Best Crypto Audit Software for CPA Firms Auditing Bitcoin

This guide is for audit seniors and managers at US firms who picked up a client with bitcoin on the balance sheet. Calendar-2025 year ends were the first full annual periods under ASU 2023-08 for most companies, so plenty of firms met the question for real this spring: what do we actually use to test this line? We looked at the tools CPA firms are offered, checked every published price on 21 September 2026, and ran the two tests that matter on real chain data so you can see what good evidence looks like.

We sell Bitcoin on-chain data, so weigh that. Where another tool is the better fit we say so.

What should crypto audit software do on a bitcoin engagement?

Four jobs, and the software you choose has to cover the ones your procedures rely on. Existence and quantity: agree the client's reported units to the chain at the cutoff. Rollforward: tie opening balance plus receipts minus disposals to the closing balance for every address. Completeness: find activity the client did not report, which usually means addresses that are missing from the list they gave you. Reproducibility: every figure in the file should come with the address and block height, so a reviewer or a PCAOB inspector can rerun it and land on the same satoshi.

Ownership is the fifth question, and no data tool answers it. Control of the keys is evidenced by the client signing a message or moving a small amount on request. Keep it in its own workpaper. A clean balance on an address the client cannot prove it controls is not their asset.

Which block is the bitcoin year-end cutoff for 31 December 2025?

Bitcoin has blocks, not dates, so "the balance at 31 December" has to become "the balance after block N". We looked up the last block before 23:59:59 on 31 December 2025 in three time zones a US client might close in.

The Eastern row has two answers, and that is the finding. A timestamp lookup on a public explorer returns block 930,371. But blocks 930,372 and 930,373 are also stamped before midnight Eastern, because block 930,372 carries a timestamp 273 seconds earlier than the block before it. The protocol allows that: a timestamp only has to beat the median of the previous eleven blocks. So "the last block before the cutoff" can mean the first block stamped after it, minus one, or the highest height stamped before it, and the two rules put 8,821 transactions in different fiscal years. Neither rule is wrong. The client's close memo should name one, and your workpaper should record the height and hash that rule produced.

For a cold wallet that did nothing on New Year's Eve, none of this moves a number. For an operating wallet that pays vendors or takes customer payments, it can, and a difference between the client's schedule and your recomputation is often a cutoff rule rather than an error.

Quarter ends are not always that messy. At 30 September 2026 no timestamp near any of the three midnights ran backwards, so the cutoffs are clean: block 969,369 in UTC, 969,391 in Eastern (EDT) and 969,409 in Pacific (PDT). The gap is still 40 blocks and 179,384 transactions between the UTC and Pacific answers, so the rule still belongs in the workpaper. The same convention has to drive the price you test: we measured more than $500 per bitcoin between a 4:00 pm Eastern and a midnight Pacific price that night, set out in our bitcoin fair value accounting guide.

What does a bitcoin rollforward that ties out look like?

Here is one we ran on 21 September 2026 on a large custodial cold address, 3M219KR5vEneNb47ewrPfWyb5jQ2DjxRP6, between blocks 967,238 and 968,066.

Opening plus the three receipts equals the closing balance exactly. Two details are worth pointing at, because both show up in real engagements. The 1,281-satoshi receipt came from an address nobody at the entity controls. It is real, it is in the balance, and it is immaterial, but a client that books only the transactions it initiated will be off by that amount, and you want to recognize the pattern rather than chase it. And a further 546-satoshi payment was sitting unconfirmed when we read the balance. It is correctly outside the closing figure. An explorer screenshot can show it next to the balance, which is one reason screenshots and height-pinned queries disagree.

The flow and the balance were computed separately, which is the point. When they tie, the rollforward checks itself. When they do not, the gap tells you where to look: a missing address, a movement in the wrong period, or a cutoff rule applied inconsistently. The Bitcoin audit trail API page shows how that query runs across a whole address list.

