Nft accounting: recognition, measurement, classification, royalties and impairment

Nft accounting for finance and accounting teams. How to account for NFTs — recognition, measurement and classification as intangible assets or inventory depending on use, royalty income, impairment and disclosures — under IFRS and US GAAP, with an auditable sub-ledger trail. This guide covers the treatment and how CryptaCount's crypto sub-ledger automates it.

General information on accounting treatment, not accounting or tax advice. Verify against the applicable standards (IFRS / US GAAP) and your auditor.

What makes NFTs hard to account for

NFT accounting is the discipline of bringing non-fungible tokens into ordinary accounting records: deciding when an NFT is recognised, at what value it is carried, which line it belongs to, and how the income and costs around it are booked. The difficulty is not that NFTs are exotic — economically they behave like unique items an entity buys, holds, creates or sells — but that each one is unique, often illiquid, and surrounded by activity (minting, royalties, gas costs, marketplace fees) that has to be untangled before a journal can be posted.

The defining feature is non-fungibility. Two units of a fungible token are interchangeable and share a market price; two NFTs are not, even within the same collection. That breaks the convenient assumption that a quoted price applies to a holding. Many NFTs trade rarely or never after acquisition, so there may be no recent, observable price at all. Valuation therefore leans on judgement — recent sales of comparable items, floor prices for a collection, or modelled estimates — and that judgement has to be documented rather than asserted.

The second difficulty is that the same token can be a completely different asset depending on why the entity holds it. An NFT bought as a long-term holding, an NFT held for sale by a marketplace or trader, an NFT created in-house and held for sale, and an NFT that conveys a licence or access right are not the same accounting object. Use drives classification, classification drives measurement, and so the business model has to be established before the books can be built.

The third difficulty is the surrounding activity. Minting an NFT incurs gas and may generate it as inventory; selling one incurs marketplace fees and may trigger an ongoing royalty entitlement on secondary sales; acquiring one bundles the purchase price with transaction costs. Each of these is an accounting event in its own right. Capturing them completely and tying each back to a transaction hash is exactly what a crypto sub-ledger is built to do, because doing it by hand across a collection does not scale.

How the activity maps to the books

As with any asset, three decisions drive the accounting: recognition (what event creates an entry and when), measurement (at what value the NFT sits on the balance sheet), and classification (which line it belongs to). The right answers depend on the entity's business model and the framework it reports under, so they are decided deliberately rather than defaulted from the wallet.

Recognition

An entity recognises an NFT when it obtains control of it — typically on acquisition or, for a creator, on minting where the token is held for the entity's own purposes. The initial carrying amount usually includes the purchase price plus directly attributable transaction costs such as gas and marketplace fees, depending on the framework and classification. For a minted NFT held for sale, the relevant costs are the costs of bringing it into existence. Recognition also applies to the income side: a sale derecognises the NFT and recognises any gain or loss, and royalty entitlements are recognised as income when earned.

Classification — intangible asset or inventory

The pivotal judgement is whether an NFT is an intangible asset or inventory, and it turns on use. An NFT held for investment, for the access or utility it confers, or otherwise not for sale in the ordinary course of business generally presents as an intangible asset. An NFT held for sale in the ordinary course of business — by a marketplace, a trader, or a creator producing items to sell — generally presents as inventory and is measured on an inventory basis. The same digital object can therefore sit in different lines on two entities' balance sheets, which is why classification has to be reasoned from the holding's purpose.

Measurement under IFRS and US GAAP

Measurement follows classification and is where the frameworks diverge. Under IFRS, NFTs held as intangible assets are generally carried at cost subject to impairment, with inventory treatment — typically the lower of cost and net realisable value — where the entity holds them for sale. Under US GAAP, classification similarly drives whether an NFT is carried under an intangible or inventory model, and the scope of fair-value measurement that applies to certain fungible crypto assets does not automatically extend to unique tokens. The practical consequence is that the same NFT activity can produce different carrying amounts, different timing of gains and losses, and different disclosures under the two frameworks. This is framework-level guidance; confirm the specific treatment against current standards and professional advice.

Royalties

Many NFTs carry a royalty that pays the original creator a share of each subsequent secondary sale. For a creator, royalties are an income stream recognised as it is earned, measured at the value received, and they continue long after the first sale. For a buyer, a royalty obligation embedded in a marketplace sale is a cost of transacting. Because royalties arrive as on-chain payments tied to third-party sales the entity did not initiate, they are easy to miss and have to be captured from the chain to be recognised completely.

