Revenue Recognition Principles: 5-Step Model, ASC 606, and IFRS 15

Revenue recognition principles under ASC 606 and IFRS 15 tell a business when to record income from a customer contract and how much to record: revenue is booked as the company transfers promised goods or services, in the amount it expects to receive in return. The mechanics live in a five-step model that both standards share. Skipping a step, or applying one loosely, is what turns an accounting question into an SEC enforcement matter or an unexpected tax bill.

The Five-Step Model

Both ASC 606 in the United States and IFRS 15 internationally apply the same sequence, in order:¹

Identify the contract with the customer.
Identify the performance obligations in the contract.
Determine the transaction price.
Allocate the transaction price to each performance obligation.
Recognize revenue when (or as) each obligation is satisfied.

The idea holding the steps together is that revenue should track the value actually delivered, not the timing of cash. A company that collects two years of subscription fees on day one does not get to report all of it in the first quarter. It earns the revenue as it performs.

Step 1: Is There a Contract You Can Recognize Against?

A contract does not have to be a formal signed document, but it does have to create enforceable rights and obligations, and it has to satisfy all five of the following before any revenue can be recognized:¹

Both parties have approved the agreement and committed to their obligations.
Each party’s rights to the goods or services are identifiable.
Payment terms are identified.
The arrangement has commercial substance, meaning it is expected to change the risk, timing, or amount of the company’s future cash flows.
Collection of the consideration is probable.

The collectibility test catches people off guard. Under U.S. GAAP, “probable” generally means a likelihood of roughly 75 to 80 percent. If a customer’s creditworthiness is shaky enough to fall below that, revenue cannot be recognized at all, even after delivery and even with a signed contract. Miss any one of the five criteria and the standard bars recognition until the gap is fixed. This is not a soft guideline; recognizing revenue on a contract that fails a criterion is the kind of thing that produces a restatement.

When the Contract Changes Mid-Stream

Scope grows, prices get renegotiated, units are added. A modification counts as a new, separate contract only when two things are both true: the added goods or services are distinct, and the price increase reflects their standalone selling prices. Otherwise the modification folds into the existing contract, and the company has to decide whether the remaining work is distinct from what was already delivered (a prospective adjustment) or continues a single ongoing obligation (a cumulative catch-up). Which one applies changes which period absorbs the revenue impact.

Step 2: How Many Promises Are in the Contract?

A performance obligation is a promise to deliver something. When a contract bundles several deliverables, each one is a separate obligation for reporting purposes only if it passes two tests:¹

The customer can benefit from it on its own or with other readily available resources.
The promise is not highly dependent on, or integrated with, other promises in the same contract.

A software license sold alongside implementation services is the standard illustration. If the customer can run the software without the implementation, those are two obligations. If the implementation is so bound up with the software that neither works independently, they collapse into one. Splitting or lumping obligations wrong shifts revenue between periods, and auditors know it.

Warranties Are Not Automatically Obligations

An assurance-type warranty, which just promises the product works as advertised, is not a performance obligation. It is accounted for as a cost accrual under existing warranty guidance. A service-type warranty gives the customer something extra, like an extended coverage plan, and it is a separate performance obligation. A portion of the transaction price gets allocated to it, and that revenue is recognized over the warranty period. If a customer can buy the warranty separately, it is almost certainly service-type.

Step 3: What Is the Transaction Price?

The transaction price is what the company expects to receive in exchange for the goods or services, excluding amounts collected for third parties like sales tax.¹ A fixed-price contract makes this trivial. Everything else takes work.

Variable Consideration and the Constraint

Discounts, rebates, performance bonuses, penalties, and rights of return all make the price variable. Companies estimate the variable piece using either an expected value (probability-weighted average of possible outcomes) or the most likely amount (a single best estimate), whichever better predicts the outcome.

Then the constraint kicks in. Variable consideration can be included in the transaction price only to the extent that a significant reversal of cumulative recognized revenue is not probable once the uncertainty resolves. In plain terms: if there is a real chance the money will have to come back, it cannot be counted yet. The constraint gets tighter when the amount depends on outside forces like market volatility, when the resolution window is long, when the company has limited experience with similar contracts, or when it has a history of broad price concessions.

When Timing of Payment Becomes Financing

A payment made two years before delivery is functionally an interest-free loan to the seller. A payment made two years after delivery is the seller financing the buyer. When the timing gap is material, the transaction price is adjusted for the time value of money. The practical shortcut: if the gap between delivery and payment is one year or less, the adjustment is skipped. Most ordinary business fits inside that year. The adjustment also does not apply where the customer pays in advance but controls delivery timing, or where a substantial portion of the price is variable and depends on future events outside either party’s control.

Step 4: Split the Price Across the Obligations

When a contract has more than one performance obligation, the transaction price gets allocated among them based on their relative standalone selling prices, meaning what the company would charge for each item sold on its own.¹

Observable standalone prices are the ideal input. When there aren’t any, two estimation methods are permitted: an adjusted market assessment of what customers would pay, or an expected cost plus a reasonable margin. A residual approach, where you back into the price by subtracting known standalone prices from the total, is available only when the selling price is highly variable or uncertain. The purpose of this step is to stop companies from front-loading revenue by loading price onto the obligations that get delivered first.

Step 5: Recognize Revenue When Control Transfers

Revenue is recognized when control of the promised good or service passes to the customer. Control means the customer can direct the use of the asset and obtain substantially all of its remaining benefits. That transfer can happen at a single point in time or over a period.¹

Point-in-time recognition is the norm for product sales. The trigger is when the customer takes physical possession, when legal title transfers, or when the buyer assumes the risks and rewards of ownership. For a manufacturer, that is often the moment the customer signs for delivery.

