strain, exposing hidden vulnerabilities and pushing the network to its limits. Thus, reliability is not an absolute but an asymptote—a goal that can be neared but never completely attained as evidenced by the performance of mainstream networks:

• Amazon Web Services (AWS) : 24 outages since 2007.​
• OpenAI : 100+ service disruptions since 2021.
• Automated Clearing House (ACH) System : In November 2023, ACH encountered a processing error that delayed deposits and transfers​.
• Bank of America : October 2024 saw users face login issues and $0 balance displays​.
• JPMorgan Chase : In April 2024, Chase’s website and app went offline, blocking account access.
• Coinbase’s Base L2 Network : 12 outages in 2024 alone.
• TON Blockchain: Two service disruptions in August 2024.

Because of this, DevOps teams use “four nines”  (99.99% uptime). “Four nines” means a goal for a network to be operational 99.99% of the time, allowing for only about 52 minutes of downtime each year. Even massive, centralized systems falter occasionally, and each outage highlights a new area for improvement.

For platforms such as OpenAI, Solana, TON and Base, reliability challenges primarily arise from their popularity. High usage can overwhelm these systems, particularly when unexpected traffic surges exceed the infrastructure’s design capacity.

The strain causes performance issues or crashes, not because the technology lacks quality but because demand has outgrown what it can currently handle. Outages become a predictable part of rapid adoption when activity levels far surpass the original design.

QUIC Implementation on Solana

One way Solana improved its network involves using Quick UDP Internet Connections (QUIC)  instead of a traditional User Datagram protocol (UDP) .

Solana began implementing QUIC in version 1.10.26 , and by 1.10.32+ , QUIC became the default for all servers and routing. Outages fell sharply: From 26 in early 2022 to one, following QUIC’s full deployment. The impact carried forward, with even fewer interruptions in 2023 and 2024.

Originally, Solana relied on a custom UDP-based system to propagate transaction data and keep nodes in sync, which allowed for faster data transmission but lacked reliability.

Think of UDP as ordering a bunch of pizzas to be delivered to a party without any tracking or delivery confirmation. The pizzas get sent out fast, but there’s no way to know if all of them arrive, or if some end up at the wrong address.

In Solana’s case, bots could launch DDoS attacks, known as UDP flood attacks (see graphic below). The attack overwhelms the network with massive volumes of transactions without identification, which makes it nearly impossible for validators to block or rate-limit the bot traffic.

Additionally, the internet heavily depends on Transmission Control Protocol/Internet Protocol (TCP/IP) . It enables connections with websites, web services and other internet-based platforms by establishing a secure and reliable communication pathway.

However, TCP uses a multi-step “handshake” to confirm both sender and receiver are ready for data transfer, including verification steps for HTTPS (secure) websites (see below). Precisely due to the multi-step process it introduces delays, because it operates as a “blocking” protocol.

For example, if someone streams a video over TCP and a data packet gets lost, the entire stream halts until that packet is recovered, a problem known as head-of-line blocking. Or picture a drive-thru, where each car should wait until the one in front has ordered, paid and received food. Simply put, TCP is, contrary to UDP, slow but reliable.

QUIC merges the speed of UDP with the security and reliability of TCP. Instead of TCP’s multi-step handshake, QUIC uses a single-step handshake to establish a secure connection. After that, it opens a “connection stream” and enables data to move as quickly as UDP.

With QUIC, Solana’s network can identify the IP addresses of incoming requests. Validators can then block or rate-limit bot traffic, which drastically reduces the risks of DDoS attacks and high transaction failure rates.

IP identification that QUIC enables applies mainly to validators and RPC nodes, which handle user requests. It doesn’t directly tie IP addresses to specific user identities within transactions.

Solana still uses cryptographic methods, like public and private keys, to secure and anonymize actual user transactions on-chain. So, while QUIC does make traffic management more effective, it doesn’t compromise user privacy, since only the source of the data packets—not the contents—is identified at the IP level.

Even if bots tried to bypass the rate limit using IP tunneling—a method that hides their real IP address—it would add latency and slow them down. Extra latency removes their incentive, as bots depend on high speed to trade effectively.

High Rate of Solana Transactions Failing

Solana’s escalating rate of transaction failures is another major concern. Early November 2024 data from Solscan shows that only 33.31% of transactions on Jupiter  and 53.79% on Raydium  decentralized exchanges have succeeded over the last month (see charts below).

Moreover, according to data gathered from Dune , around 30-40% of non-vote transactions fail across the chain (chart below).

But what does “failure” actually mean in this context?

Have you ever encountered an HTTP error 404 ? Imagine you’re browsing an online shopping site. You click on a product, and the site displays the details. Technically speaking, your browser (the client) sends a request to the shopping site’s servers, which then provides the necessary data.

If the product link no longer exists, you will get a “404 — Not Found” message. In this case, the server successfully receives and processes your request, but informs you that something about it was incorrect.

On Solana, transaction “failures” often operate in a similar manner. So, why do the transactions fail, in the first place?

Reason 1: Memecoins and Low Liquidity

One major factor involves the ongoing boom in memecoins. Thousands of new tokens appear daily, with many without sufficient liquidity. When users attempt to buy or sell low-liquidity tokens, transactions fail due to the lack of available buyers or sellers.

