Is there an accurate way to calculate the number of Bitcoin users

[Husires]

Every day I hear many estimates showing the number of Bitcoin users all of them are distracted.

LoyceV says there are about 36 million funded addresses, which means that the number of Bitcoin users is much smaller than that.
Can this number be accurate, especially since the number of users of this forum, Reddit and others equals 5 million users, and if each user has 10 addresses, the number of Bitcoin funded addresses must be higher than 50 million.

According to data analyst Willy Wu, there are an estimated 135 million Bitcoin users worldwide. This is less than 2% of the global population. To put that into perspective, PayPal has 361 million user accounts, Facebook has 2.8 billion monthly active users, and the internet in total has an estimated 4.7 billion users.

To put this into perspective: there are 36 million funded addresses and it's pretty safe to assume many Bitcoin users have more than one funded address. That means he counts at least 100 million "Bitcoin users" who don't own anything on-chain. Maybe he includes custodial wallets, so people speculating on Bitcoin price increases on exchanges or even Paypal nowadays, but I wouldn't call them Bitcoin users.

Remember: "not your keys not your coins"! So if you don't have keys, you can't be own Bitcoin, and if you don't own Bitcoin, you can't be a user.

[1715207828]

1715207828

[1715207828]

1715207828

[1715207828]

1715207828

[1715207828]

1715207828

[ranochigo]

LoyceV says there are about 36 million funded addresses, which means that the number of Bitcoin users is much smaller than that.

That is assuming that there is an insignificant amount of users holding their Bitcoins at exchanges and services.

Can this number be accurate, especially since the number of users of this forum, Reddit and others equals 5 million users, and if each user has 10 addresses, the number of Bitcoin funded addresses must be higher than 50 million.

No. You cannot assume that the active users on Reddit and Bitcointalk are all holding any amount of Bitcoins. Some users may very well just be participating in the discussion or are interested in holding coins other than Bitcoin.

There really isn't any way to calculate the number of unique Bitcoin users due to how its structured.

[NotATether]

No, because each wallet has a variable number of receiving and change addresses associated with it, and there's no way to determine who owns which wallet, it could be that one person has a dozen of these addresses and another person only has three for example. Also to further complicate things people can hold multiple wallets and there are several addresses that belong not to a single person but to a company or service (eg. a casino that gives you a unique deposit address).

Also there are several people who have abandoned their wallets by losing the private keys to them so that's a non-negligible number that belong to nobody.

PS. Please include the source for this so we can see how he got this data:

According to data analyst Willy Wu, there are an estimated 135 million Bitcoin users worldwide. This is less than 2% of the global population. To put that into perspective, PayPal has 361 million user accounts, Facebook has 2.8 billion monthly active users, and the internet in total has an estimated 4.7 billion users.

Remember: "not your keys not your coins"! So if you don't have keys, you can't be own Bitcoin, and if you don't own Bitcoin, you can't be a user.

Why? (Yes I know this is LoyceV's quote) There could be more people who have a bitcoin account on an exchange than there are offline wallets. So while I agree that it can't truly be their bitcoin, I disagree about them not being users.

[Husires]

There really isn't any way to calculate the number of unique Bitcoin users due to how its structured.

No, I am talking about the maximum number. above data told us that the number of 100 million users is an exaggerated number.

PS. Please include the source for this so we can see how he got this data:

I am using my phone browser. it is hard to open many tabs and add quote or source. will update it later

[HeRetiK]

There really isn't any way to calculate the number of unique Bitcoin users due to how its structured.

No, I am talking about the maximum number. above data told us that the number of 100 million users is an exaggerated number.

Probably. But there's no way of knowing since there's presumably more people that keep their funds on exchanges, custodial wallets or even casinos than care to admit.

At face value the number does look exaggerated. Anecdotally speaking I'm pretty sure that much more than 2% of my social circles got into Bitcoin at one point or another though, however there's likely some bias due to demographics (unrelated to me, since I barely talk about crypto IRL).

[NeuroticFish]

1. There's an overwhelming number of so-called Bitcoin users that have their coins only at exchanges, web wallets or micro-payment wallets (especially related to faucets)
2. There are most probably Bitcoin traders that don't care about Bitcoin and they only deposit/withdraw fiat (do they count as Bitcoin users? I'd say yes).
3. I think that the chosen number of 10 addresses/wallet is wrong. I think that people who actively uses the wallet will either have maybe 100 addresses, either only one (not using at all change addresses); on the other hand there are plenty of services having huge number of addresses in their wallets; however, especially because of points [1] and [2] it's hard to determine a good and fair number of users.

Still, I'd say that the number of Bitcoin users is rather "small". We "live" in this "closed box" called Bitcointalk and somehow expect the rest of the world be similar. Well, it's not. Impressive number of people of every kind, color, education, age and so on don't care or don't trust in Bitcoin enough to use it. I would be happy with a 35-50M number, really (that's still the equivalent of Spain's population), and I think that the 135M number is a bit too optimistic.

...But this is only my perception...

[ranochigo]

No, I am talking about the maximum number. above data told us that the number of 100 million users is an exaggerated number.

Hmm? I'm getting the idea that my post answers your topic subject as well as the general direction of your post.

