How 2,100+ SEC Filers Actually Report Revenue in XBRL: Seven Years After ASC 606
Seven years after ASC 606 became effective, our analysis of 2,100+ SEC filers shows revenue reporting in XBRL remains deeply fragmented. Only 46% of filers use the ASC 606 element exclusively, 25% still use legacy Revenues, and 11% have no standard revenue element at all.
How 2,100+ SEC Filers Actually Report Revenue in XBRL: Seven Years After ASC 606
Published: February 7, 2026 Data Currency: SEC XBRL filings processed through February 2026. Dataset covers 2,100+ companies across NYSE, Nasdaq, and OTC markets. The full SEC XBRL filing universe is larger (~10,000+ active filers per the SEC company tickers exchange registry).
Key Findings
We analyzed XBRL financial facts across 2,100+ SEC filers from FY 2019 through FY 2025. Seven years after ASC 606 became effective:
- Only 46% of filers use the ASC 606 revenue element (
RevenueFromContractWithCustomerExcludingAssessedTax) exclusively - 25% still use the legacy
Revenueselement with no sign of switching - 14% report both elements simultaneously (dual reporting)
- 11% have no standard revenue element at all — predominantly banks and financial institutions
- The legacy
Revenueselement is not declining — it grew from 416 to 802 companies (FY 2019–2025)
This fragmentation is not unique to revenue. We found the same pattern across COGS (6+ elements), CapEx (3 fundamentally different concepts), and net income (4 overlapping elements).
The XBRL Revenue Standardization Problem
When FASB introduced ASC 606 (Revenue from Contracts with Customers) in 2014, effective for public companies in 2018, the intent was clear: standardize how companies recognize and report revenue. The US-GAAP XBRL taxonomy introduced RevenueFromContractWithCustomerExcludingAssessedTax as the primary revenue element, replacing the older Revenues and SalesRevenueNet elements.
Seven years later, the data tells a different story.
Across 2,104 companies in our dataset that filed 10-K or 10-Q reports for fiscal year 2025, we identified six distinct revenue reporting patterns:
| XBRL Element | Companies | Share of Dataset |
|---|---|---|
RevenueFromContractWithCustomerExcludingAssessedTax (ASC 606, exclusive) | 978 | 46% |
Revenues (legacy US-GAAP, exclusive) | 516 | 25% |
Both ASC 606 + legacy Revenues simultaneously | 286 | 14% |
RevenueFromContractWithCustomerIncludingAssessedTax | 181 | 9% |
RevenuesNetOfInterestExpense | 26 | 1% |
Revenue (IFRS filers via 20-F) | 12 | <1% |
| No standard revenue element | 225 | 11% |
The 225 companies without any standard revenue element are not data gaps — they are banks, credit unions, investment companies, insurance firms, and broker-dealers whose business models don't map to the concept of "revenue" as the taxonomy defines it.
Seven Years of ASC 606 Adoption Data
The adoption curve reveals something unexpected: the legacy Revenues element is not being replaced — both elements are growing in parallel.
| Fiscal Year | ASC 606 Element (RevenueFromContract...) | Legacy Revenues | Ratio (ASC 606 : Legacy) |
|---|---|---|---|
| 2019 | 443 | 416 | 1.06 |
| 2020 | 562 | 467 | 1.20 |
| 2021 | 633 | 472 | 1.34 |
| 2022 | 675 | 477 | 1.42 |
| 2023 | 721 | 516 | 1.40 |
| 2024 | 977 | 714 | 1.37 |
| 2025 | 1,264 | 802 | 1.58 |
The ASC 606 element has grown at approximately 19% CAGR since 2019, reaching 1,264 companies by FY 2025. But the legacy Revenues element has also grown at approximately 12% CAGR — from 416 to 802 companies — as our dataset coverage expanded to include more filers.
The critical insight is in the dual-reporting behavior: 286 companies report both elements simultaneously. They adopted the ASC 606 element but did not remove the legacy Revenues element from their filings. This accumulation pattern means the total number of revenue elements in the XBRL ecosystem is increasing, not converging.