Reason 2: Inexperienced Users and Low Slippage

The recent hype around Solana has drawn in many first-time crypto users. They are excited by potential gains and the buzz around new tokens, but they don’t always understand how DEXs work.

New traders frequently set their slippage tolerance too low. Slippage refers to the percentage of price movement a user is willing to accept for a transaction. If the price shifts beyond this set limit, the transaction will fail. The network simply follows the user’s instructions, but on-chain data still records the transaction as a failure.

Reason 3: Front-running Bots

When a trader’s transaction fails, they often increase slippage to complete it. However, higher slippage attracts bots, which detect the change and initiate new transactions with higher fees.

Consequently, the bot’s transaction is processed first, while the user’s transaction is not, resulting in unfavorable terms or even a failure for the user (see illustration below).

However, this behavior typically targets low-volume tokens and involves significant inventory risk for the bots involved.

In low-volume markets, where limited buyers and sellers create quick, unpredictable price shifts, bots that front-run trades rely on flipping their tokens for profit soon after purchase (see illustration below).

If this doesn’t happen, the bot may find itself stuck with a token that no one wants to purchase at its inflated price. Inventory risk, therefore, refers to the possibility of holding an asset that could be worthless or extremely difficult to sell.

If a bot detects another trader preparing to sell a token right after it’s bought, it may front-run this transaction as well, placing a sell order to offload the token just before the other trader’s order hits. In doing so, the bot minimizes potential losses.

A different option is when bot operators program automated limits and multiple exit points to cap potential losses, even if another trader tries to take advantage by timing their trades.

On Solana, the cost of running these strategies is low, which means bots might escape with only a minor loss, e.g. a few dollars, instead of a large hit (see below).

On a different note, our analysis of swap volume on Solana revealed that approximately 79.4% of the volume is generated by bots rather than human users.

Assuming the data applies consistently, we can estimate that from the 30-40% of failed transactions on Solana previously discussed, roughly 20.6% involve human users, while the remaining 79.4% are attributed to bot-driven activity.

In other words, only 6.18% to 8.24% of total transactions are human-related failures, while 23.82% to 31.76% of failed transactions come from bots.

• Human-Related Failures (Low) = (30% x 20.6%) / 100 = 6.18%
• Human-Related Failures (High) = (40% x 20.6%) / 100 = 8.24%
• Bot-Related Failures (Low) = (30% x 79.4%) / 100 = 23.82%
• Bot-Related Failures (Low) = (40% x 79.4%) / 100 = 31.76%

Let’s rephrase that: Less than 8.24% of actual users on Solana encounter network issues, indicating that the vast majority enjoy smooth operations. The failure data is primarily skewed by bot activity.

This pattern closely mirrors data from Jito’s Solana analysis in March 2023 , which showed that during epoch 414, 58% of block compute was used by arbitrage transactions, with 98% of those transactions ultimately failing (see illustration below).

The issue with MEV bots on any chain is that they represent a perennial problem. While DEXs may implement countermeasures, bot operators continuously adapt to them and bypass the changes.

The decentralized, permissionless nature of blockchain networks makes it difficult to impose permanent solutions, since DEXs can’t fully prevent bots from accessing and monitoring trades.

Reason 4: API Overload and DEX Limitations

DEXs on Solana are stretched to their capacity. Numerous bots exploit free APIs from Jupiter and Raydium, depending on fast RPC nodes to outpace regular traders. Many bypass rate restrictions by running multiple servers, which increases costs and congests the network for other users.

Popularity only exacerbates the problem, as high demand causes both DEXs to struggle with processing the surge of transactions, resulting in even more failures.

Reason 5: QUIC

Unlike traditional networks with a mempool, where transactions wait before being added to a block, Solana’s continuous block production means that losing the connection between users and the block leader results in the transaction not making it into the block at all.

QUIC enables block leaders to cut or rate-limit certain user connections based on predefined criteria. During periods of high demand, block leaders can drop some connections, which allows them to manage network congestion without halting the entire blockchain.

The setup improves resilience. So, even if the network becomes heavily congested, it won’t completely stop, as it might have in the past.

However, despite QUIC’s theoretical advantages, the actual implementation has proven problematic. Block leaders now have the ability to throttle connections, but the logic that decides which connections to drop remains flawed and buggy.

Rather than selectively limiting connections based on specific criteria, such as dropping those with fees below a certain threshold, the system currently drops connections indiscriminately.

During periods of high activity, randomness becomes an issue. Each block leader can handle a limited number of connections, but when demand spikes, it receives requests far beyond this capacity, sometimes 10 to 100 times the usual volume.

Bots flood the network with connection requests, forcing regular users to compete with automated systems just to connect to the block leader.

To successfully get a transaction through, users or their bots must spam more than the others, which makes it increasingly difficult for genuine users to establish a connection and submit their transactions successfully.

As Yakovenko explained in the Defiant interview :

“The vast majority of transactions are not generated by users but by machines. The machine can construct a transaction that has some probability of success, and if it succeeds, whoever wrote that script and algorithm earns a dollar. If they send a transaction and it fails, it costs them one penny.