Cost basis and gains and losses

When an NFT is sold, the gain or loss is the difference between the proceeds (net of marketplace fees and any royalty) and the NFT's cost basis — its initial carrying amount plus capitalised acquisition costs, less any impairment already recognised. Because each NFT is unique, basis is tracked per item rather than pooled, which actually simplifies one thing: there is no need to choose between FIFO and average-cost for a single non-fungible token, since the specific item disposed of carries its own basis.

Where an entity holds many NFTs, or trades the fungible tokens used to buy and sell them, a consistent cost-basis method still matters for the fungible side of each transaction — the token paid or received. As an illustrative example, an entity acquires an NFT for 5,000 plus 200 of gas and fees, carrying it at 5,200; it later sells for 9,000 net of a 300 marketplace fee and a 450 royalty paid to the creator, realising proceeds of 8,250 and a gain of 3,050 (all figures illustrative). Booking only the headline 9,000 would overstate the gain and miss the fee and royalty entirely.

Revaluation and impairment considerations

At each reporting date, NFTs carried at cost have to be tested for impairment: where there are indicators that an NFT's recoverable value has fallen below its carrying amount, a write-down is recognised. NFTs are particularly exposed to this because collection values can fall sharply and liquidity can evaporate, leaving an item that cannot be sold near its carrying amount. Inventory held for sale is similarly written down where net realisable value drops below cost.

The hard part is evidence. With no continuous quoted price, an impairment test relies on the best available indicators — recent sales of the same or comparable items, the collection's floor price, marketplace listing data, or a documented model. Whatever approach is used has to be applied consistently and disclosed, so a reader can judge how much estimation underlies the figure. Reversals of previous impairments, where a framework permits them, follow the same evidential discipline. The objective is that two people applying the policy to the same facts reach the same number.

Controls and audit trail

Auditable NFT accounting needs an unbroken chain from each reported figure back to the on-chain event behind it. The control questions are specific to non-fungible assets: is every NFT the entity holds captured, including ones airdropped or received without purchase — completeness. Does the ledger's inventory of NFTs reconcile to the tokens actually held in the entity's wallets at the cut-off. Is each item's carrying amount supported by its acquisition cost and any impairment evidence. Have all royalty receipts and marketplace fees been captured from the chain.

• completeness — every NFT in the entity's wallets recorded, including airdrops and items received without a purchase;
• reconciliation — the ledger's NFT inventory agreed to on-chain holdings at each measurement date;
• valuation evidence — the basis for each carrying amount and impairment retained, given the absence of a continuous quoted price;
• income completeness — royalty receipts and secondary-sale entitlements captured from the chain rather than relied on to be reported;
• classification support — the purpose of each holding documented so its intangible-or-inventory presentation can be defended.

Because each NFT is identifiable on a public ledger, the existence and ownership of every item can be confirmed independently against the chain — a strong position for the existence assertion. A difference between the ledger's inventory and the wallet's actual holdings points at an unrecorded transfer, an uncaptured airdrop or a sale that never hit the books, and resolving it at close keeps it out of the statements.

How CryptaCount handles NFTs

CryptaCount is a crypto sub-ledger that sits in front of the general ledger and turns NFT activity into accounting records. It ingests on-chain transactions across the entity's wallets, recognises each NFT acquisition, mint, transfer and sale, captures the gas, marketplace fees and royalty receipts that accompany them, and carries each item with its own basis so a disposal measures gain or loss against the right number. Because every NFT is tracked individually, the sub-ledger maintains a per-item inventory that reconciles directly to the wallet.

On that activity, CryptaCount applies a consistent measurement basis, supports impairment recognition with the supporting data retained, recognises royalty income as it is earned, and posts summarised period journal entries to the general ledger. Each GL balance decomposes back into the individual NFTs and events behind it, so an auditor can select a figure, trace it to a specific token, and confirm its ownership and movement on the public chain. The same engine presents the activity under IFRS or US GAAP and supports DAC8, CARF and MiCA reporting, so a marketplace, fund, creator-business or web3 treasury records each NFT once and reports from it under whichever basis it needs. See how the broader cost-basis engine handles the fungible tokens used to buy and sell those NFTs.