At current adoption rates, both elements will likely coexist indefinitely. Any system extracting revenue from XBRL filings must account for this dual-reporting reality.
What Dual Reporting Means in Practice
When Oracle files its 10-K, the XBRL instance document contains both RevenueFromContractWithCustomerExcludingAssessedTax and Revenues — each reporting the identical $57.4 billion figure. This is not an error. SEC filing preparation platforms (Workiva, Donnelley Financial) allow companies to tag the same line item with multiple XBRL concepts, and many companies do so to maintain backward compatibility with data consumers that may rely on either element.
For data extraction systems, dual reporting creates an ambiguity problem: if you query for both elements and find matches for both, you must deduplicate to avoid double-counting revenue. If you query for only one element, you miss companies that only report the other. The 286 dual reporters in our dataset occupy an uncomfortable middle ground that complicates any simple extraction strategy.
The dual-reporting trend also suggests that XBRL element deprecation in practice is much slower than taxonomy evolution. The FASB can introduce new preferred elements, but companies have no obligation to stop using older elements — and their filing preparation workflows actively encourage maintaining both for broad compatibility.
How America's Largest Companies Report Revenue
Looking at specific companies makes the fragmentation concrete. Among the most widely followed names:
| Company | Primary Revenue XBRL Element | FY Revenue |
|---|---|---|
| Apple (AAPL) | RevenueFromContractWithCustomerExcludingAssessedTax | $416.2B |
| Microsoft (MSFT) | RevenueFromContractWithCustomerExcludingAssessedTax | $281.7B |
| Amazon (AMZN) | RevenueFromContractWithCustomerExcludingAssessedTax | $716.9B |
| Alphabet (GOOGL) | RevenueFromContractWithCustomerExcludingAssessedTax | $350.0B |
| ExxonMobil (XOM) | Revenues (legacy) | $413.7B |
| Uber (UBER) | Revenues (legacy) | $44.0B |
| Oracle (ORCL) | Both elements (dual reporter) | $57.4B |
| JPMorgan Chase (JPM) | Multiple elements | $158.1B |
A clear pattern emerges: large technology companies have uniformly adopted the ASC 606 element, while energy companies and platform companies continue using the legacy Revenues element. Oracle is a dual reporter — filing both RevenueFromContractWithCustomerExcludingAssessedTax and Revenues with the same dollar value.
JPMorgan Chase illustrates why "revenue" is a contested concept for financial institutions. JPM's XBRL filing includes Revenues ($158.1B), RevenuesNetOfInterestExpense ($158.1B), plus the component elements InterestIncomeExpenseNet and NoninterestIncome. The appropriate "revenue" figure depends entirely on context — total revenue, net interest margin, fee income, or some combination.
The 225 Revenue Orphans
Approximately 11% of filers in our dataset — 225 companies — file 10-K and 10-Q reports but contain no standard revenue XBRL element in their consolidated financial facts. These companies report income through industry-specific elements instead:
| XBRL Element Reported Instead of Revenue | Companies | Typical Filer Profile |
|---|---|---|
InterestIncomeExpenseNet | ~100 | Banks, credit unions, thrifts |
InterestAndDividendIncomeOperating | ~86 | Registered investment companies, closed-end funds |
NoninterestIncome | ~86 | Banks (fee income: card fees, advisory, trading) |
InterestIncomeExpenseAfterProvisionForLoanLoss | ~83 | Banks (net of credit loss provisions) |
NetInvestmentIncome | ~20 | Insurance companies (portfolio income) |
InvestmentIncomeNet | ~14 | Asset managers, holding companies |
BrokerageCommissionsRevenue | ~3 | Broker-dealers |
InsuranceServicesRevenue | ~2 | P&C insurance carriers |
This is not a data quality issue — it reflects a fundamental mismatch between the US-GAAP taxonomy's concept of "revenue" and how financial institutions generate income. A bank's primary income is the spread between interest earned on loans and interest paid on deposits (InterestIncomeExpenseNet). An insurance company's income is investment returns on its float (NetInvestmentIncome). These concepts don't map to Revenues or any ASC 606 element.