They can send 100 of them, and that 1 in 100 will earn a dollar. A 1% probability of success is still net positive for them to do that.

So whatever strategy they came up with, because Solana is so cheap and fast, you can do this at these, like, micro levels, where you send 10,000 transactions that cost you $1 to do that, and 1 in 10,000 succeeds and earns you a dollar.

The net effect of increasing throughput and lowering the cost of using the network makes it more accessible for machines. And machines will submit lots and lots of transactions. Some of them will have small probabilities of success.”

Additionally, on the Unchained Podcast, he elaborated:

“Those fees are inescapable, to some extent. There are obviously engineering reasons to make it better, but there’s a limit to how far they (Jupiter and Raydium) can go because you have a single piece of state that’s globally replicated, low liquidity, and high volatility that many different users are trying to touch. There’s no way to fix that problem.

The only way to resolve this is to have people pay priority and get ahead. That person gets to trade, but their trade pushes everyone else to failure because their slippage setting is set too low.”

Is Solana Centralized?

There are different schools of thought on what truly defines “decentralization” in blockchain. Determining whether a blockchain is centralized is more complex than it might initially seem.

Everyone tends to select their own preferred metrics for gauging decentralization, and there is no single consensus. Some argue that true decentralization is found in token distribution, while others believe it hinges on node diversity or point to metrics like the Nakamoto Coefficient.

We’ll examine several indicators of centralization, including:

• initial token allocation;
• the Nakamoto Coefficient;
• Herfindahl-Hirschman Index;
• Gini Coefficient;
• development concentration;
• node geographical distribution;
• client diversity;
• and the cost of running a validator.

It’s essential to highlight that online data for these metrics—particularly the Nakamoto, Herfindahl, and Gini coefficients—is often inaccurately calculated. Instead of depending on external sources, we conducted these calculations ourselves using on-chain data from blockchain explorers and custom Dune queries to ensure accuracy.

Solana’s Initial Token Allocation

Solana’s initial token distribution  leaned heavily towards venture capital, with 30.23% of tokens allocated to investors, which is notably high.

We can look at this from two perspectives. The negative angle suggests the team prioritized profit, while the positive angle sees it as a necessary step for rapid scaling.

Additionally, the team and foundation’s allocations at 12.79% and 10.46%, respectively, are in line with typical blockchain projects. Community distribution made up a substantial 44.07%, while the public sale received only 1.64%.

Increasing the public sale by 15-20% and reducing investor allocation accordingly would have created a more balanced distribution.

Solana may have been aiming for a compromise to satisfy both community and investor interests, but ultimately, the public’s allocation still stands at just 45.71%. This isn’t insignificant, yet it’s not nothing-burger either.

Concerns about centralization are somewhat alleviated when considering the unlock schedule, as most allocations have fully vested, except for the significant case of FTX/Alameda Research.

FTX and Alameda Research Unlock Schedule

It is reported that FTX/Alameda Research is holding  a significant amount of SOL, ready to unload it on the market and tank the price. It sounds alarming, but a closer examination reveals that these fears may be overstated.

According to the Solana Foundation, Alameda Research bought  a substantial portion of SOL, which unlocks according to the following schedule:

This brings the total to 54,086,686 SOL, with 14,541,365 SOL remaining to unlock starting from November 2024.

The largest single unlock will occur in March 2025, with 8,146,178 SOL being released in just two weeks—equivalent to approximately $1.63 billion at $200 per SOL.

Why does this matter?

The year-to-date average and median daily trading volumes for Solana stand at $3.42 billion and $2.95 billion, respectively. So, this unlock would represent about 48-55% of daily trading volume. While the market could absorb it, a sudden sale could easily create a short-term price dip.

FTX Solana Holdings Pose Little Threat

FTX and Alameda Research are deep in Chapter 11 proceedings and need funds to repay creditors. In 2024, they began aggressively auctioning off assets from their portfolio, and one of those assets was Solana.

According to Bloomberg, in April 2024, FTX sold roughly two-thirds  of a $2.6 billion SOL holding at a discounted rate of $64 per SOL. Another tranche in May fetched $110 per token.

Additionally, according to Arkham Intelligence  as of Nov. 13, 2024, FTX wallets now hold only 15,079 SOL, or around $2.7 million.

Nakamoto Coefficient

The Nakamoto Coefficient measures the minimum number of independent entities—validators or nodes—required to disrupt a blockchain’s consensus, essentially, a way of quantifying decentralization.

In a proof-of-stake network, a Nakamoto Coefficient above 33% indicates that a blockchain is resistant to attacks by any single controlling group that holds over a third of the stake. It’s equivalent to asking, “How many people does it take to flip the system on its head?”

By the end of October, Solana’s Nakamoto Coefficient was 19. In comparison, BNB and ETH had coefficients of 2, highlighting their centralization levels. Meanwhile, Cardano (ADA) stood out with a coefficient of 108, showcasing a distinct level of decentralization (see below).

So, where does Solana fall?

When compared to other chains that weren’t originally designed for full decentralization, Solana’s score of 19 places it in a strong upper tier. External data supports this; according to Nakaflow, Solana ranks above 17 out of 22  measured projects.

Moving forward, we should also consider the blockchain trilemma .