Much of the audit evidence a client hands over arrives as a wallet statement rather than raw data, and statements built by different tools disagree on cutoff time zone, change and whether they cover one address or the whole wallet. Our guide to Bitcoin wallet statement tools for accountants and lenders shows how far apart two honest statements for the same wallet can be.

Best crypto audit software for CPA firms, compared

Prices below were checked on each vendor's own pricing page on 21 September 2026.

LedgerLens is built for auditors rather than for the company being audited, which is its real advantage. Its pricing page describes an Auditor's Workbench for the "hundreds of point-in-time balance queries" an engagement needs, with a 7-day trial on Starter and proof of reserves modules held back for Enterprise. It covers 20+ blockchains. If your client list spans Ethereum tokens and several chains, start there.

Cryptio and Bitwave are subledgers the client runs to book crypto activity into its general ledger. Cryptio says it holds SOC 1 and SOC 2 Type 1 and Type 2 reports and lists Big 4 and national firms as partners. Neither publishes a price. You do not buy these as the auditor. You test them: agree their period-end units to the chain independently, and evaluate the controls report. That independent agreement is where a data tool comes in.

BitcoinDatabase is Bitcoin only, and that is the trade-off. On Bitcoin you get address history back to the 2009 genesis block over REST, a balance at a chosen height within the indexed chain, which grows with every new block, every input and output with its confirming block, and the balance history of an address over time, queried directly by SQL from the Growth plan. Before you rely on any vendor's historical balances, read our test of Bitcoin balance history APIs, where one well known feed refused P2WSH cold wallets and another ignored the height it was given. Full history needs Growth; the $49 Developer plan carries one year. Bulk CSV export for working papers is on Scale. For a client holding bitcoin with dozens or hundreds of addresses, the bulk address balance checker runs the whole list at one height in one pass.

BigQuery works if someone on the team writes SQL comfortably and understands that you pay per byte scanned, so an unfiltered query over the full input and output tables costs real money. Bitquery covers many chains through GraphQL; its cheapest tier is marked personal use only, so a firm starts at $79 a month.

Is blockchain data sufficient audit evidence?

It is strong evidence for existence and for movements, because it is external to the client and anyone can reproduce it. It is not evidence of control, and it is only as complete as the address list the client gave you. So pair it with ownership procedures, and design a completeness test: trace the funding of each known address one hop back and look for sibling addresses the client did not list. Change outputs returning to an unlisted address are the usual tell.

Remember the cash leg too. Every bitcoin purchase has a USD outflow on a bank statement, and matching the two is often the fastest completeness check you have. Clients rarely hand those statements over in a usable shape, so it helps to convert the PDF bank statements into a spreadsheet before you start matching dates and amounts against the on-chain receipts.

How much does crypto audit software cost?

Published prices run from $49 to $199 a month for the tools that list one, with the subledgers and enterprise tiers quoted on request. The cost that matters more is staff time. A senior pasting 200 addresses into an explorer, screenshotting each one and hoping the screenshots were taken at the cutoff, burns a day and produces evidence that cannot be rerun. A height-pinned query over the same list takes minutes and can be rerun by the reviewer.

Which crypto audit software should your firm use?

If your crypto clients span many chains and tokens, trial LedgerLens, because it is the one tool here designed around the auditor's workflow. If your clients hold mainly bitcoin, and especially if they hold it across many addresses, query the chain directly: agree units at the cutoff height, roll every address forward, and keep the query and the height in the file. That is what our audit trail API for bitcoin is built for, and it is the same data your client's treasury team can use, as covered in our guide to the best Bitcoin API for corporate treasury teams. Either way, record the cutoff rule, the height and the hash, because that is what makes the number defensible a year later.

When the engagement turns from attest to investigation, say a client disputes where coins went, the tool list changes. Our guide to crypto tracing software for forensic accountants covers following change and co-spent inputs, and why a fixed address list overstates outflows.

This is general information about audit data sources and a published accounting standard, not audit, accounting or legal advice.