Revenue accrual vs. revenue recognition: what’s the difference and why it matters

Revenue recognition is the accounting decision about how much revenue a firm has earned in a period and when it can report it, governed by ASC 606 and IFRS 15. Revenue accrual is the bookkeeping entry that records that earned amount on the balance sheet when invoicing has not caught up yet. Recognition sets the number; accrual books it.

In month-end conversations, “we accrued it” and “we recognized it” often describe the same situation. That overlap is exactly why the two terms get blurred, and why the blur gets expensive once a firm runs fixed-fee work, milestone billing, and monthly closes at the same time.

The recognition side is well covered. The five-step model behind professional services revenue recognition explains how a firm decides what it has earned, and the playbook on revenue recognition for service projects shows why invoice dates and revenue dates drift apart on real engagements. The question both leave open is narrower: once revenue is recognized but not yet billed, what physically happens in the general ledger, and what is that action called?

That gap matters for the firms most likely to feel it. In Birdview’s analysis of 40+ discovery calls with professional services firms (March 2025 to March 2026), about half described invoicing as manual, and QuickBooks was the most requested accounting integration. When invoices are late and assembled by hand, the space between “earned” and “billed” grows every month, and the accrual entry is the only thing holding the books honest in the meantime.

What is revenue recognition?

Revenue recognition is a policy decision, not an entry. It answers how much revenue a firm has earned in a period and when it is allowed to report it, regardless of when an invoice is sent or cash arrives.

In the US the rules sit in ASC 606, Revenue from Contracts with Customers; internationally, IFRS 15 says nearly the same thing. Both apply one five-step sequence: identify the contract, identify the performance obligations, set the transaction price, allocate it, and recognize revenue as each obligation is satisfied. For a services firm, step five is where nearly all the judgment lives.

What matters for this comparison is who does the work and what they produce. A controller or finance lead reads the contract, picks a measure of progress (hours, cost-to-cost, milestones), and arrives at a figure. On a $100,000 fixed-fee engagement judged 40% complete, that figure is $40,000 of earned revenue. Nothing has been posted yet. Recognition has produced a number and a date, and the books still have to catch up.

What is revenue accrual?

Revenue accrual is the journal entry that records earned revenue in the period it was earned, before an invoice exists. The entry credits revenue and debits a balance sheet asset, usually labelled accrued revenue or unbilled revenue.

The word “accrual” carries two meanings, and separating them clears up most of the confusion.

Accrual as a basis of accounting. A firm either keeps its books on the cash basis or the accrual basis. IRS Publication 538 puts the difference plainly: under the cash method, income is reported when it is received; under the accrual method, income is reported when it is earned, regardless of when payment arrives. GAAP financial statements, and therefore ASC 606, assume the accrual basis. A cash-basis ledger has no place to hold “earned but not yet paid,” so it cannot produce ASC 606-compliant revenue at all.

Accrual as a specific entry. Inside an accrual-basis ledger, an accrual entry is posted whenever recognized revenue for a period runs ahead of what has been invoiced. If recognition says $40,000 is earned and invoices total $25,000, the accrual books the missing $15,000.

ASC 606 has its own vocabulary for that balance. Deloitte’s ASC 606 roadmap notes the standard uses “contract asset” but lists unbilled receivables as a common alternative label. The distinction finance reviewers care about: a contract asset is a right to payment that still depends on future performance, while a receivable is unconditional and only waiting on time or an invoice run. A T&M month that is fully approved but not yet invoiced is usually an unbilled receivable; a fixed-fee balance waiting on the next milestone is usually a contract asset.

Either way, the accrual decides nothing. It records an amount that recognition already determined.

Revenue recognition vs. revenue accrual: side by side

Recognition and accrual are sequential steps, not competing methods. Recognition produces the earned amount; accrual records whatever part of that amount has not been invoiced yet.

The table explains a pattern that shows up in month-end reviews. A firm can have a careful recognition policy and still produce messy books if accrual entries are posted late, posted twice, or never reversed. The reverse also happens: tidy, well-reversed accruals built on a progress estimate nobody has checked since kickoff.

The balance itself is worth tracking as a number. Accrued revenue as a share of recognized revenue is one of the clearer project financial metrics for spotting billing lag. When that share keeps climbing over two or three closes, invoicing is falling behind delivery, and the cause usually sits in time approvals or milestone sign-offs rather than in accounting.

Ownership of the inputs is often split across systems. Approved hours and percent complete typically live in project or PSA software, while the journal entry lives in the ledger. Deciding which system owns which financial data is what keeps the recognized figure and the accrued figure from being calculated twice, in two places, with two answers.

Worked example: recognition and accrual in sequence

The clearest way to separate the two is to watch both happen on one contract, in order. Recognition runs first and produces a figure; accrual runs second and books the gap; a reversal runs third when the invoice finally goes out.

The scenario. A consulting firm signs a $100,000 fixed-fee engagement. At the end of March, the project is 40% complete on a cost-to-cost measure. The client has been invoiced $25,000 so far, on a milestone schedule agreed at signing.

Step 1: recognition decides the amount. Applying the firm’s policy gives $100,000 × 40% = $40,000 of revenue earned through March. This figure does not depend on the $25,000 invoice. If the progress measure is wrong, everything downstream is wrong too, which is why recognition is a controller’s call and not a data-entry task.

Step 2: accrual books the gap. Invoices already put $25,000 into revenue (debit accounts receivable, credit revenue). The remaining $40,000 − $25,000 = $15,000 has no invoice behind it, so an accrual entry records it:

The income statement now shows the $40,000 recognition decided on, and the balance sheet carries $15,000 of earned value the client has not been billed for yet.