The trilemma suggests that achieving full decentralization, security and scalability simultaneously is nearly impossible. Blockchains must make compromises, as optimizing for two of these aspects typically restricts the third.

As a result, finding the right balance involves making tough choices: Prioritizing high decentralization usually decreases speed, while focusing on extreme scalability can make a network more susceptible to attacks.

It seems that most blockchains, including Solana, have recently moved from attempting to completely solve the trilemma to aiming for a more pragmatic middle ground. Rather than being perfectly secure or fully decentralized, they aim to be secure enough and decentralized enough to support scalability without major trade-offs.

Naturally, each chain will excel in some areas and lag in others, but the compromise ensures that every aspect remains at least functional.

Herfindahl-Hirschman Index

The Herfindahl-Hirschman Index (HHI) measures the concentration of ownership within a blockchain. It’s used to assess how evenly token ownership is spread, with lower scores indicating a more balanced distribution and higher scores signaling concentration in the hands of a few.

The index is calculated by squaring each wallet’s percentage of total holdings and then summing these squared values.

We used the top 100 wallets to determine the HHI. Solana achieved a low score of 214.7, which represents the lowest concentration of ownership among major blockchains, outperforming ETH, BTC, ADA, and BNB in decentralization.

Gini Coefficient

The Gini Coefficient, initially developed to evaluate economic distribution, measures income or wealth inequality. In the context of blockchain, it indicates how evenly tokens are distributed across wallets, with a score of 0 signifying perfect equality and a score of 1 indicating total concentration.

The Gini Coefficient is calculated by analyzing the distribution curve of wallet holdings and measuring the disparity between perfect equality and actual distribution.

Using the top 100 wallets, we determined Solana’s Gini Coefficient to be 0.61, the highest among the blockchains measured. This points at a relatively high concentration of holdings, though it’s not as extreme as scores typically above 0.8-0.9.

However, Solana’s score is quite similar to those of BTC, ADA and particularly ETH. Surprisingly, BNB recorded the best Gini Coefficient at 0.22. This reiterates the point we discussed earlier: People can select the metrics or data that best supports their narrative.

Developer Concentration

Developer concentration is crucial and often underestimated. If only a small number of developers are responsible for maintaining and updating a network, decision-making becomes centralized. When power is concentrated among a few individuals, the ecosystem can become susceptible to changes in control or influence.

There’s no definitive threshold for centralization in terms of developers, but, generally speaking, having fewer than 10 to 20 active core developers can be seen as highly concentrated. This is particularly true if these developers have significant influence over decision-making.

With approximately 338 weekly active core developers , Solana ranks just below Ethereum and significantly above most other blockchains. This means that Solana’s development is not dominated by a small group of individuals, making it less vulnerable to centralized control.

It’s worth noting that weekly core active developer numbers have been on a downtrend across most chains since 2021, likely tied to the cooling of the bull market. If the typical cycle pattern repeats, we could see developer activity increase again by 2025. However, this remains uncertain.

Nodes

Geographical Distribution

When nodes are spread across different countries, cities, and data centers, a network becomes less vulnerable to localized risks, such as regulatory restrictions, natural disasters, or power outages. A geographically dispersed network helps ensure that no single country or region can easily disrupt or control the chain.

As of November 2024, Solana’s network counts 4,619 nodes  across 45 countries, 206 cities, and 503 unique data centers. In practical terms, such a spread enhances both security and decentralization.

For comparison, Bitcoin and Ethereum have 19,538  and 5,720  nodes, respectively, which span across 94 and 74 countries.

Software Diversity

If nodes rely on multiple versions of client software, the network can withstand potential bugs or targeted attacks on a single version. A network heavily reliant on one or two versions becomes vulnerable; any flaw or exploit in those versions could impact a significant portion of the network.

Solana’s software diversity  falls short of this ideal. By the end of October 2024, 69.69% of validators were using version 1.18.25 of the Solana Labs Client, while 20.55% were on version 1.18.23. Combined, these two versions account for over 90% of the network, resulting in limited redundancy and resilience.

The concentration becomes even more visible when considering that only two clients exist in Solana’s ecosystem. The second client, JITO, held  a 31.45% share of the network, according to the Solana Foundation, as of October 2023, with the remaining majority relying on the Solana Labs client.

The issue may see improvement soon, as new clients, such as Firedancer and Sig , are under development. Their release and adoption could enhance Solana’s software diversity, which would bring greater resilience to the network.

Currently, Solana’s dependence on just two main clients, Solana Labs and Jito Labs, exposes the network to risks associated with limited software diversity. In comparison, Bitcoin and Ethereum handle software diversity more effectively.

Bitcoin’s nodes use nine different versions, with 87% of the network spread across them and only the top three versions controlling just over 50%, led by Satoshi 27.1.0 at 33.9%.

Ethereum shows some concentration, but, still, distributes over 90% of its nodes across three main clients: Nethermind at 45.08%, Geth at 39.10%, and Erigon at 7.09%, each maintaining a balanced distribution of software versions.

Cloud Provider Distribution

Cloud provider hosting operates similarly to client software versions or geographic locations: Greater diversification leads to better outcomes. Solana’s network demonstrates a relatively balanced distribution  among its top providers.

