What Makes a Blockchain Truly Decentralized? Key Decentralisation Metrics Explained
- Jan 30
- 11 min read
Updated: Feb 2
The majority of blockchain projects self-identify as decentralised. The term is used in a variety of business documents, including white papers, press releases and investor decks. In practice, however, the reality often differs. Control can be allocated to a small group of validators. The governance of these projects can be delegated to a small group of developers. The supply of tokens can remain highly concentrated.
This discrepancy between claims and reality is a significant consideration. Decentralisation is not a branding feature. It has implications for security. It has been demonstrated to affect resistance to censorship. It affects whether users must trust a small set of actors. Recent network outages and protocol freezes have demonstrated this risk with notable clarity.
Measuring decentralisation is a complex process. There is no single figure that proves this. Different networks decentralise different parts of the system. Some focus on nodes. Other stakeholders may prioritise aspects such as ownership or governance. The absence of clear metrics results in the term becoming meaningless.
Key Highlights:
Decentralisation is a multi-faceted concept, encompassing validators, consensus power, token ownership, governance, and infrastructure. It cannot be measured by a single metric.
High validator or node counts can obscure actual concentration when it comes to stake, mining power, or hosting infrastructure clusters among a small number of actors.
Both Proof-of-Stake and Proof-of-Work systems demonstrate a tendency to concentrate power through capital or pooling.
Token ownership, delegation, and liquid staking have been shown to significantly influence real governance power, often beyond that of formal voting mechanisms.
Infrastructure dependencies, such as cloud hosting, RPC providers and MEV middleware, introduce centralisation risks outside the protocol itself.
What Decentralisation Actually Means in Practice
Decentralisation is often reduced to a simple idea. This phenomenon is often characterised as a lack of control. In practice, it is more specific. The matter pertains to the criteria for participation. Furthermore, it is pertinent to consider the parties with the authority to amend the established regulations. The question of asset ownership is also relevant in this context. A network may be open to users but closed to decision-making.
Note that control, participation and ownership are not always aligned. Many networks permit any individual to initiate transactions. It is possible that validation rights remain with a limited group. Token ownership can remain concentrated among early investors. In such cases, user access exists without the necessary authorisation. This structure is commonplace among large blockchains in the current landscape.
Decentralisation also operates on a spectrum. This is not a binary state. Some systems distribute power widely. Others do so partially. Even Bitcoin, frequently regarded as highly decentralised, exhibits concentration in mining pools. As of 2024, the top five mining pools controlled over 65% of Bitcoin's hash rate at various points. This should not be interpreted as an indication of failure. It demonstrates the trade-offs involved in real systems.
Decentralisation is a multifaceted concept with numerous layers. The technical layer encompasses nodes, validators and clients. The economic layer encompasses the supply of tokens and the associated incentives. The social layer encompasses governance and influence. It is possible for a network to achieve a high score in one layer, yet a low score in another. Ethereum demonstrates notable client diversity. The governance influence continues to be concentrated among a small group of teams and stakeholders.
No blockchain system has been able to achieve full decentralisation across all layers. There is a natural tendency towards concentration, driven by hardware limitations, capital requirements and coordination costs. This does not invalidate the model. This means that decentralisation must be measured, rather than assumed. The implementation of clear metrics is instrumental in distinguishing between structure and narrative.
Validator and Node Distribution Metrics
One of the most visible signals of decentralisation is the validator and node distribution. This demonstrates the significant contribution of independent actors in the network's operations. It also demonstrates the difficulty of joining that group. A higher count generally indicates greater participation. This does not guarantee the outcome.
Note that large blockchains report very different numbers. As of 2024, Bitcoin had over 15,000 reachable full nodes. These nodes verify the integrity of the blocks and transactions. Note that not all of them produce blocks. Following the transition to Proof of Stake, Ethereum had more than 900,000 validators. These validators are divided across numerous operators. Solana reported approximately 2,000 active validators. Avalanche has reported that it has over 1,700 validators across its primary network.
The geographic spread of the business is also an important factor. Nodes that are concentrated in a single country are exposed to shared legal risks. According to data from Bitnodes, the United States was home to over 35% of Bitcoin nodes in 2024. Germany and the Netherlands followed at a distance. Ethereum nodes demonstrated a comparable trend, exhibiting substantial concentration in North America and Western Europe. This creates jurisdictional exposure even when node counts appear satisfactory.
The hardware and bandwidth requirements determine the participants. Bitcoin nodes can run on modest hardware. A consumer laptop is often sufficient for this purpose. In order to become a validator, a minimum of 32 ETH is required, as well as stable uptime. Solana validators require high-end CPUs, large memory, and fast network links. These costs have the potential to limit participation. They advocate for the centralisation of validation processes within data centres and the hands of professional operators.
