a16z: Distinguishing the true from the false, estimating the number of real active users in crypto to be between 30 million and 60 million
This number accounts for only 14% to 27% of the 220 million monthly active addresses we measured in September.Authors: Daren Matsuoka & Eddy Lazzarin, a16z
Compiled by: Deep Tide TechFlow
In writing our 2024 Crypto Industry Report, our team spent considerable time assessing the scale of the crypto industry. As the industry matures and more applications come online, we want to understand how many people are actually using cryptocurrencies. This is a complex question because the most obvious and easily quantifiable usage metric—active addresses—can easily be manipulated. Here are our thoughts.
In the traditional software space, the concept of "users" is quite clear. Of course, there are many ways to measure the quality of users—indeed, the entire field of growth analytics studies this issue—but at the most basic level, users can be counted as "daily active users" (DAUs), "monthly active users" (MAUs), and so on.
In the crypto space, however, the situation is much more complicated. This is because user identities on the blockchain are pseudonymous. A person can easily create and manage what is known as a sybil on the blockchain—a set of different identities referred to as "public addresses." (There are many perfectly legitimate reasons for doing this, such as privacy, security, or other purposes.) Therefore, it is difficult to determine how many addresses a single person might use. (The reverse is also true, as multiple people can share an address through multi-signature, aggregated accounts, and various account abstraction protocols.)
Until recently, the most popular blockchains had very limited capacity, leading to high transaction fees. This naturally created a barrier to people creating and using hundreds or thousands of addresses, as doing so was costly. However, recently, with the emergence of L2 rollups and new high-throughput L1s, crypto infrastructure has become more scalable, making transaction costs on many blockchains nearly zero.
In traditional internet applications, isn't the cost of creating multiple identities also nearly zero? This is true in most cases. For example, a person can easily create and use multiple email addresses. But the key difference is that in the crypto space, there are strong incentives for this behavior.
The crypto industry has long rewarded early users through token incentive programs. Nowadays, new protocols often launch their tokens into circulation through "airdrops"—a reward activity that provides token incentives to a predefined set of addresses. These address lists are typically derived from historical on-chain transaction records. Some individuals may attempt to manipulate the system by creating multiple different identities and leveraging them for transactions. This strategy is known in the industry as "airdrop farming."
Given these behaviors, it is clear that the 220 million unique monthly active addresses we measured in September 2024 do not directly equate to 220 million people or users. (It is worth noting that addresses active across multiple EVM chains are only counted once in the total of 220 million.)
So, what is the actual number of active users? Is it 10 million, 50 million, or 100 million? This is precisely the question we aim to explore. Here is our methodology.
Method 1: Filtering Active Addresses
One approach we adopted is to filter out addresses suspected of being controlled by bots or belonging to sybil attacks. Through on-chain analysis and forensics, we explored various methods to achieve this:
Filter out addresses whose funds come from decentralized contracts—decentralized contracts are a type of smart contract whose sole purpose is to receive funds and automatically distribute them to multiple different addresses. While there may be some false positives, this activity indicates that these target addresses are receiving funds from the same source and are therefore somewhat interconnected.
Filter out addresses with balances close to zero at the beginning and end of a specific time period. For example, if you are looking for real monthly active users in September 2024, you might try to exclude addresses with balances close to zero on September 1 and September 30. This criterion suggests that these addresses are temporary. While bots and sybil attackers may attempt to "clean up" their balances after operations, real users typically keep some balance in their wallets to pay for future transaction fees.
Analyze the distribution of addresses that conducted one, two, three, four, five, or more transactions during a specific time period. Addresses with only one or two transactions during this period are at best low-quality users and at worst could be bots or sybil attackers. This method works best in longer-term aggregated analyses.
Filter out addresses that conducted a large number of transactions in a very short time. Human users can only reasonably handle a certain number of transactions within a given time when using wallets or application interfaces, while bots can transact at a higher frequency.
From an optimistic perspective, consider including addresses associated with identity protocols that typically require some setup cost. For example, addresses with ENS names, Farcaster IDs, and other associated social identities are likely to be actual human users.
Method 2: Inferring from Wallet Users
Another way to estimate the number of monthly active users is to look at off-chain data sources, with wallet users being the most obvious starting point.
In February 2024, the popular crypto wallet MetaMask reported that its monthly active users reached 30 million. They defined monthly active users as "users who loaded the MetaMask extension page or opened the mobile app at least once in any rolling 30-day period."
If we want to estimate the actual number of users conducting transactions, the next step is to determine what proportion of MetaMask users ultimately conduct transactions. According to data from 2019, MetaMask reported that on any given day, about 30% of active users would confirm an on-chain transaction. (This is the most recent estimate available.) If we apply this ratio to the monthly active users (MAU), then approximately 9 million users conduct transactions monthly through the MetaMask wallet product.
Next, we need to understand MetaMask's overall market share among wallets across all blockchains. While there is no direct precise data, we can make some reasonable estimates based on known information. For example, we can estimate MetaMask's share of the mobile wallet market based on data from mobile analytics company Sensor Tower. (Due to commercial service agreements, we cannot disclose specific numbers here.)
Once we estimate MetaMask's market share, we can simply extrapolate the total number of crypto users based on the previously derived 9 million monthly active transaction users. We can then compare this estimate with the results from Method 1 to see if they fall within the same range.
To further refine our estimate, we can analyze data from other wallet and infrastructure providers willing to share their proprietary data and cross-validate it with the above data.
Other Considerations
It is important to note that some users will use and transact with multiple addresses and wallets. While this is unlikely to significantly increase the total number (because, unlike bots and sybil attacks, the number of wallets a person can reasonably use is limited), further deduplication based on reasonable assumptions may be worthwhile.
On the other hand, there is also a situation where a single address may be associated with multiple users, such as in the case of exchange aggregated accounts. By the way, as account abstraction protocols and smart contract wallets become more prevalent, this situation will become more complex. We have not included these factors in our analysis.
Estimation Results: 30 Million to 60 Million Real Monthly Active Transaction Users
Based on our analysis using multiple methods, we currently estimate that there are 30 million to 60 million real monthly active crypto users. While this range is broad, it is the best estimate derived from the existing data.
Please note that this number represents only 14% to 27% of the 220 million monthly active addresses we measured in September. At the same time, it also accounts for only 5% to 10% of the 617 million global cryptocurrency holders reported by Crypto.com in June. (Global cryptocurrency holders refer to those who own cryptocurrencies but may not necessarily transact on-chain.) This gap indicates a significant potential to convert existing cryptocurrency holders—most of whom are passive holders—into active users. With improvements in major infrastructure, enabling new and engaging applications and user experiences, those dormant cryptocurrency holders may become active users on-chain once again.
Accurately measuring the number of active crypto users is indeed challenging, but by employing some of the methods detailed in this article, we can begin to arrive at a reasonable estimate. This article is our attempt to share our thought process and calculations. These methods will continue to evolve over time, and we will keep sharing our latest insights.