Step 3: the reversal most teams forget. In April, the firm invoices the next $15,000 milestone. That $15,000 is already in revenue from the March accrual, so it must not land in revenue a second time. There are two clean ways to handle it:

• Auto-reversing accrual. The March entry reverses on April 1, and the April invoice posts to revenue as normal. Net effect: no double count.
• Relieve the asset. The April invoice debits accounts receivable and credits accrued revenue instead of revenue, clearing the $15,000 balance directly.

Both are acceptable. The failure is doing neither. If the April invoice credits revenue and the accrual stays on the books, the contract shows $55,000 of revenue against $40,000 actually earned, and the extra $15,000 quietly survives into next quarter’s numbers. Scale it up: on a firm reporting $2 million of quarterly revenue, three missed $15,000 reversals overstate revenue by $45,000, or 2.25%, and every dollar of it drops straight into reported margin.

The example uses one fixed-fee contract on purpose. How the same gap behaves differently on fixed-fee versus T&M projects, and how it is managed as a running WIP balance, depends on the billing model rather than on the definitions here.

What happens when billing runs ahead? Deferred revenue

When invoices exceed recognized revenue, the excess is deferred revenue, a liability, not an accrual. It is the same recognition-versus-booking logic running in the opposite direction.

Flip the scenario. The same $100,000 contract is 40% complete, so $40,000 is still the recognized figure. This time the client paid a front-loaded milestone, and invoices total $50,000. The $50,000 − $40,000 = $10,000 excess is not revenue yet. It sits on the balance sheet as deferred (or unearned) revenue, which ASC 606 calls a contract liability: work the firm still owes the client.

One rule surprises people who learned these terms separately. Under ASC 606, contract assets and contract liabilities from the same contract are presented net, so a single engagement shows either a net unbilled position or a net deferred position at period end, never both. Receivables stay separate from that netting.

Deferred revenue is also why a project can look cash-rich and margin-poor at once. A $10,000 deposit is cash in the bank but zero revenue until the work is delivered, which is one of the main reasons cash flow and profit diverge in project businesses. Teams that confuse “accrued” with “recognized” tend to make the mirror-image mistake here, reading “billed” as “earned.”

Which mistakes come from conflating the two?

Most errors happen when a team treats the accrual entry as if it were the recognition decision, or treats an invoice as if it were either. Five patterns show up repeatedly in services finance.

1. Treating the accrual as proof of compliance. An accrual entry only records a number. If the progress estimate behind it is stale, the books are precisely wrong. Cost: revenue that looks policy-driven but fails the first audit question about how percent complete was measured.
2. Running accrual logic on cash-basis books. Some smaller firms keep a cash-basis ledger and track “accruals” in a spreadsheet beside it. The spreadsheet never ties out to the ledger. Cost: two versions of revenue and two different accounts receivable aging reports, with finance spending each close reconciling them instead of reviewing them.
3. Skipping the reversal. The $15,000 example above shows the arithmetic. Cost: revenue counted twice, usually discovered at year-end, when unwinding it means restating several closes.
4. Using “accrued” and “billed” as synonyms in reports. A project report that labels unbilled value as “billed to date” tells a delivery lead the client has been invoiced when it has not. Cost: collections and client conversations based on invoices that do not exist.
5. Accruing from unapproved time. On T&M work, the accrual is only as good as the hours behind it. Hours that are logged late, never approved, or written off later inflate the accrual first and erode it afterwards. Cost: the unbilled time that leaks out of service firms shows up as a shrinking accrual nobody can explain.

The fix for all five is the same discipline: decide the recognized amount from approved delivery data, book only the gap against invoices, and reverse or relieve every accrual when the invoice lands. The billing method on each contract determines how often that gap opens, which is why fixed-fee and milestone work needs the closest watch.

The short version

Recognition decides how much revenue is earned and when; accrual records the earned part that has not been invoiced; deferred revenue records the billed part that has not been earned. One is a judgment under ASC 606 or IFRS 15. The other two are entries that keep the balance sheet consistent with that judgment until invoices catch up.

In practice, the quality of all three depends on the same upstream inputs: approved time, an honest measure of progress, and milestones that are signed off when the work is done rather than weeks later. Firms that get those inputs into finance before the close are the ones whose margin is visible while the project is still running, instead of only after it closes.

FAQ

What is the difference between accrued revenue and recognized revenue?

Accrued revenue is a balance sheet amount: revenue already earned and reported but not yet invoiced. Recognized revenue is the income statement total for the period, whether or not it was invoiced. Recognized revenue includes accrued revenue. On a contract with $40,000 recognized and $25,000 invoiced, recognized revenue is $40,000 and accrued revenue is $15,000.

Can you recognize revenue without accruing it?

Yes. When invoices for a period exactly match the revenue recognized for that period, there is no gap to book and no accrual entry is needed. This is common on T&M work invoiced monthly from approved hours. Accruals appear only when recognition runs ahead of billing, which happens most on fixed-fee, milestone, and percentage-of-completion contracts.

Is accrued revenue the same as unbilled revenue?

In everyday use, yes: both describe earned revenue that has not been invoiced. Under ASC 606 the precise label depends on the payment right. If payment still depends on future performance, the balance is a contract asset. If only time or an invoice run stands in the way, it is an unbilled receivable. Auditors expect that split to be visible.

How is deferred revenue different from accrued revenue?

Deferred revenue is the opposite position. It arises when a client is invoiced or pays before the work is delivered, so it is a liability representing work still owed. Accrued revenue is an asset representing work delivered but not billed. Under ASC 606, the two positions on a single contract are netted, so the contract shows one or the other.