The 12 documents your bank will ask for

You have decided to move crypto into a bank account. The money is yours and you earned it.

The question is whether a bank can see for itself where the money came from. Not whether it believes you - whether it can check. And an unverifiable file is worse for you than an incomplete one. If your explanation conflicts with the underlying data, the bank now has an inconsistency it needs to investigate rather than a question it can close, and that is a slower and less comfortable conversation than the one you were trying to have.

Below are twelve areas a bank may need to understand in a complex crypto source of wealth review. Not every case touches all of them, and no two banks ask in the same words. Eleven are evidence. The twelfth is the only one you have to write yourself. It comes last in this list, but it is the one to write first. Two things are assumed to be on file already: a valid identity document and a recent proof of address.

1. Exchange history that comes from the exchange

Every account you have used, including the dormant ones.

Whether supported or not, offer the reviewer access to source-authenticated exchange history rather than a spreadsheet you generated yourself. An export you produced is a file you could have edited, and several banks say so out loud when they decline one. What settles a question is history the exchange itself stands behind, arriving with the platform’s authentication behind it instead of yours.

In practice that is a client-authorised connection: read-only, created by you rather than on your behalf, withdrawn when the review closes. The timing matters more than the mechanics. You set nothing up in advance and you attach nothing. What belongs in your file is one sentence saying source-authenticated access can be provided on request, for every account on your list. If it is wanted, the bank asks, and it is arranged then, through whatever secure route the bank or the firm preparing your file uses.

Keep your own copies if you like. If a platform closes, your export may be the only thing left of it. Just do not send them unasked.

A note on margin and futures. This is the weakest part of every exchange’s data, by connection and by export alike. Profit and loss on leveraged positions is often missing, partial, or expressed in a way nothing else reconciles against. Only at a few exchanges do the numbers come out whole. If you traded on leverage at any scale, expect manual work here, and expect it to be the part where professional help pays for itself.

The common mistake. A screenshot of the exchange profile page showing your name. In 2026, a picture of a name on a screen proves nothing, as AI can create it in seconds. A verification e-mail from the exchange is better, but it still does not tie you to this account rather than to some account.

2. Every wallet address you control, or controlled

Give the chain and the address together, as a pair. A bare list of addresses makes the reviewer work out which network each one belongs to before they can look at anything, and that is work you can remove with one extra column.

Include the empty ones, the abandoned ones and the wallet on the phone you replaced.

What makes the list convincing is not a signature on each address. It is that the addresses connect: to each other, and to exchange accounts already verified in your name. A wallet sitting inside that web is yours in a way no document can assert on its own.

Some banks will still want the ownership proved directly, because their policy says so, or where it cannot be established from the surrounding transaction history. Wait until you are asked. It may never be needed, and if it is, the bank may want a particular method, such as a transfer of a specific amount, that nothing prepared in advance can supply.

On extended public keys. A wallet’s extended public key exposes every address it has ever generated, and that is usually the point, because it is what lets the counterparties behind your history be examined properly. You can decide not to share it and to give specific addresses instead, for privacy reasons, provided you say plainly that this is what you are doing. Where the part you have held back is not decisive to the story, that explanation is often accepted.

The common mistake. The wallet emptied in 2019 and forgotten. It is the one most likely to be missing from the list and the one a reviewer is most likely to find anyway, because the coins that left it went somewhere.

3. Nothing to export from the chain

The blockchain is public. Given your addresses, the history can be rebuilt from the chain itself, independently of anything you supply, using tools that cost nothing. That reconstruction is worth more than a spreadsheet you exported, because it does not depend on you being right or on the file being untouched. Your addresses matter. Your homemade blockchain spreadsheet does not.

This item is also a test, and it runs in the other direction. If your bank is not able to use a public block explorer, it is not really able to start work on your case. That is not a criticism of the person reading your file. It is the reason files sit for months without a question being asked.

The common mistake. Treating a blockchain export as the deliverable. The addresses are the deliverable.

4. Bank statements for the fiat that left, and for the account it is coming back to

These are the bridge between the regulated world the bank understands and the one it cannot see. Official PDF statements for the months in which you funded the exchange, from the first purchase to the last, plus the account the proceeds will arrive in.

A statement is a PDF and a PDF can be edited too, which is a fair objection and a smaller problem than it sounds. There is a good chance the bank receiving the money is the same one that sent it out years ago, in which case it is checking its own records. That is the practical argument for off-ramping at the bank that on-ramped you, particularly where the gains are not extreme. Where it is a different bank, banks have their own channels for confirming a payment with another institution.

There is also something on those statements worth using rather than burying. Banks look at what went out against what is coming back, as a ratio. If your own figures make that relationship look sensible, put them in front of the reviewer rather than waiting to be asked.

The common mistake. Sending only recent statements because the old ones are harder to find. The desk asks about the first purchase, not the last. The other one: a funding account that is not the receiving account, because the money left account A in 2017 and is arriving in account B in 2026. That is fine. It has to be said rather than discovered.

5. Where your first euros came from

Crypto did not create your wealth out of nothing. You bought it with money that came from somewhere: salary, savings, a property sale, an inheritance, a business. This is source of wealth in the strict sense, and it is the part almost nobody prepares.

