The ASC 606 five-step revenue recognition model is the US GAAP framework that governs when and how an entity records revenue from contracts with customers. The five steps run in order: identify the contract, identify the performance obligations within it, determine the transaction price, allocate that price to each performance obligation, and recognize revenue as each obligation is satisfied. FASB issued the standard jointly with the IASB (as IFRS 15) in May 2014, and it has been effective for all entities since 2019. The model replaced a patchwork of industry-specific guidance with one principles-based process, which means the arithmetic is rarely the hard part. Judgment is.
What ASC 606 Covers (and What It Doesn’t)
ASC 606 applies to every contract with a customer for goods or services that are an output of the entity’s ordinary activities. A customer is the party that has contracted to obtain those goods or services. Before you reach for the five steps, confirm the arrangement is not in one of the standard’s explicit carve-outs.
Out of scope:
- Leases, which are accounted for under ASC 842.
- Insurance contracts, governed by ASC 944.
- Financial instruments covered by their own codification topics (receivables, debt and equity securities, equity method investments, derivatives, transfers and servicing, among others).
- Non-warranty guarantees, which fall under ASC 460. Product and service warranties stay inside ASC 606.
- Nonmonetary exchanges between entities in the same line of business made to facilitate sales to end customers, such as two oil companies swapping inventory to meet regional demand.
Getting scope wrong means applying the wrong model from the start, so the boundaries matter. Guarantees in particular are not swept in with debt securities; they have their own home.
Step 1: Identify the Contract
A contract is an agreement that creates enforceable rights and obligations. Under ASC 606, one exists only when all five of these criteria are met at the same time:
- The parties have approved the contract and are committed to their obligations.
- Each party’s rights regarding the goods or services to be transferred can be identified.
- The payment terms are identifiable.
- The contract has commercial substance, meaning the risk, timing, or amount of future cash flows is expected to change.
- Collection of substantially all of the consideration is probable.
If any one of those is missing, you cannot start the five-step process. You reassess as circumstances change, and if the criteria are later met, that is the point at which the contract exists for accounting purposes.
Watch for related contracts. Two or more agreements entered into around the same time with the same customer may need to be combined and treated as a single contract when they were negotiated as a package, when the consideration in one depends on the other, or when the goods or services across them form a single performance obligation.
Step 2: Identify the Performance Obligations
Performance obligations are the distinct promises in the contract to deliver goods or services. A good or service is distinct when both of these are true:
- The customer can benefit from it on its own or together with other readily available resources.
- The promise to transfer it is separately identifiable from other promises in the contract.
When goods or services are highly interdependent, or when one significantly modifies another, they are not separately identifiable and must be bundled into a single performance obligation. A construction contract where design, engineering, and building are deeply integrated is the classic case. The bundling call directly drives how much revenue gets recognized at each milestone, so a shaky conclusion here distorts everything downstream.
Step 3: Determine the Transaction Price
The transaction price is the total consideration the entity expects to receive for delivering the promised goods or services. Fixed fees are easy. Variable elements are where the work lives.
Variable consideration includes rebates, performance bonuses, penalties, price concessions, and refund rights. Estimate it one of two ways. The expected value method is a probability-weighted calculation across possible outcomes, which fits situations with many scenarios. The most likely amount method picks the single most probable outcome and often suits binary situations such as hitting or missing a milestone.
Whichever method you use, the estimate is subject to a constraint. You can include a variable amount in the transaction price only to the extent it is probable that doing so will not cause a significant reversal of cumulative revenue once the uncertainty resolves. Factors that raise the reversal risk include amounts heavily influenced by forces outside the entity’s control, uncertainty that will not resolve for a long time, limited experience with similar contracts, a history of offering broad price concessions, and a wide range of possible outcomes.
One exception matters. Sales-based and usage-based royalties promised in exchange for a license of intellectual property are recognized only when the later of two events occurs: the underlying sale or usage happens, or the related performance obligation is satisfied. The general variable consideration constraint does not apply to these royalties; the royalty-specific timing rule overrides it. This stops entities from estimating and front-loading royalty revenue before any sales materialize.
Finally, when the timing of payments gives either party a significant benefit of financing, the transaction price must be adjusted for the time value of money. A practical expedient lets you skip the adjustment when the period between transfer and payment is expected to be one year or less.
Step 4: Allocate the Transaction Price
With the transaction price in hand and the performance obligations identified, allocate the total across those obligations based on their relative standalone selling prices. The standalone selling price is the amount the entity would charge if it sold that good or service separately.
If observable standalone prices exist, use them. If they do not, ASC 606 allows three estimation approaches:
- Adjusted market assessment, which looks at what the market would pay and adjusts for entity-specific costs and margins.
- Expected cost plus a margin, which forecasts the cost of satisfying the obligation and adds an appropriate margin on top.
- Residual approach, which backs into the standalone selling price by subtracting the observable standalone prices of the other obligations from the total transaction price. This one is permitted only when the selling price for the item is highly variable (no representative price is discernible from past transactions) or the entity has not yet established a price and has never sold the item separately.