Vultr hosts 206 nodes, TeraSwitch Networks hosts 110, OVH SAS has 106, Latitude-sh has 98 and UAB Cherry Servers also has 98 nodes. Together, these five providers control about 47% of the network, which creates a healthy spread and reduces centralization risks.

On the other hand, Ethereum has a more concentrated distribution among providers. Amazon.com hosts 618 nodes (28.3%), and Hetzner Online GmbH hosts 421 nodes (19.3%). Combined, these two providers control 47.6% of Ethereum’s network, making it more vulnerable and less decentralized.

Cost of Running a Validator

Solana is also viewed as a centralized platform partly due to the high entry barriers for validator operators. According to Solana Labs , the baseline requirements to operate a validator include:

The issue here is that the cost of running a validator shapes who can participate in maintaining a blockchain’s security and decentralization. Validator expenses—hardware, electricity and network fees—act as a barrier to entry.

When costs are high, participation becomes limited to wealthier individuals, companies or institutions with the resources to support these operations.

As a result, a concentration of wealth and power within the validator set leads to centralization, as a small number of participants maintain significant influence over the network.

Conversely, when validator costs are accessible, it allows a broader range of participants, including smaller independent operators, to join. This distribution of control among many strengthens the network.

There are ways to build and operate a Solana validator node without relying on cloud servers, but this approach remains somewhat unreliable and costly.

Operating a node independently requires that the operator possesses technical proficiency and can manage complex configurations and troubleshooting. Even for those with the necessary technical skills, the process can present numerous challenges, potentially causing stability issues or network performance problems.

Moreover, the bandwidth requirements for a Solana validator are substantial, ranging between 50 to 100 TB per month, according to estimates from discussions within the Solana validator Discord community.

Achieving this level of bandwidth at home usually needs a dedicated line from an ISP, which can be expensive and not easily accessible to everyone. Even with such a setup, many operators have reported ongoing issues and interruptions when running nodes independently. This often means they struggle to meet the performance standards required for consistent validation.

Therefore, at present, server hosting remains the primary and most practical option for those seeking to reliably operate a Solana validator.

Running a node requires hardware costs between $552 and $879 monthly , depending on the location. Additionally, bandwidth can cost an extra $70 to $360 each month . Annually, the total operating cost amounts to roughly $7,500 to $15,000.

Each vote a validator submits costs 0.000005 SOL, which accumulates to about 2.16 SOL per epoch, or 394.2 SOL per year. At $180 per SOL, that translates to approximately $71,000 annually in voting costs, bringing the total annual cost to around $78,500 to $86,000.

Even with a conservative estimate, cutting this figure in half still results in costs over $43,000 per year, which is quite high for the average person. Ideally, for true decentralization, the cost structure would be much lower, enabling people to set up their own hardware once and operate it at home with minimal ongoing expenses.

This suggests a “buy once, forget” model with the most affordable viable hardware. Or simply, a setup that would make running a node financially attractive for a large portion of the global population.

It’s worth noting that there are ways for operators to reduce costs and achieve profitability more quickly, such as:

• Using programs like the Solana Foundation’s Delegation Program, which covers voting costs for the first year.
• Commission rate optimization.
• Targeting liquid staking pools to boost the staked amount without requiring substantial initial investments.
• Opting for JitoLabs’ MEV capture, which offers an additional revenue stream through off-chain bribes for processing certain transactions.

Let’s assume  that the operator uses every available benefit, including:

• the cheapest possible hardware;
• a 5% standard commission rate, according to Solana Beach;
• an 8% Jito MEV commission;
• Solana’s inflation rate at 4.939%;
• voting costs totaling 394.2 SOL per year;
• an average block reward of 0.0597 SOL for the last week of October 2024;
• an average MEV tip of 0.1028 SOL per block for Oct. 30, 2024;
• and a skip rate of 4%.

Then the operator would need at least 27,114 SOL in staked assets to break even. That translates to around $4.88 million worth of SOL in stake at $180 per SOL.

There are multiple ways for operators to achieve sufficient stake without massive initial capital.

For example, stake pools like Marinade Finance and SolBlaze algorithmically distribute delegated tokens across a large number of validators.

They provide a liquid staking model, which means users can stake their SOL with these pools and receive liquid tokens, such as mSOL or bSOL, which can still be used across DeFi platforms. Both pools enable validators to attract stake based on their operational quality, even if they lack a strong reputation or significant self-stake.

The Delegation Program is another major avenue for new and small validators to build up their stake and reduce operating expenses.

It’s designed to support validators who meet certain criteria, with a big focus on decentralization and performance. Eligible validators can receive up to 100,000 SOL in delegated stake from the Solana Foundation.

​​Finally, Jito Labs has a different delegation program that centers on Maximally Extractable Value (MEV) rewards. By using the JITO client, validators can capture MEV on Solana, allowing them to earn “off-chain bribes” from transaction ordering and block building on the Solana network.

Validators running the JITO client share the MEV bribes they earn with their delegators. It increases the returns for those who stake with them and helps validators earn additional income outside of standard block rewards and transaction fees.

