Begin with the person and the period.
Consider Mara, a fictional US citizen who has lived and worked in Colorado throughout the relevant years. She bought Bitcoin through an exchange that later closed, withdrew it to self-custody, consolidated several wallets, and sold part through a different broker. Her current wallet opens normally. The old exchange login does not. A software import therefore begins with a deposit whose acquisition cost is blank.
Those facts make US federal guidance relevant to this example. They do not establish a rule for every holder with the same wallet software. Someone who moved countries, received units as compensation, inherited them, or held them for a company has additional questions. Before collecting prices, write down the owner, acquisition periods, residence history, and proposed reporting years. This stops a technically impressive reconstruction from answering the wrong taxpayer’s problem.
The IRS treats digital assets as property for federal income tax purposes and requires records that support reporting. That framework makes both acquisition and later disposition relevant. It does not allow a blockchain explorer to supply facts that the chain never recorded, such as the identity of a purchaser or the terms of an off-chain trade. Our task here is to prepare evidence for the applicable analysis, with conclusions left to the scoped review.
Preserve originals before improving the spreadsheet.
Mara starts with an evidence inventory. She has bank statements, a few emailed trade confirmations, withdrawal notifications, two wallet exports, and the new broker’s sale confirmation. She saves the original files before renaming columns or converting timestamps. An index records where each file came from, when it was obtained and the account or period it covers. A screenshot may help explain a screen, but it should not replace an available underlying export.
Keep a working copy separate. Changes belong there: normalizing a date, mapping an account nickname, adding a transaction reference, or proposing that two entries describe the same movement. A simple change log is enough if it records the old value, proposed value, reason, and supporting file. Silent corrections create a second evidence problem when the preparer cannot reproduce the result.
Access boundaries matter too. A preparer may need transaction data and a statement explaining ownership. They do not need seed words or a signing key to reconstruct a purchase. Export the relevant records through an agreed secure channel. If a public address connects other family or business activity, discuss the necessary scope before distributing a complete wallet history to everyone involved.
Trace the holdings through twelve records.
The following ledger is fictional. Quantities illustrate how records connect; no historical market prices are supplied. A reference to a bank amount means the amount is retained in the source document, not estimated from a chart. Blank acquisition fields remain blank until evidence or an explicitly reviewed reconstruction supports them. The twelve rows are evidence events, not twelve taxable disposals.
Notice what the bank transfer does not prove. A payment to an exchange could fund several trades, cover fees, or remain as cash. Allocating the whole transfer to Bitcoin would create a false purchase cost if any of those alternatives occurred. Likewise, the withdrawal quantity cannot automatically be treated as the purchased quantity. Withdrawals can aggregate earlier purchases and deduct charges.
Mara’s ledger gives each source event its own identifier, then links related events. It does not force every source row into a single accounting category immediately. That makes it possible to correct one connection without rebuilding the entire file. The sale can remain linked to a pool of candidate acquisition records while the applicable identification question is reviewed.
| Row | Event or record | Units or amount | What it supports | Open question |
|---|---|---|---|---|
| 01 | Bank transfer to old exchange | Fiat amount in bank record | Funding left Mara’s account | Was the whole amount used to buy Bitcoin? |
| 02 | Purchase confirmation A | 0.60 BTC | Acquisition quantity and trade reference | Locate complete charge and fee details |
| 03 | Purchase confirmation B | 0.40 BTC | A second acquisition event | Confirm execution currency and timestamp |
| 04 | Exchange account statement fragment | Fiat balance shown | Some funding remained unspent | Match balance to the funding period |
| 05 | Exchange withdrawal notice | 1.00 BTC requested | Movement out of exchange custody | Reconcile withdrawal fee and net units |
| 06 | First wallet receipt | Net units in wallet export | Receipt at an identified address | Connect address to Mara’s ownership |
| 07 | Second wallet contribution | Separate identified holding | Additional units entered consolidation | Preserve its different acquisition history |
| 08 | Consolidation transaction | Inputs and outputs recorded | Wallet reorganization | Identify change and transaction fee |
| 09 | Transfer to new broker | Units in transfer record | Movement into custodial account | Match broker deposit reference |
| 10 | Broker deposit credit | Net credited units | Broker recognized the deposit | Explain any timing or quantity difference |
| 11 | Partial sale | 0.25 BTC sold | Disposition quantity and proceeds | Determine applicable identification and cost |
| 12 | Remaining position check | Reconciled unsold balance | Continuity after the sale | Carry forward supported acquisition links |
Separate three kinds of missing evidence.
Missing price evidence is different from missing quantity evidence. If the quantity and exact trade time are known but the confirmation omitted the purchase amount, bank movements, surviving order emails, and counterparty records may narrow the inquiry. A historical price series could be relevant to a documented reconstruction, but it is not the original execution record and should never be presented as one.