Payslips or an employment contract from the years you were buying, an income tax assessment for those years, the notary deed for a sale or an inheritance, the share purchase agreement for a business.

Not all of it went through a bank. Crypto bought at an ATM, on a card, with a prepaid voucher or in cash leaves no statement anyone can pull, and those purchases are the ones people forget to mention because there is nothing to attach. Whatever survives is worth keeping: the ATM receipt, the line on a card statement, the voucher record. Where nothing survives, say so and say roughly when and how much, because an amount that appears on chain from nowhere is the question you want to have answered in advance.

The common mistake. Answering “where did the 20,000 come from” with “my bank account”. The bank can already see the bank account. It wants to know how the money got into it, which means the trail has to reach one step further back than the exchange deposit.

6. Coins you did not buy

Mining, payment for work, a gift, an OTC deal, a loan repaid in crypto. All legitimate, and from the outside all identical: coins arriving from an address that is not yours.

Each needs its own evidence. The invoice or contract if you were paid in crypto. A written gift record, with the giver’s own trail behind it. A written OTC agreement naming the counterparty. Payout history from the mining pool.

The common mistake. Calling every unexplained inflow an airdrop. Reviewers have heard it. The second: a gift with no paper. “My brother sent me 3 ETH” is a sentence, not evidence, and the bank may end up asking about your brother’s wealth as well.

7. Staking, liquidity pools and what sits inside a contract

Documenting a DeFi history is tricky every time. Not so much the swaps, which are visible enough, but everything around them: the stretches where value left your wallets and sat somewhere else for a while.

Rewards are visible when they are paid. What you had staked, where, and for how long is not. That history lives inside the contracts, and reconstructing it means going through them period by period. Current positions can be read straight off. The history behind them has to be rebuilt, and that is a different kind of job.

The common mistake. Assuming that because today’s balance is visible, the history behind it is too. A position is not an explanation, and a balance that appears without one is exactly the thing a reviewer stops at.

8. The platform that no longer exists

No connection, no export, nothing to authorise. What survives is your bank statement showing the money leave, whatever confirmation e-mails you kept, and the chain record of coins arriving somewhere if you ever moved them off.

So build the list yourself, from the two sides that still exist. For fiat, every deposit and withdrawal: date, bank, amount. For crypto, every transfer in and out: date, network, the address at the other end, amount. That list is the reconstruction, and it is what a reviewer can actually work from.

Search old mailboxes for the rest. Then say plainly what cannot be documented, and why.

The common mistake. Silence. Banks understand that platforms fail. An unexplained gap where a platform used to be is a different matter entirely.

9. Accounts that are not only yours

An account opened by a partner, a parent or a friend, because they already had one, turns your file into theirs. Banks routinely ask for a separate source of wealth file from each person involved rather than one that covers both.

Say so in your claim, at the start.

The common mistake. Hoping it will not come up. It comes up at the point where the money moved, which is the point a reviewer looks at hardest.

10. The sale and the withdrawal

This is the transaction the bank is about to receive, and the one it works backwards from if the money arrives before the file does. Which coins were sold, at what price, on which platform, on which date, and then the withdrawal: what amount, to which account, in whose name.

The common mistake. Cashing out by a route nobody can follow. A peer to peer platform, a crypto card, a friend’s account, or twenty small transfers instead of one. Small repeated transfers are not invisible; they trigger a different kind of attention. Withdraw to an account in your own name, in as few transfers as make sense, and tell the bank before the first one lands.

11. Tax, if it applies to you

Nothing here is tax advice. Whether anything is owed depends on your country, your holding period and what you actually did. Someone who bought one Bitcoin and never sold it may have nothing to declare at all, and a file built on the assumption that tax documents are always required wastes weeks proving something that was never in question.

What matters for the bank is consistency. If a return exists, it should agree with everything else in the file. If none exists, the reason belongs in writing, because a blank reads as an omission rather than an exemption.

Where returns do exist they come from the national portal, with the assessment that followed: MyMinfin in Belgium, ELSTER in Germany, impots.gouv.fr in France. Those are examples, not a list, and every country does this differently.

If the tax side is not settled yet, one set of connected data can produce both a tax calculation and the evidence behind a source of wealth file. How that works, and which calculators it runs through, is set out on the tax page.

The common mistake. “I did not have to declare anything,” left unexplained. Sometimes it is correct. It still has to be written down, with the reason.

12. Your claim

Eleven areas are evidence, and not one of them says what happened. The twelfth is the two pages only you can write: when you first bought, with what money, what happened since, what you hold now, and what you want to do with it. Everything else on this list exists to be tested against it. How to write it, and what goes on each of the two pages, is set out separately.

What connects the claim to the evidence is a picture of how value actually moved: which accounts, which addresses, in which direction, and where it entered or left. A reviewer who can follow that does not have to read eleven attachments in sequence to understand the story. That picture is what the interactive Source-of-Wealth map is, and it is the reason a good file is read rather than weighed.

A file assembled after the questions start arrives late and answers only what was asked. One prepared before the money moves answers the question first - or arrives with the notice that the money is coming, and puts control of the conversation in your hands.