Node count alone can be misleading. A significant number of nodes may share the same cloud provider. It has been demonstrated that outages at major providers are a risk. In 2023 and 2024, cloud-related incidents impacted several blockchain networks simultaneously. True resilience is contingent on diversity. This includes hardware, software clients, and hosting environments.
Decentralisation Metrics Across Major Blockchains
Network | Validators / Nodes | Power Concentration Signal | Infrastructure Risk | Source |
Bitcoin | ~15,000+ reachable nodes | Top 3 mining pools >55% blocks | Mining pool coordination | |
Ethereum | ~900,000+ validators | Top staking entities ~45% stake | Cloud + MEV reliance | |
Solana | ~2,000 validators | High hardware barriers | Data-center hosting | |
Avalanche | ~1,700 validators | Validator weight skew | Subnet operator overlap |
Consensus and Power Concentration
In the field of blockchain, consensus design is instrumental in determining the distribution of power. Proof of Work links influence to computing power. Proof of Stake is linked to capital. Both systems have been found to create pressure towards concentration. The source of power may differ, but the outcome can be similar.
In Proof of Stake networks, the concentration of stake is a pivotal issue. Large holders can run many validators. Smaller holders frequently delegate. In mid-2024, the top five staking entities on the Ethereum network held a combined stake of over 45% of the active stake. Lido alone accounted for approximately 30%. This structure gives rise to concerns regarding coordinated behaviour. Furthermore, it has been demonstrated that this will result in an increased reliance on a small number of operators and governance forums.
Proof of Work demonstrates a different pattern. Mining pools aggregate hash power in order to reduce income variance. Consequently, the focus of block production shifts. In 2024, the top three Bitcoin mining pools consistently accounted for over 55% of blocks in a given week. Pool dominance does not necessarily equate to ownership of hardware. It should be noted that this process still creates coordination risk at the block level.
Cartel formation is a known risk in both models. Validators or pools can align on transaction ordering. They have the capability to censor addresses. They have the ability to delay protocol changes. These actions do not require malicious intent. Economic incentives are often effective in encouraging cooperation among actors. Should a small number of players exercise control over the final outcome, the risk will increase accordingly.
Finality and slashing introduce additional trade-offs. The enhanced speed of the finalisation process is known to improve the user experience. Furthermore, it has been demonstrated that this results in an increased reliance on a smaller active set. Implementing Slashing protocol is an effective measure to ensure the security of your network and protect against potential cyber-attacks. It also has the potential to introduce operational risk. This situation tends to favour large, professional operators who can call on legal and technical resources. The same principle often applies to governance power. Those who validate the most blocks tend to shape upgrades.
Consensus does not necessarily equate to decentralisation. It determines the accumulation of influence. It is imperative to understand this dynamic when comparing networks.
Economic Decentralization: Token and Stake Distribution
Economic decentralisation is the process of examining the ownership of value within a network. Furthermore, an analysis is provided on the subject of the entities exerting control over the distribution of staking power. It is common for token supply to be concentrated at the outset. Founders, funds, and early users tend to hold large shares. This has a long-lasting impact on the company's influence.
The data indicates a clear concentration of major blockchains. On the Ethereum network, the top 100 addresses held approximately 41% of the total ETH supply in 2024. A significant proportion of these addresses are associated with exchanges and staking pools. On the Solana blockchain, the top 20 wallets held over 50% of circulating SOL at several points in 2024. Bitcoin also displays a comparable trend. It is a fact worthy of note that the top 2% of addresses have control over more than 90% of the total supply of BTC. These figures do not indicate active control by a small number of individuals. This is indicative of uneven power distribution.
Delegation has a significant impact on the practical functioning of stake. Note that many users do not run validators. These tokens are then delegated to large operators. Liquid staking is set to further boost this trend. Protocols such as Lido issue tokens that represent staked assets. This improves liquidity. It also centralises voting power. When one protocol aggregates a significant share of stake, it gains considerable influence over validation and governance.
Exchanges play a pivotal role in this layer. The company holds custody of user assets. They frequently act as representatives for their clients. This gives them voting power without direct ownership. In 2024, centralized exchanges controlled a significant portion of staked ETH and SOL. This has the potential to create regulatory and governance risk. Decisions may be influenced by compliance requirements rather than network health.
The distribution of wealth has a subtle effect on network control. Large holders have a significant influence on proposals. They provide funding for client teams. These individuals play a pivotal role in shaping the discourse within the public sphere. Even without the right to vote, their voice is influential. Economic pressure can influence outcomes to the same extent as formal rules.
This underscores the importance of token metrics. Supply charts alone are not sufficient. Analysts must consider delegation flows, custodial shares, and staking concentration as a cohesive unit. These factors demonstrate the location of genuine influence.
Governance and Upgrade Control
The governance framework determines the entities permitted to modify a blockchain. In many cases, this is more significant than transaction throughput. Most networks are decentralised in their use. In contrast, a decentralised approach to decision-making is characterised by a distribution of authority and responsibility across multiple individuals or entities.