Any overall discount in the contract is allocated proportionally across all performance obligations unless there is observable evidence the discount relates only to a specific one or subset. The residual approach is the method practitioners most often misapply, usually reaching for it as a shortcut when adjusted market or cost-plus would produce a more faithful allocation.
Step 5: Recognize Revenue When the Performance Obligation Is Satisfied
Revenue is recognized when control of the promised good or service transfers to the customer. Control means the customer can direct the use of the asset and obtain substantially all of its remaining benefits. Transfer happens either over time or at a point in time, and that distinction decides the entire pattern of recognition.
A performance obligation is satisfied over time if any one of these is true:
- The customer simultaneously receives and consumes the benefits as the entity performs. Routine service contracts typically fit here.
- The entity’s work creates or enhances an asset the customer controls as work progresses, such as building on the customer’s property.
- The entity’s work does not create an asset with an alternative use to the entity, and the entity has an enforceable right to payment for work completed to date.
If none of those apply, the obligation is satisfied at a point in time, usually on delivery, customer acceptance, or transfer of legal title.
Measuring Progress for Over-Time Obligations
When an obligation is satisfied over time, you need a method to measure progress. The method must faithfully depict performance and must be applied consistently to similar obligations.
Output methods measure progress by the value delivered to the customer, using units produced, milestones reached, surveys of work completed, or time elapsed. Input methods measure progress by the entity’s effort, using costs incurred, labor hours, or machine hours relative to total expected inputs. Neither is inherently better. Output methods tie directly to what the customer receives, but outputs are sometimes hard to observe. Input methods are easier to track but can misstate progress when inputs do not correlate proportionally with the transfer of control, a frequent issue when significant materials are consumed early in a contract. The choice is a judgment about which pattern best reflects actual performance, not a free selection.
When Contracts Change Mid-Stream
Change orders, scope expansions, and price renegotiations are routine, and how you account for them cuts back through the five-step model. A modification is treated as a separate, independent contract when both of the following are true:
- The modification adds goods or services that are distinct from those already promised.
- The price increase reflects the standalone selling prices of those additional goods or services, adjusted for the circumstances of the particular contract.
When both are met, the modification is essentially a new deal layered on the original, and the two are accounted for independently.
When a modification does not qualify as a separate contract, the treatment depends on what remains. If the remaining goods or services are distinct from what was already transferred, the entity accounts for the modification prospectively, as if the original contract was terminated and a new one created, with the transaction price reallocated at updated standalone selling prices. If the remaining goods or services are not distinct, the entity records a cumulative catch-up adjustment, recalculating total progress on the combined obligation and recognizing or reversing revenue in the period of the modification. The catch-up approach is common for partially completed single-obligation contracts, such as a construction project where both scope and price shift midway through.
Where Judgment Most Often Goes Wrong
Three application areas generate most of the practical difficulty under the model.
Principal Versus Agent
When a transaction involves a third party in delivering goods or services to the customer, you have to decide whether you are acting as the principal or the agent. The call reshapes the income statement: a principal recognizes revenue at the gross amount collected from the customer, while an agent recognizes only its fee or commission. Actual profitability is unchanged, but reported revenue can swing dramatically.
The test is control. An entity is the principal if it controls the good or service before it is transferred to the customer. Indicators of control include being primarily responsible for fulfilling the promise, bearing inventory risk (including before the customer orders or after a return), and having discretion in setting the price charged. No single indicator is determinative, and the weight of each depends on the arrangement. The question gets especially contentious in platform businesses, marketplace models, and drop-shipping, where the entity may never physically handle the product. Auditors scrutinize these conclusions closely for exactly that reason.
Licenses of Intellectual Property
ASC 606 has dedicated guidance for IP licenses because the economics differ fundamentally from delivering a physical product. The question is whether the license gives the customer a right to access the entity’s IP or a right to use it, and the answer depends on the nature of the IP.
Functional IP has significant standalone functionality. It can process transactions, perform tasks, or be played and consumed on its own. Software, completed media content, and patented drug formulas are typical. A license to functional IP generally grants a right to use the IP as it exists at the point the license is granted, so revenue is recognized at that point in time. Symbolic IP lacks standalone functionality, and its value derives from the entity’s ongoing or past activities. Brand names, team logos, and franchise rights sit here. Because the entity has an implicit obligation to support or maintain the IP, a license to symbolic IP grants a right to access, with revenue recognized over the license period.
There is a narrow exception for functional IP. If the entity’s activities are expected to substantively change the IP’s functionality during the license period, and the customer is contractually or practically required to use the updated version, the license shifts to a right-to-access model with revenue recognized over time. For any license tied to sales-based or usage-based royalties, the royalty exception from Step 3 governs timing instead.
Transaction Price Allocation
Beyond the residual approach problem already flagged, allocation errors often trace back to a weak link in Step 2. If the performance obligations were not identified correctly, allocating the price across them cannot produce a faithful result no matter how carefully the standalone selling prices are estimated. When Step 5 revenue looks off, the problem is often two steps upstream.
The five steps are sequential by design. The judgments at each one carry forward, and reworking a later step rarely fixes an error made earlier. The discipline the model asks for is to make each decision deliberately, document the reasoning, and reassess when the facts change.