For anyone building systems that extract financial data from SEC filings, this means revenue extraction cannot rely on a single element or even a single family of elements. Banks require a fundamentally different approach — computing total revenue as the sum of net interest income and noninterest income.
The 145 companies reporting both InterestIncomeExpenseNet and NoninterestIncome are almost certainly banks where total revenue must be derived rather than read directly. Another 86 companies report InterestAndDividendIncomeOperating, which is common among registered investment companies and closed-end funds that derive income primarily from portfolio returns rather than operational revenue.
This finding has significant implications for financial data standardization efforts. The US-GAAP taxonomy's assumption that all companies have a concept called "revenue" breaks down for roughly one in ten SEC filers. Any comprehensive financial database must either exclude these companies (losing coverage of the entire banking sector) or implement industry-specific revenue computation.
Revenue Is Not Unique: Fragmentation Across the Income Statement
The fragmentation we found in revenue reporting extends to every major financial line item. Each metric has its own element distribution, its own industry-specific patterns, and its own edge cases.
Cost of Goods Sold: 6+ Elements by Industry
| XBRL Element | Companies | Common Filer Profile |
|---|---|---|
CostOfGoodsAndServicesSold | ~833 | Manufacturing, retail, consumer goods, industrials |
CostOfRevenue | ~398 | Technology, SaaS, digital services, media |
CostOfGoodsAndServiceExcludingDepreciationDepletionAndAmortization | ~126 | Platform and marketplace companies |
InformationTechnologyAndDataProcessing | ~85 | Banks, financial data companies |
DirectCostsOfLeasedAndRentedPropertyOrEquipment | ~35 | REITs, property management |
CostOfSales | ~16 | IFRS filers (20-F), some consumer staples |
CloudServicesAndLicenseSupportExpenses | 1 | Oracle (unique element) |
The COGS element a company uses is strongly correlated with its industry. We found several cases where standard XBRL extraction produces misleading results:
-
ExxonMobil reports cost of revenue as
CrudeOilAndProductPurchases(~$229B), a company extension element that does not appear in the standard US-GAAP taxonomy. Standard extraction that only checks the six common COGS elements returns null for XOM's cost of revenue — producing a misleading 100% gross margin for the world's largest public oil company. Other major oil companies (Chevron, ConocoPhillips, Marathon Petroleum) use the standardCostOfGoodsAndServicesSoldelement. -
Uber reports
CostOfGoodsAndServiceExcludingDepreciationDepletionAndAmortization— separating depreciation from cost of revenue in a way that standard extraction misses. Without checking this element, UBER also shows an incorrect 100% gross margin. -
Oracle uses
CloudServicesAndLicenseSupportExpenses— a company extension element used by exactly one company in our entire dataset — to report its cloud and software cost of revenue.
Capital Expenditures: Balance Sheet vs. Cash Flow
| XBRL Element | Companies |
|---|---|
PaymentsToAcquirePropertyPlantAndEquipment | ~1,482 |
PaymentsToAcquireProductiveAssets | ~348 |
PropertyPlantAndEquipmentAdditions | ~38 |
The capital expenditure data reveals a conceptual mismatch, not just an element naming issue. 38 regulated utilities report CapEx via PropertyPlantAndEquipmentAdditions — a balance sheet concept — rather than through cash flow statement elements like PaymentsToAcquirePropertyPlantAndEquipment.