To summarize, operators have two options: Run a technically challenging node at home with high risks of downtime and failure or opt for a stable, reliable but costly server option. Both scenarios require a significant stake to cover the costs.

For most, reaching a break-even point with an initial stake requirement of $4.88 million makes decentralization impractical at this level, as it limits participation to those with substantial capital.

Despite various methods to acquire the initial stake or reduce costs, the barriers remain significant, making widespread, decentralized participation difficult to achieve.

It’s important to highlight insights from the Solana validator community, which suggest that achieving profitability in less than a year is possible. This perspective is based on effectively using available delegation programs, optimizing commission rates, and earning potential from MEV.

However, even if an operator manages to reduce or offset a significant portion of recurring expenses, the hardware costs alone present a substantial barrier.

The current economic landscape of Solana benefits those who can afford the initial costs and skillfully manage staking. This reinforces concerns that validator operations may lean toward centralization out of necessity rather than choice.

By assigning a score of 1 for each positive point and 0 for each negative, we can evaluate Solana’s decentralization as follows:

Using this scoring method, Solana scores 6 out of 9 on the decentralization scale, leaning slightly towards decentralization but with some notable trade-offs.

We gave a 0.5 score to both the initial token distribution and the Gini coefficient because each represents a mixed scenario, not clearly fitting into either centralization or decentralization.

Adjusting the scores by counting the two 0.5s as zero, the tally changes to 5 out of 9. This places Solana on a fine line, halfway between a centralized and decentralized system.

Solana’s Reported Numbers Are Misleading

Daily Active Wallets

In October 2024, Solana’s daily active wallets reached an all-time high of 9.4 million, a figure many believe to be inflated. Let’s explore why that might be the case.

There’s no data available on Solana’s total unique addresses. This is important because knowing the total address count provides a clearer picture of how many wallets are genuinely active.

Consider Bitcoin: It has more than 1.34 billion addresses , but most are inactive or empty. Wallets with a non-zero balance total about 52.55 million , and the peak for daily active wallets  was around 1.36 million, in April 2021.

This indicates that only 4.15% of total addresses hold a balance, and daily active wallets represent just 0.1% of the total addresses or 2.59% of the non-zero wallets.

However, Bitcoin might not be the best comparison, since it is often viewed as a store of value rather than a platform for high transaction activity. Ethereum serves as a better benchmark in this context.

With 226.28 million  total addresses, 130.05 million  of which hold a balance, Ethereum saw a DAU peak of 1.42 million  in December 2022. This means 57.47% of addresses are non-zero, and DAUs account for 0.63% of total addresses or 1.09% of non-zero addresses.

The framework gives us a baseline for evaluating activity levels.

Since no public source provided the total number of Solana addresses, we turned to Dune Analytics. With some guidance from @21co and @cryptokoryo, we adjusted their code to estimate Solana’s address data. The results were quite surprising.

A total of 1.942 billion Solana addresses were found. With a peak DAU of 9.4 million, according to the Artemis Terminal, this represents only 0.48% of total addresses.

By using similar methods to estimate non-zero balances, it was found that there are 633.21 million addresses with a balance, or 32.61% of the total. The DAU among these non-zero addresses is 1.49%, which is lower than Bitcoin’s and higher than Ethereum’s ratios but still close enough to be considered valid.

Solana’s Staggering Number of Wallets Raises the Question: Why so many?

Bitcoin, being around since 2009, naturally has over a billion addresses as the most established cryptocurrency.

Ethereum, being relatively newer, limits excessive wallet creation and trivial transactions through its transaction fees. These costs act as a filter, discouraging users from spamming the network or creating unnecessary wallets.

Solana operates differently. Its low fees and high throughput remove nearly all barriers, making it easy for bots to be active. With transaction costs being almost negligible, there’s little to stop the creation of countless wallets or running high-frequency transactions.

The vast number of Solana wallets is concerning. Data from Oct. 22, 2024, by Hello Moon shows  that most daily active wallets hold very little or no SOL:

Among the daily active wallets, nearly 8 million have a balance of 0 SOL. This means that most of Solana’s wallets are not actively holding or transacting in SOL.

These 0-balance wallets likely serve as disposable, single-use accounts or placeholders for other purposes, which reduces the meaningful user count and raises concerns about the quality of activity on the network.

Several factors contribute to this unusual distribution. Firstly, Solana’s minimal transaction fees encourage both users and bots to create multiple wallets without significant costs.

With no financial penalties, creating and discarding wallets remains low. This makes it easy to automate wallet creation for various purposes, from decentralized applications to spam or manipulation attempts.

For instance, some decentralized applications may use multiple wallets to increase transaction throughput or simulate user activity, creating the illusion of engagement on the network.

Additionally, bots are drawn to Solana’s low costs and high throughput for high-frequency trading, arbitrage, or speculative activities. These strategies often need rotating wallets to avoid detection or exploit inefficiencies.

Transactions and Volume

The debate around Solana’s transaction and volume data splits into two schools of thought. One camp believes Solana’s numbers are inflated, driven largely by bots and fake trades, while the other celebrates these figures as proof of a thriving network. Reality, however, leans closer to skepticism.

We examined the top token by number of transactions on Dexscreener and chose SAW, which launched just 12 hours before our analysis. The transaction pattern revealed an odd repetition: 13.83 SAW was traded repeatedly.