Protocol changes are typically initiated by a small group. Core developers are responsible for writing improvement proposals. The proposals will then be subjected to a rigorous review and testing process. On the Ethereum platform, Ethereum Improvement Proposals are developed and discussed publicly. The final decisions will be made by the client teams and validator adoption. There is no formal on-chain vote that binds outcomes. Social consensus plays a central role.
Some networks utilise on-chain governance. Token holders have the opportunity to vote directly on upgrades. Examples of such chains include Cosmos-based ones and Tezos. This model offers clarity. It also ties power to token ownership. The influence of major stakeholders can be a key factor in determining outcomes. It is common for there to be low voter turnout. In many cases, participation levels are less than 10%. This raises questions about the representativeness of the sample.
Even in circumstances where they lack formal authority, foundations can still exert a significant influence on governance. They fund research. They provide support for client development. They are instrumental in shaping strategic direction. The Ethereum Foundation has played a key role in major upgrades such as the Merge and Shanghai. Solana Labs and the Solana Foundation have implemented several performance-focused changes. This support accelerates development. Furthermore, it has been demonstrated that this practice concentrates agenda-setting power.
It is evident that the trade-off is clearly revealed by emergency upgrades. In the event of an exploit or outage, it is imperative that the network responds promptly. In 2022 and 2023, several blockchains halted or coordinated restarts after failures. These measures were implemented to safeguard the interests of users. Furthermore, the evidence suggests that a small group can intervene. Full decentralisation would reduce the speed of such responses.
Governance models are reflective of organisational priorities. Strong coordination improves safety and speed. This can result in a reduction of independence. Weaker coordination improves neutrality. It has the potential to increase risk during times of crisis. There is no such thing as a perfect design. In a business context, transparency and realistic assessment are of the utmost importance.
Infrastructure and Dependency Risks
The infrastructure is not yet aligned with the consensus. In many cases, it has a significant impact on the decentralisation of a network in real-world conditions. Many blockchains rely on shared services. These services are not part of the protocol. These factors continue to influence access and reliability.
Cloud providers are a key dependency. A significant proportion of nodes are run on a small number of platforms. Amazon Web Services, Google Cloud, and Hetzner host a significant number of validators and RPC nodes. In 2024, it was estimated that over 60% of Ethereum nodes were using major cloud providers. In the event of an issue being encountered by one provider, there is the potential for multiple nodes to become unavailable simultaneously. This risk has already manifested itself during regional outages.
RPC services add an additional layer of complexity. It is rare for users to connect directly to their own nodes. They rely on providers such as Infura, Alchemy, or QuickNode. These services are responsible for the routing of transactions and queries. It is possible for them to throttle traffic. In order to meet legal demands, they have the capability to block regions. In 2022, an Infura outage led to disruption of access to Ethereum-based applications. The protocol remained active. User access did not.
Another key concern is the diversity of the client base. When most validators run the same software, bugs can become systemic. Ethereum has been working to reduce this risk. It supports multiple execution and consensus clients. However, client share remains uneven. In 2024, two execution clients maintained their dominant market share. Solana has been subject to more intense scrutiny due to its limited client diversity. The dominance of a single software solution can increase the risk of service interruption.
The introduction of MEV infrastructure serves to further consolidate the market. Block builders, relays, and searchers sit between users and validators. On the Ethereum network, the majority of blocks now pass through MEV-Boost relays. A small number of relays are responsible for most of the flow. This approach, however, carries with it inherent risks of censorship and coordination. Validators generally adhere to relay rules to ensure they do not miss out on potential rewards.
These layers demonstrate the limitations of validator count as a metric. A network can have thousands of validators. It should be noted that control can still cluster in middleware. True decentralisation requires diversity across hosting, software, and access paths. In its absence, failures will rapidly propagate.
Conclusion
Decentralisation cannot be reduced to one metric. The number of validators alone is not a sufficient metric. All of the following aspects are of equal importance: token distribution, governance control and infrastructure dependencies. Each blockchain decentralises different layers to varying degrees.
When comparing networks, it is important to consider a range of factors. The technical structure of a system is indicative of its management. Economic data is a valuable tool for understanding the balance of power. The governance paths illustrate the individuals with the authority to modify the established rules. If any layer is ignored, this may result in false conclusions being drawn.
Decentralisation is an evolving concept. Early networks often start out centralised. Maturity can improve distribution. It has also been demonstrated that increased concentration can be achieved through scale and capital.
"Good enough decentralisation" means that no single actor can control outcomes. This approach ensures that failures remain within the localised area. This indicates that users do not rely on trust.
This content is for informational purposes only and should not be taken as solicitation, recommendation, endorsement or investment advice. It is crucial for you to conduct your own research and due diligence to make informed decisions, as any investment will be your sole responsibility. Please review our disclaimer and risk warning.