This matters because the balance sheet concept and cash flow concepts can report dramatically different values. For example, American Electric Power (AEP) reports $7.5B in annual capital expenditures through $150M) rather than the consolidated total. This is a 50x discrepancy caused by a concept-level mismatch, not a data quality error.PropertyPlantAndEquipmentAdditions, but the standard cash flow concepts return only segment-level dimensional data (
Net Income: The Parent vs. Consolidated Overlap
| XBRL Element | Companies |
|---|---|
NetIncomeLoss | ~2,049 |
ProfitLoss | ~1,282 |
NetIncomeLossAvailableToCommonStockholdersBasic | ~904 |
ProfitLossAttributableToOwnersOfParent | ~18 |
Most companies report multiple net income elements simultaneously. The overlap is not redundancy — these elements represent different concepts:
NetIncomeLossis consolidated net income including non-controlling interestsNetIncomeLossAvailableToCommonStockholdersBasicis the parent-attributable figure, required for EPS calculations under ASC 260ProfitLossis the IFRS equivalent (used by 20-F filers)
For companies with significant non-controlling interests — Philip Morris International, Walmart, UnitedHealth Group — these elements report materially different dollar amounts. Which element you read determines the earnings figure you report.
Dividends Paid: 6 Elements, No Clear Primary
The fragmentation extends even to straightforward concepts like cash dividends. We found six distinct XBRL elements for dividend payments:
| XBRL Element | Companies |
|---|---|
PaymentsOfDividendsCommonStock | ~692 |
DividendsCommonStockCash | ~586 |
PaymentsOfDividends | ~462 |
DividendsCommonStock | ~347 |
PaymentsOfDividendsPreferredStockAndPreferenceStock | ~138 |
PaymentsOfOrdinaryDividends | ~75 |
Unlike revenue — where the ASC 606 transition explains much of the fragmentation — the dividend element distribution reflects a taxonomy that has always offered multiple equivalent concepts for the same economic event. PaymentsOfDividendsCommonStock and DividendsCommonStockCash both represent cash dividends paid to common shareholders, yet companies split roughly evenly between them.
The 75 companies using PaymentsOfOrdinaryDividends include major dividend payers like Johnson & Johnson. Missing these companies when screening for dividend stocks would introduce significant survivorship bias into any quantitative analysis.
The Scope of Fragmentation
Across revenue, COGS, CapEx, net income, and dividends, the pattern is consistent:
- No financial line item has a single universal XBRL element
- Industry drives element selection more than accounting standards
- Companies accumulate elements over time rather than migrating
- Extension taxonomy elements create long-tail fragmentation that only surfaces at scale
- Concept-level mismatches (balance sheet vs. cash flow for the same economic concept) are harder to detect than missing elements
This means any system claiming to extract "revenue" or "net income" from XBRL filings using a single element is systematically excluding 10-25% of SEC filers. The true coverage rate of a single-element approach varies by metric — revenue is 46% at best (ASC 606 only), COGS is 57% (standard element only), and CapEx is 79% (primary cash flow element only).
The Current Filing Window
This research reflects XBRL filings processed through February 2026. The current filing window includes:
| Form Type | Fiscal Period | Companies Filed |
|---|---|---|
| 10-K (Annual Report) | FY 2025 | 108 |
| 10-Q (Quarterly) | Q1 FY 2026 | 73 |
| 10-Q (Quarterly) | Q2 FY 2026 | 73 |
| 10-Q (Quarterly) | Q3 FY 2025 | 55 |
| 10-K/A (Amended Annual) | FY 2025 | 8 |
Calendar-year companies have until approximately March 1, 2026 to file their FY 2025 10-K annual reports (60 days after fiscal year end for large accelerated filers, 75 days for accelerated filers). The 108 annual reports already filed represent early filers; the majority of FY 2025 10-K filings will arrive in late February and March.
What We Learned Building This Dataset
We process SEC XBRL filings programmatically and have refined our extraction approach through multiple iterations — each driven by a real-world data quality discovery like the ones documented in this research.