The most accurate estimation of the maximum number as of now is less than 8 billion. It probably is exaggerated but as stated, it doesn't count those who don't store Bitcoins on their own addresses but rather on custodial wallets either for quicker trading or a higher perceived security. It's far more common than you think. I'm going out on a limb and say that each user probably doesn't have 10 addresses with funds at a given moment.

[LoyceV]

Can this number be accurate, especially since the number of users of this forum, Reddit and others equals 5 million users, and if each user has 10 addresses, the number of Bitcoin funded addresses must be higher than 50 million.

I assume "my" number of funded Bitcoin addresses is correct. The data comes from Blockchair.
I don't know about Reddit, but most of the 3 million registered users on Bitcointalk are inactive, banned or waiting for evil fees to be paid. BPIP shows 319,780 active users (meaning they logged in within the last three months). Many of those will be alt-accounts, and I don't expect many bounty spammers with a (centralized) Eth Shitcoin address in their profile to hold any Bitcoin.

I don't think most users will own 10 funded Bitcoin addresses. For many users it's probably a lot closer to only one address. But some people (or services) will have a lot more. Many addresses also hold only dust:

2,821,479 addresses hold 1000 sat or less. Total value: 1,403,085,278 satoshi (~14 BTC).
30,649,056 addresses hold more than 1000 sat. Total value: 1,858,148,094,500,905 satoshi (~18.58 million BTC).

That is assuming that there is an insignificant amount of users holding their Bitcoins at exchanges and services.

I'd love to see numbers for this

There's a much more practical limitation: to use Bitcoin, you need to be able to get a transaction confirmed. Block space is very limited, and the number of possible transactions per day is very low. If 135 million people would "use" Bitcoin, they can't make more than one transaction per 6 months on average. That's like saying I use shorts, even though it's winter now.

There could be more people who have a bitcoin account on an exchange than there are offline wallets. So while I agree that it can't truly be their bitcoin, I disagree about them not being users.

So it comes down to the definition of "a user". In my earlier quote I assumed anyone with a funded Bitcoin address to be a user. I wouldn't go as far as saying that HODLers aren't using it, and someone who uses an exchange to send Bitcoin to another exchange should probably be considered a user too. But I have a hard time calling someone who only trades Bitcoin on an exchange a user.

Food for thought: would you call someone who buys Tesla stock a Tesla user?

Theoretically government could ask all exchange for identity of every user who perform KYC, AML or other identity verification method.

There isn't a single government that holds power over all exchanges.

[HeRetiK]

Food for thought: would you call someone who buys Tesla stock a Tesla user?

You can use a Tesla without owning Tesla stock but you can't use Bitcoin without owning Bitcoin The question would be if you are a Tesla user if you bought a Tesla and then just left it in the garage.

Theoretically government could ask all exchange for identity of every user who perform KYC, AML or other identity verification method. With big amount of data, they could obtain minimum number Bitcoin user (with proper data reprocessing).

Jesus, don't give them any ideas please. The orifices of my financial life get probed enough as it is, thank you very much. Any more and I'll start developing a fetish.

[Theb]

Not unless we have an actual census regarding our real number we won't really know the exact number of Bitcoin users we have. Counting the number of accounts such as Coinbase, Binance, or even Bitcointalk accounts would be irrelevant and highly inaccurate since there are chances that 1 person can own multiple accounts and it doesn't actually depict the real number of us users. The same goes with the number of wallets created as one person can own and create multiples of it. So if we really want to know an accurate number of us users it must be done through a census and not through digital data we have. Maybe one day if KYC might be strictly enforced all through out the world for the crypto industry then we can go to the digital route for statics but until then we can't rely on any existing data.

[DannyHamilton]

There's a much more practical limitation: to use Bitcoin, you need to be able to get a transaction confirmed. Block space is very limited, and the number of possible transactions per day is very low. If 135 million people would "use" Bitcoin, they can't make more than one transaction per 6 months on average. That's like saying I use shorts, even though it's winter now.

As you've pointed out, the key here is first to define what it means to "use bitcoin". Once that is accomplished, and everyone in the discussion agrees on the definition being used, we can begin to consider how that number might be estimated.

For the temporary purposes of this post alone, I'm going to use a definition of a "user" of bitcoin being:
Someone that either has engaged in an on-chain transaction within the past 365 days (consumers) OR is holding some amount of bitcoin in an address that they have direct control over (holders).

I'm aware that this leaves out "holders" that are "holding" their bitcoins in a custodial service, as well as those "consumers" that engage in off-chain transactions (either Lightning Network or transactions entirely within a custodial service where both the merchant and consumer are using the same service)

I'm too lazy to count a year's worth of transactions (or even to Google to see if anyone has a count posted somewhere), but I've got enough motivation to make up an estimate that "feels right" in my head.  Let's say an average of about 3000 transactions per block?

Using my arbitrary definition, it seems that we should be able to just add up the number of blockchain transactions, and add to that the number of funded addresses to achieve an upper bound:
3000 (tx/block) X 52,560 (blocks/yr) = 157,680,000 unique transactions per year.
Add to this the 36 million funded addresses mentioned earlier and we have an upper bound of 193,680,000 "users".

If we want to refine that upper bound a bit, we could throw out addresses that are funded with less than some minimum threshold, as well as transactions that use previously used inputs. We could also eliminate from the "funded addresses" count any address that was used in a transaction in the past year (since that user will have already been counted in the "transactions" portion of the formula, and we don't want to count them twice).