The Privacy You Give Up When You Use a Public Address

Most guides handle this by reassuring you that blockchains are pseudonymous and moving on. That is technically accurate and practically misleading, because pseudonymity is a property that degrades on contact with normal life, and it degrades permanently.

Here is the version that respects your intelligence: a wallet address is a published financial record with your holdings in it, and the only thing protecting you is that nobody has connected it to your name yet. That protection has no lock on it. It is a fact about what other people currently know, and facts like that only ever move in one direction.

What is actually published when you use a wallet?

Everything except your name.

An address's current balance, every token it holds, every transaction it has ever sent or received, the exact timestamp of each, the amount, the fee, and the address on the other side. Free, instantly, to anyone on earth, through a block explorer like Etherscan or mempool.space. No account required, no permission asked, no record of who looked.

It is also permanent and retroactive. Transactions from 2017 are as readable today as this morning's, and they will be readable in twenty years. There is no expiry, no right to erasure that a chain will honour, and no way to withdraw something once it is confirmed.

Account-model chains make this worse than Bitcoin does. An Ethereum address is typically reused indefinitely, so a single string accumulates your complete financial history in one place. Modern Bitcoin wallets generate a fresh receiving address for each payment, which helps, though the common-input-ownership heuristic reassembles most of it: when a transaction spends from several addresses at once, those addresses almost certainly share an owner. Analysis firms have run that clustering at scale for a decade, and it works.

The same-address problem repeats across chains. An EVM address is identical on Ethereum, Arbitrum, Base, Polygon and every other chain sharing the format, so activity on one is trivially joined to activity on all of them. Tracking a multichain identity treats that as a portfolio convenience, which it is. It is also a privacy consequence.

Why is linkage the real problem?

Because an unlinked address is a number, and a linked address is a bank statement with your name on it.

The link gets made by ordinary behaviour, usually more than once. Withdrawing from a verified exchange to your personal wallet writes the connection into that exchange's records permanently, and those records are subpoenable, breachable and in some jurisdictions routinely reported. Registering an ENS name attaches a human-readable label. Posting a donation address, displaying an NFT bought with the same wallet, claiming an airdrop that requires a social account, signing into a site that also has your email, or paying a person who knows who you are all do the same job.

A single screenshot is enough. So is being on the receiving end of a transaction from someone who later gets compromised.

Once any one of those links exists, everything follows from it. Your balance is known. Your buying and selling is known, including the timing, which means someone can reconstruct how you behaved during a crash. Counterparties are known, so your associations are visible. And the linkage propagates: addresses clustered with the linked one by transaction patterns inherit the identification.

There is a physical dimension worth stating rather than skirting. Publicly known large balances have led to targeted robbery and coercion of individuals. It is a small category of harm and a severe one, and it argues specifically against advertising a linked address alongside a visible balance.

Dusting attacks exist to manufacture linkage cheaply. Send a trivial amount of a token to thousands of addresses and watch which ones later combine that dust with other funds, revealing shared ownership. This is one reason not to interact with unfamiliar tokens that appear in your wallet, on top of the more common reason that many of them are approval traps. Spam and dust tokens covers filtering them out of your view without touching them.

Which mitigations actually work?

Ranked honestly, including what each costs.

Address separation by purpose. The most effective thing available to an ordinary user, and it requires only discipline. The principle: an address that is publicly attached to you never holds meaningful value, and an address that holds meaningful value never touches anything public.

1. Designate a public-facing address. ENS name, donations, NFT displays, anything you post. Keep a small working balance in it and expect it to be fully attributed to you.
2. Keep a savings address that receives nothing public. No ENS, no NFT displays, no site logins where you have an account, no direct withdrawals from a KYC exchange if you can route otherwise.
3. Never combine funds across the two in a single transaction. On Bitcoin this is what unites them under common-input-ownership. Wallets with coin control let you enforce it.
4. Use a fresh receiving address per payment where the wallet supports it, which most Bitcoin wallets do by default and most EVM wallets do not.
5. Keep a written note of which address does which job. The failure mode of this scheme is not attack; it is forgetting and using the wrong one. Multi-wallet setup covers keeping several addresses legible.

Not withdrawing from a verified exchange straight to long-term storage. The exchange holds your identity documents and the destination address in the same record. Understand that you are choosing to create that link when you do it, and that no later action removes it.

Silence. Not posting screenshots, balances, transaction hashes or addresses. Unglamorous, free, and more effective than most technical measures.

Mixing and privacy protocols. These genuinely break on-chain linkability, and the trade is real: several major exchanges flag or freeze deposits with mixing history, and sanctions actions against mixing services have created live legal questions in some jurisdictions. This is not a reason to dismiss the category. It is a reason to read the current position where you live before using one, rather than after.

What does not work is worth listing too. Switching wallet software changes nothing, because the address comes from the keys. Moving funds between your own addresses creates a traceable edge rather than erasing one. Bridging to another chain is followed routinely. And assuming a modest balance is beneath notice ignores that clustering is automated, indiscriminate and permanent.

What does this mean for tracking a wallet in an app?

It means the tracker is not the exposure, and treating it as one gets the risk backwards.

Adding an address to a portfolio app publishes nothing, because the address was already public and its history was already readable by anyone who cared to look. What the app adds is a link inside one company's database between that address and your email account. That is a genuine cost and a bounded one, and what a tracker can see sets out exactly how far it extends.

If that cost is more than you want to pay for a specific holding, the alternative is straightforward. Record it as a manual position with quantity and acquisition price. Your total stays correct, your allocation stays correct, your cost basis stays correct, and no address enters the picture. Long-term cold storage is the obvious candidate, and tracking a hardware wallet covers the choice between watching its address and entering it by hand.

The framing that holds up is this: privacy on a public ledger is not a setting you enable. It is a set of links you decline to create, one transaction at a time, with the knowledge that every link you do create is permanent. Decide now which addresses are allowed to be you.