This is known as wash trading. It artificially inflates both the transaction count and volume. For example, if a token trades at $100 through thousands of repetitive trades, the volume calculation can reach millions. However, the actual volume behind these trades is fake.

Other examples reveal similar patterns, with more elaborate trades appearing in neat multiples, like 1.74, 1.75, 17.5, 173.29, or 87.52 and 8.66.

These are not signs of organic activity. Sometimes, bots make small random changes to these figures to look legitimate, but the overall pattern is still clear.

Liquidity data from Raydium shows an even clearer distortion (Figure 31). Tokens, like LUCE, FLOCK, and XENO, sit in liquidity pools of $1, $19, and $3, respectively, yet show daily volumes in the hundreds of thousands.

The mismatch between liquidity and volume shows how bot-driven activity inflates numbers.

What’s the Harm in Inflated Metrics Like This?

The primary issue is the distortion of real user engagement. Fake volume gives the false impression of organic growth, misleading those comparing Solana to more established chains, where most transactions come from genuine users, despite some bot activity.

Claiming that one chain is superior based on inflated adoption metrics distorts the perception of its actual popularity.

This situation also poses a real risk for users. High-visibility pools with very little liquidity often lure new investors who may fall victim to rug pulls. Solana seems particularly affected. Over the past 14 days, Solana made up more than 72% of bot trades across major chains, including Arbitrum, Avalanche, Base, BSC, Ethereum, and Fantom.

It’s important to note that issues like wash trading and low-liquidity pools are not unique to Solana—they occur across the entire blockchain ecosystem. However, Solana’s low fees and high transaction speeds make it especially attractive for bot activity, which makes these issues more pronounced on this network.

The chart below provides an overview of bot volume across major blockchain networks. Bot activity is also present on chains, such as Fantom, Arbitrum, Avalanche, BSC, Base, Ethereum, and Blast.

Yet, Solana consistently dominates this metric, accounting for 65% to 83% of bot volume on different days.

On the other hand, whether transactions come from bots or real users, the network stays active, and transaction fees keep adding up. This means Solana’s system is still in demand, as each transaction—whether genuine or automated—uses network resources and financially supports the ecosystem.

If Solana slowed down its network and raised fees, it would reduce bot activity. But that would defeat its main advantage of being fast and cheap, essentially harming itself.

Another option involves increasing anti-bot measures, but there’s little incentive for this approach as bots contribute transaction fees that support the network’s economy.

The middle ground would ensure that bots don’t waste network resources, disrupt operations, or interfere with the experience of genuine users.

Solana has begun adding features to reduce bot activity and improve the network for real users. One of the initiatives is the QUIC protocol we talked about earlier.

Another good example is the Stake-Weighted Quality of Service (QoS) , which allows block producers to prioritize transactions from validators based on their stake.

This means that validators with higher stakes can transmit a proportional share of the transaction load, thus reducing the impact of low-stake—or even no-stake)—low-quality nodes that might otherwise flood the network.

Transactions per Second

Solana’s theoretical peak  stands at 65,000 transactions per second (TPS), making it one of the fastest blockchains. Only a few chains, like Aptos at 160,000 TPS and Internet Computer at 209,708 TPS, exceed this performance.

While Solana hasn’t reached this theoretical peak in practice, it ranks high in the market with a documented maximum TPS of 7,229 tx/s , according to Chainspect, ranking it third overall. Solana’s historical average TPS hovers closer to 3,000, according to Solscan.

So, what makes up these numbers?

The TPS figure often cited as Solana’s “real” throughput includes vote transactions. These are basically internal messages that validators send to maintain network consensus. They ensure smooth block production, but don’t count as economic transactions or actual user activity.

Both Solscan and Chainspect have addressed this issue. Solscan offers two different metrics for Solana’s performance: “standard TPS,” which counts all transactions, and “true TPS,” which excludes vote transactions. Chainspect focuses only on non-vote transactions in its calculations.

Another factor to consider is failed transactions, which are included in the TPS count. Even when a transaction fails, the network still processes it and uses up computational resources.

Each failed transaction still takes up space in the blocks and adds to the TPS count, even though it doesn’t change the chain. So, Solana’s high TPS reflects the total computational effort, not just successful user activity.

To get a more accurate measure of real economic transactions, we analyzed Block #298758484. Out of 2,574 transactions, only 304 met the criteria, and 51 of those were failed transactions.

We defined “real” transactions as those that move assets, create accounts, modify positions, initiate trades, or interact meaningfully with DeFi protocols. In other words, transactions that involve true asset movement or state-changing activity.

It’s important to understand that in Solana, TPS measures the real-time demand for space in each block. Unlike other blockchains, Solana doesn’t have a “mempool” or queue where transactions wait to be processed.

Each transaction either enters the current block immediately or it doesn’t, which means TPS reflects the exact number of transactions users are actively pushing through the network at any given moment.

For blockchains with mempools, like Ethereum and Bitcoin, transactions accumulate in a queue until there is enough space in a block. The TPS of these chains typically reflects the available block space rather than the actual demand.

