Last Updated: 2026-10-26
The decisive question is whether a product can reconstruct your actual history and produce the output you need for the relevant country and year. A large integration count does not establish correct treatment of your bridge, margin account or reward. A software subscription also does not automatically include a professional preparing or filing your return.
This is a desk comparison of the providers' public pages, checked on 26 October 2026. No paid account, import accuracy test, filing or support-response study was conducted. The four providers below remain useful candidates to investigate; none has been awarded a performance or accuracy score here.
Dated public plan details
Amounts below are the dollar prices displayed by the linked pages when checked, before any checkout-specific tax, promotion or regional variation. Verify the applicable year, transaction-count definition, renewal terms and included export at purchase.
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Provider and official page | What the page displayed | Decision this does not settle |
|---|---|---|
| Koinly | Newbie $49 per tax year for 100 transactions; Hodler $99 for 1,000; Trader displayed $199 with higher-volume choices. Free portfolio tools and paid tax reports are distinguished. | Exact higher-volume quote, treatment of your protocol and the jurisdiction-specific report you need |
| CoinTracker | Base $59/year for 100 transactions; Prime $199/year for 1,000; Ultra $499/year for 10,000. Free tracking excludes downloadable tax forms. | Renewal terms, counting rules, import completeness and whether a particular professional service is included |
| TokenTax | Basic $49 per tax year for 100 transactions; Premium $199 for 5,000; Pro $1,999 for 20,000. The page separates crypto-tax software and full filing. | Which service engagement includes preparation or filing, the client's jurisdiction and the actual review scope |
| CoinLedger | One-time report purchase: Hobbyist $49 for 100 transactions, Investor $99 for 1,000, Pro $199+ for 3,000+. The page listed report years 2010–2025. | Availability of a later tax-year report, additional transaction blocks and treatment of complex operations |
These are provider statements, not independent accuracy findings. In particular, do not infer that a 2026-labelled comparison means every provider already offers a finished 2026 report. The report year shown at checkout matters.
Compare the same job
Start with a written requirement: jurisdiction, tax year, accounts, networks, transaction types, approximate raw activity and desired output. List difficult cases separately: margin settlements, rewards, bridges, liquidity positions, wrapped assets, missing purchase history and closed accounts. Do not upload private records to four services merely to obtain a price comparison; first inspect documentation and ask whether a sanitized sample or supported trial can answer your questions.
For each candidate, record five answers:
- Coverage: does the exact account type, chain and date range import, including fees and older acquisitions? “Exchange supported” may not answer whether that exchange's derivatives history is supported.
- Reconciliation: can you inspect unmatched transfers, missing basis and manual corrections? Can you export the underlying transaction ledger rather than only a final total?
- Output: is the required country/year report available, and can your professional use it? A generic gains CSV is different from a jurisdiction-specific form or filing service.
- Commercial scope: what counts toward the plan limit, what is paid again next year, and are human reconciliation, amended reports or filing separate services?
- Data handling: what data is shared, what access is requested, how can access be revoked, and what deletion/retention terms apply? An integration does not need withdrawal permission to read a transaction export.
A small import acceptance test
Use the synthetic recordkeeping example as a transparent test design, adapting it to an import format a provider documents. It begins with 100 fictional T, transfers 80 to a second account with a separate 1 T fee, and sells 60. Its closing balances are 19 T in A and 20 T in B. Cash paid is $202; cash received is $177. No tax result is specified.
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Test | Expected observation | What failure reveals |
|---|---|---|
| Import both sides of the transfer | One matched movement; 80 T change location | Duplicate acquisition or missing transfer matching |
| Include the 1 T transfer fee | Combined closing balance 39 T | Fee asset omitted or quantity handling wrong |
| Import the same file twice | No silent duplication; behavior is explained | Repeat-import workflow needs correction |
| Omit the purchase deliberately | Missing acquisition is surfaced | A plausible total may conceal incomplete history |
| Add a manual classification, then refresh | Change is preserved or its replacement is clearly reported | Review work may be lost during resync |
| Export the ledger | Source IDs, units and corrections remain traceable | Final figures cannot be checked back to inputs |
This is a proposed test, not a claim that the four products passed it. A passing simple example would not validate DeFi interpretation or a real tax return. Include your genuinely difficult operation types in any later authorized test, and have the relevant tax classification reviewed separately.
Price the reconciliation work as well as the report
For an invented comparison, suppose Tool A costs $99 and leaves 60 unresolved rows. Tool B costs $199 and leaves 10. If each row takes four minutes to investigate, the workload is four hours versus forty minutes. Valuing that time at an illustrative $30/hour gives combined costs of $219 and $219. This is a break-even calculation, not evidence about any named vendor. Rows may take very different amounts of time, and an automated label can be wrong even when it is not marked unresolved.
Use your own count and time estimate. Separate software price, extra years, transaction blocks, reconciliation service, professional review and filing. Do not compare a DIY report with a full-service engagement as though they deliver the same work.
When to ask for a written service scope
If you need a professional rather than software alone, ask who performs the work, which jurisdiction they cover, what records you must supply, whether they reconcile missing history, who reviews the result, and who actually files. Confirm how corrections and additional years are charged. A CPA or enrolled-agent reference on a website does not mean every plan includes that person's review of your account.
If a necessary account or year is unsupported, stop that comparison at the requirement gap. Do not award an attractive overall score that hides a missing mandatory report. If several products meet the requirements, choose using the verified cost, reconciliation workflow and data-handling terms that matter to you.
The linked provider pages are direct sources, not test logs. No personal use, best-in-class tagging or guaranteed compliance is claimed. This is not tax advice; current treatment depends on your facts and jurisdiction. AITR may earn commissions from commercial links elsewhere on the site; that does not establish a product's suitability.