Several findings were not obvious from the XBRL taxonomy documentation alone:
-
Dual reporting persists indefinitely. We expected the legacy
Revenueselement to decline as ASC 606 adoption increased. Instead, companies add the new element alongside the old one. This means the number of distinct revenue elements in the wild increases over time, not decreases. -
Company extension elements create long-tail fragmentation. ExxonMobil's
CrudeOilAndProductPurchasesand Oracle'sCloudServicesAndLicenseSupportExpensesare company-specific XBRL elements that no taxonomy documentation will prepare you for. These only surface when you process filings at scale and validate outputs against known financials. -
Concept-level mismatches are harder to detect than missing elements. The utility CapEx case — where
PropertyPlantAndEquipmentAdditions(balance sheet) andPaymentsToAcquirePropertyPlantAndEquipment(cash flow) report different values for the same economic concept — only becomes visible when you compare across data sources within a single filing. -
Label linkbases from different filing preparation platforms produce different XML structures for the same XBRL concepts. Filings prepared by Workiva and filings prepared by Donnelley Financial (DFIN) use different naming conventions in their label linkbase files, requiring parsers to handle both formats.
-
Monthly dividend declarations can appear as quarterly-duration facts. Short-duration facts (under 70 days) — such as monthly preferred dividend declarations or partial-period income allocations — need to be filtered during period classification to avoid contaminating quarterly financial data with sub-quarterly artifacts.
Methodology and Dataset
Data Source: SEC EDGAR XBRL filings (10-K, 10-Q, 10-K/A, 10-Q/A, 20-F), processed by MetricDuck.
Dataset: Approximately 2,100 companies with processed XBRL financial facts, covering fiscal years 2019 through 2025. This represents a cross-section of NYSE, Nasdaq, and OTC-listed companies. The full SEC XBRL filing universe is larger — approximately 10,000+ active filers per the SEC company tickers exchange registry.
Filtering criteria:
- Consolidated facts only (
is_consolidated = TRUE) — excludes segment-level and dimensional breakdown data - Duration-type facts for income statement and cash flow items (revenue, COGS, net income, dividends)
- Instant-type facts for balance sheet items (CapEx via
PropertyPlantAndEquipmentAdditions)
Revenue orphan definition: A company is classified as a "revenue orphan" if it filed a 10-K or 10-Q for FY 2025 but no standard revenue XBRL element (Revenues, RevenueFromContractWithCustomerExcludingAssessedTax, RevenueFromContractWithCustomerIncludingAssessedTax, RevenuesNetOfInterestExpense, SalesRevenueNet, SalesRevenueGoodsNet, SalesRevenueServicesNet, TotalRevenues, or Revenue) appears in its consolidated duration facts.
Limitations: Our dataset does not cover the full SEC filing universe. Company counts and percentages reflect our processed dataset and may not be representative of all ~10,000 SEC filers. We continue to expand coverage.
Explore the Data
The financial data analyzed in this research powers MetricDuck's company analysis platform. Explore earnings, ROIC, and filing intelligence for individual companies:
- Apple (AAPL) Earnings | Filing Intelligence
- Microsoft (MSFT) Earnings | Filing Intelligence
- Amazon (AMZN) Earnings | Filing Intelligence
- JPMorgan Chase (JPM) Earnings | Filing Intelligence
- ExxonMobil (XOM) Earnings | Filing Intelligence
Related Research:
- Earnings Quality: The Complete Framework — How to evaluate whether reported earnings convert to cash
- Bank Earnings Quality: Loan Loss Reserves — Deep dive into JPM, BAC, and WFC
- ROIC: The Complete Investor Guide — Return on invested capital analysis across 1,300+ companies
- 10-K Analysis Framework — Professional approach to annual filing analysis
Disclaimer:
This analysis is for educational and informational purposes only. It does not constitute investment advice, and you should not rely on it as such.
Important considerations:
- This is a research analysis of XBRL data patterns, not a recommendation to buy or sell any security
- Company revenue figures cited are extracted from SEC XBRL filings and may differ from figures reported in press releases or other sources due to rounding, restatements, or element selection
- Our dataset covers approximately 2,100 companies and does not represent the full SEC filing universe
- Data is sourced from SEC filings, which may contain errors or be subject to restatement
- Always verify financial data against original SEC filings on EDGAR
Conflict of Interest Disclosure: MetricDuck provides financial analysis tools that process SEC XBRL filings. This research demonstrates our data capabilities and domain expertise.
MetricDuck Research
SEC filing analysis and XBRL data extraction for fundamental investors