Over-time recognition applies when the customer simultaneously receives and consumes the benefits as the company performs, when the work creates or enhances an asset the customer controls, or when the work has no alternative use to the company and the company has an enforceable right to payment for what it has completed. Long-term construction and recurring service arrangements are the classic cases. Progress is measured using input methods (like costs incurred against total expected costs) or output methods (like milestones or units delivered).

Principal or Agent Changes the Number You Report

Whether a company is a principal or an agent changes revenue dramatically even when the underlying transaction volume is identical. A principal controls the good or service before transferring it and reports the gross amount. An agent arranges for someone else to provide it and reports only its fee or commission.

The determination hinges on control, and three indicators help sort it: whether the company is primarily responsible for fulfilling the promise, whether it bears inventory risk, and whether it has discretion in setting the price. No single indicator is decisive, and their weight shifts with the nature of the transaction. A marketplace that never takes title, does not set the final price, and is not on the hook for defects looks like an agent. A retailer that buys goods, warehouses them, prices them, and handles returns is a principal.

Rights of Return

When customers can return products, revenue is recognized only for the units the company does not expect to be returned. For the expected returns, the company records a refund liability for the cash it expects to pay back and a separate asset for its right to recover the returned goods.¹ The recovery asset starts at the former carrying amount of inventory, minus expected recovery costs and any expected decline in value. Both figures get updated each reporting period as the return estimate is refined, with the adjustments flowing through revenue and cost of sales.

Capitalizing the Cost of Winning the Contract

Sales commissions and similar incremental costs of obtaining a contract are not expensed immediately. If the company would not have incurred the cost without winning the deal, and it expects to recover the cost, the cost gets capitalized as an asset. A commission paid only when the deal closes is the textbook case.

The asset is then amortized over the period the company expects to benefit, which can run past the initial contract term if renewals are anticipated. Amortization follows a pattern consistent with how the underlying goods or services transfer to the customer. There is a practical expedient: if the amortization period would be one year or less, the cost can be expensed immediately. The election has to be applied consistently to contracts with similar characteristics.

How This Flows Into Federal Taxes

Revenue recognition is not confined to the financial statements. Under IRC Section 451(b), added by the Tax Cuts and Jobs Act, accrual-method taxpayers with an applicable financial statement must recognize income for tax purposes no later than the year that income appears as revenue on that statement.²

So if ASC 606 causes revenue to hit the books earlier than it would have under prior rules, the tax obligation moves up with it. The applicable financial statement follows a hierarchy: SEC filings first, then audited statements used for credit or shareholder reporting, then statements filed with other federal agencies. Some methods are carved out, including the installment method and long-term contract methods, but for most accrual-method businesses the tie between book and taxable income is now tight.

Adopting these rules counts as a change in accounting method, which requires filing IRS Form 3115.³ Eligible businesses can use automatic change procedures with no user fee. The form must be attached to a timely filed return for the year of change, with a copy sent to the IRS National Office.

Where ASC 606 and IFRS 15 Diverge

The two standards were built together and are largely converged, but a handful of differences matter if you report under both.⁴ The collectibility threshold is the biggest: U.S. GAAP reads “probable” as “likely to occur,” while IFRS reads it as “more likely than not,” a lower bar. The same contract can clear the test under IFRS and fail under U.S. GAAP. Other differences show up in intellectual property licenses (U.S. GAAP classifies them as functional or symbolic based on standalone functionality; IFRS looks at whether the customer controls the license at grant), in shipping and handling after control transfers (U.S. GAAP permits a policy election to treat these as fulfillment activities), and in impairment reversals on contract cost assets (required under IFRS, prohibited under U.S. GAAP).

What Has to Be Disclosed

Recognizing revenue correctly is only half of what the standard demands. Financial statement users have to be able to see the nature, amount, timing, and uncertainty of revenue and the related cash flows. The required categories:

Disaggregated revenue, broken out by categories that show how economic factors affect the business, such as product line, geography, customer type, contract duration, or sales channel.
Contract balances, including opening and closing balances of receivables, contract assets, and contract liabilities, with explanations of significant changes.
Performance obligation details, covering when obligations are typically satisfied, significant payment terms, the nature of what is promised, and any return or warranty obligations.
Remaining obligations, meaning the total transaction price allocated to obligations not yet satisfied and when the company expects to recognize that revenue. Contracts with an original expected duration of one year or less are exempt.

Nonpublic entities get some relief. They can opt out of quantitative disaggregation, though they still have to disclose revenue split by timing of transfer and give qualitative context on the economic factors driving revenue.

What Happens When Companies Get It Wrong

The SEC treats improper revenue recognition as an enforcement priority. Monsanto paid an $80 million penalty for misstating earnings through rebate programs that recognized revenue without properly accounting for associated costs. Three executives paid individual penalties ranging from $30,000 to $55,000, and two were barred from participating in the financial reporting or audits of public companies.⁵

Pareteum Corporation paid $500,000 to settle fraud charges after overstating revenue by roughly $42 million across six quarters. The company said it was following ASC 606 but was in fact recognizing revenue on non-binding purchase orders with no regard to whether any performance obligations had been satisfied. Claiming compliance in a 10-K while ignoring the five-step model in practice adds a layer of misrepresentation on top of the underlying accounting error, and the sanctions in these cases extended past fines to professional bars that ended careers.

1
Financial Accounting Standards Board. Revenue from Contracts with Customers (Topic 606)
2
Office of the Law Revision Counsel. 26 USC 451 – General Rule for Taxable Year of Inclusion
3
Internal Revenue Service. Instructions for Form 3115
4
Financial Accounting Standards Board. Comparison of Topic 606 and IFRS 15
5
U.S. Securities and Exchange Commission. SEC Charges Monsanto With Accounting Violations