Missing quantity is harder to repair with a fiat payment alone. A bank transfer does not establish how many units were bought, especially where multiple trades occurred. Seek order history, withdrawal records, and account balances that constrain the answer. If the resulting quantities still cannot reconcile, leave a difference rather than create a balancing purchase called miscellaneous. That invented event will contaminate subsequent holdings and sales.
Missing owner evidence raises another problem. A transfer into a wallet Mara controls might represent her purchase, a spouse’s property, company funds, or a loan repayment. The chain does not decide which explanation applies. Relevant agreements, account names, and contemporaneous correspondence may establish the context. If ownership remains disputed, identify that issue before treating every input as Mara’s personal tax basis.
Give each uncertainty a responsible owner.
An uncertainty register prevents a reconstruction from becoming a folder of unanswered emails. Each entry identifies the affected records, competing explanations, the next action, and who will decide whether the remaining evidence is sufficient. Confidence labels describe the evidence, not the desired tax result. A well-supported high-cost figure and a well-supported low-cost figure should receive the same evidential treatment.
Set a review date for each open item. The aim is not endless searching for a perfect export that may no longer exist. It is a proportionate account of what has been obtained, what remains possible, and what the professional needs to decide. A failed request is evidence of the search, not evidence of the missing trade itself.
| Gap | Current evidence | Alternative explanation | Next action | Decision owner |
|---|---|---|---|---|
| Purchase A fee missing | Confirmation and matching bank funding | Fee paid separately or deducted from units | Request archived statement and reconcile balance | Preparer reviews permitted treatment |
| Withdrawal quantity difference | Requested amount exceeds wallet credit | Provider fee, separate transfer, or incomplete export | Compare notice, transaction, and account history | Mara gathers; preparer assesses |
| Consolidation input ownership | Wallet label says family | Spouse or company may own part | Locate acquisition and ownership records | Relevant owner and professional |
| Trade time uncertain | Email received after apparent execution | Email delivery delay or different timezone | Preserve original headers and provider definitions | Preparer selects supported timing |
| No old exchange response | Closed account and request evidence | Archive may exist with successor or administrator | Make a bounded documented follow-up | Mara records outcome |
Test the import before trusting its totals.
A common failure begins when the old exchange withdrawal and the wallet receipt are imported as unrelated events. Software may label one as a disposal and the other as an acquisition with no cost. A second import of the same wallet can then double the units. Review event identifiers, asset quantities, and account boundaries before accepting a gain calculation. The first test is whether the holdings reconcile, not whether the tax number looks plausible.
Reconcile opening units, additions, removals, and closing units for each account, then across the owner’s accounts. Explain fees separately. Treat a transfer classification as a conclusion supported by common ownership and connected records, rather than by a convenient matching amount alone. Two unrelated events can have identical quantities. Conversely, a genuine transfer may have different sent and received quantities because a fee intervened.
Save the software version, settings, and import files used for a calculation. Unit identification and fee treatment require rules appropriate to the taxpayer and year; they are not merely preferences in a menu. Ask the preparer which choices have been made and how they interact with records from earlier returns. Changing a setting can alter the result without changing a single transaction, which is precisely why the setting belongs in the handoff.
Hand over a reviewable conclusion, including the limits.
The handoff should contain a short narrative, the twelve-event ledger, an original-file index, a transaction reconciliation, and the uncertainty register. Distinguish documented facts from reconstructed entries and professional conclusions. A reconstructed row should identify its method, inputs, and remaining limitations. Avoid a generic note saying records reconstructed where only some rows need that description; the reviewer needs to know which ones.
Ask for decisions on the applicable identification method, acquisition costs, transaction charges, and any unresolved ownership question. Also ask whether the new evidence affects a previous filing. Discovering an old confirmation can change more than the current sale if earlier disposals used the same holdings. Preserve the earlier version and let the responsible professional consider whether any correction is needed rather than silently overwriting the history.
The file is ready for review when another person can follow the principal acquisition-to-sale links and see exactly where they stop. It need not pretend to contain every missing fact. An unexplained zero is not a conservative substitute for an evidence gap, and an attractive estimate is not a recovered record. Both obscure the actual question the preparer needs to resolve.
One final cross-check is useful: ask a reviewer to select a disposal row at random and trace it backward without explanation from the person who built the file. If the references stop at a nickname, an unsupported balance, or an unavailable account, improve that connection or mark the limitation. A reconstruction that can be followed independently is easier to assess than one that works only while its author narrates it.
The whole picture
The Bitcoin Lawyer take
Mara can support parts of the history from independent records even after the exchange closes. She should hand over each supported quantity with its date, fees, and source document.
Leave the unresolved purchase price visible. The preparer can assess what the evidence supports under the applicable rules before relying on it in a return.