When transactions exceed block capacity, they end up waiting in the mempool. This queue management system means that for blockchains with mempools, TPS reflects the supply limit of block space rather than the immediate demand, as it does in Solana’s case.

Inflation and Staking

The “high” inflation rate is a frequent point of criticism among skeptics. However, a common misconception arises from confusing the circulating supply with the total supply.

Total supply refers to the entire amount of SOL tokens within the Solana ecosystem. This includes tokens that are currently in circulation, staked, held by institutions, or otherwise locked or vested.

Circulating supply, on the other hand, refers to the SOL tokens that are currently available in the market for trading or spending.

Solana’s inflation is guided by a structured schedule, defined by two key parameters: The initial inflation rate and the disinflation rate.

When Solana launched in 2020, the initial annual inflation rate  was set at 8%. This high rate aimed to encourage early staking and strengthen validator participation. However, sustaining such a high rate long-term would cause excessive dilution, leading to a disinflation rate of 15%.

Each year, the inflation rate decreases by 15%, gradually moving towards a long-term target of 1.5%. This process leads to an increase in the total supply. Over the past four years, the average inflation rate has been around 4.73%.

However, the circulating supply has grown much faster, averaging around 15.86% per year. This difference comes from the initial token distribution, where a large portion of SOL was allocated to the team, early investors, and other stakeholders, but remained locked at launch.

As these tokens unlocked over time, they entered the circulating supply, leading to a growth rate that exceeded new token issuance through inflation.

Even though the increase in circulating supply isn’t technically inflation, it still impacts token holders’ purchasing power. The introduction of more tokens into the market dilutes the value of existing holdings.

As more tokens enter circulation, the increased supply puts downward pressure on the price. Consequently, the token’s value trends toward depreciation, even with the controlled inflation rate of the total supply.

To counterbalance the impact of increasing supply, Solana uses deflationary token burning and staking incentives. Token burning permanently removes 50% of transaction fees from circulation, helping to align value with supply growth.

While this doesn’t completely counteract inflation, it does create a deflationary effect that becomes stronger as transaction volume increases.

↑Transactions → ↑Transactions Fees → ↑SOL Burned

Increased transaction activity results in higher fees and more tokens being burned. If token burning regularly offsets or surpasses new token issuance through inflation, the circulating supply would begin to decrease.

However, this can only be achieved if transaction volume stays consistently high enough to burn tokens at a rate that matches or surpasses inflation.

Staking is another way to lessen the impact of inflation on holders. By staking SOL, users lock up their tokens to support network security and, in return, earn rewards.

Staking yields consistently outpace the inflation rate, allowing participants who stake to preserve or even increase the value of their holdings relative to inflation.

Historically, Solana’s staking rate has varied between 5.5% and 9.3%. Currently, the nominal staking rate is around 6.21%.

The staking yield is influenced by the proportion of the total SOL supply that participants commit to staking. As the percentage of SOL staked increases, the staking yield decreases and vice versa.

Revenue and Financial Statements

Another point to address is the perception that Solana is an “unprofitable” blockchain network. This notion is supported by financial statements from Token Terminal and referenced in Messari’s quarterly reports. However, describing Solana as unprofitable misunderstands its purpose and operational structure.

The problem stems from using traditional business financial metrics to evaluate a decentralized protocol. This framework is simply not suitable.

Unlike traditional businesses, Solana and similar blockchains do not aim to maximize profit. Instead, their goal is to secure and decentralize their networks. Viewing token incentives as expenses overlooks the fundamental way these networks operate.

Solana’s token issuance does not represent a loss but is a mechanism to incentivize validators, maintain security, and drive participation. Without these incentives, decentralized protocols would lose the very properties that make them resilient and functional.

By conventional profitability standards, Solana, Bitcoin, and Ethereum could only seem profitable if they stopped all issuance or aggressively burned tokens. However, they would always appear to “operate at a perpetual loss” because they continuously issue tokens to maintain security.

Eliminating token issuance would undermine the security and decentralization that are fundamental to these networks. Token burning might increase scarcity, but it acts as a supply control mechanism rather than a tool for generating revenue.

Conclusion

Solana finds itself at a crossroads between ambition and reality. Despite its reputation for speed and low fees, the network struggles with high bot activity, which inflates transaction metrics and distorts perceptions of adoption.

Regardless of whether the activity is artificial, it generates fees and keeps the network operating as intended.

The idea that bot-driven activity might vanish, leaving Solana inactive, overlooks a crucial point: Solana’s appeal to both bots and users lies in its speed and low costs. Without these benefits, neither group would find the network worthwhile.

Although Solana leans towards decentralization, it still faces challenges, especially in terms of software diversity and the high costs associated with running a validator.

The key question is whether Solana can maintain a balance between accessibility and performance without sacrificing genuine activity or decentralization.

Currently, Solana’s progress seems to be tied to managing bot traffic, refining interactions to ensure bots don’t disrupt user experience. If Solana can handle bot activity without bothering actual users, it might strike a balance, combining both automated and human engagement to keep the network lively and active.

CCN Reports is a regular series that delves into the details to provide in-depth analysis of cryptocurrencies and the companies associated with them. We aim to engage a global audience interested in what’s what, who’s who and perhaps even why’s that.