High-bandwidth memory powers every AI accelerator from NVIDIA's H100 to AMD's MI300X—but the most revealing data about HBM supply geography doesn't appear in any chip maker's filing. It's in FormFactor's annual report, where a probe card geographic revenue table shows South Korea overtook the United States as the dominant buyer of HBM test equipment in 2025, adding $53 million while US revenues fell $32 million in a single year. This analysis cross-reads 135 operational SEC filings across four companies and three SIC codes—tracing HBM demand from process equipment through memory production, test infrastructure, and demand driver—to surface a supply chain concentration that no single 10-K can reveal.
Basel III endgame never became law, yet 419 substantive SEC filings prove it functioned as an operative market signal across three distinct industries simultaneously. Morgan Stanley's 10-K disclosed that its risk-weighted assets would rise 40% under the Expanded Approach—nearly three times State Street's 15%—while the G-SIB surcharge offset buried in that same filing contradicts the universal narrative that large trading banks faced unlimited capital pain. The real story emerges only from reading across the chain: bank capital constraints flow downstream through ICE's clearing infrastructure and into the Agency MBS market where mortgage REITs like Bimini Capital hold assets.
AI infrastructure capex has crossed from strategic investment into arms race. Amazon's FY 2025 10-K records $128.3 billion in capital expenditures — up 65% in a single year. Alphabet spent $91.4 billion, up 74%. Meta committed $115-135 billion for 2026 before a dollar has been spent. Reading all five filings together reveals three structural patterns investors are still underestimating: the spending is accelerating, not plateauing; Microsoft's reported margins already show the cost compression that peers only discuss in the future tense; and NVIDIA's own 10-K names the same data centers and power grids that hyperscalers are racing to build as the binding constraint on its revenue growth — completing a self-reinforcing cycle that no single filing captures alone.
On March 11, 2026, Stryker Corporation filed its first 8-K disclosing a cybersecurity incident that disrupted its global Microsoft environment — one of 916 material cybersecurity 8-K disclosures filed with the SEC in the past 12 months. But the initial 8-K is the least financially informative disclosure in the entire lifecycle. Cross-company analysis of four non-tech companies — a medical device maker, a donut chain, a mortgage lender, and a hospital system — reveals that the most financially specific SEC disclosures appear 12–24 months after the incident. Insurance recoveries of $21.5M to $35M, class action consolidations, and settlement disclosures mark the true cost — and they arrive long after the headlines fade.
Across 351 operational SEC filings, the lithium battery supply chain reveals a counter-intuitive pattern: falling lithium prices did not flow evenly through the value chain. Albemarle grew volumes 8% even as pricing collapsed $627 million. Rivian grew deliveries while flagging battery supply as a material risk. But Aspen Aerogels — the company whose thermal barriers physically protect lithium-ion batteries — lost 45% of its revenue despite holding multi-year OEM contracts. Mid-chain component suppliers absorbed the worst of the shock.
Data center power demand appears in 404 SEC filings across 30+ companies and 15 industries, but the most important pattern isn't the breadth — it's the convergence. Bitcoin miners are becoming data center landlords. Solar infrastructure companies are rewiring for server racks. A semiconductor firm cut 19% of its workforce to chase power conversion chips. From opposite ends of the value chain, companies that historically had nothing to do with computing are reorganizing around a single commodity: electricity access.
Nearshoring Mexico appears in 201 operational SEC filings from 30+ companies — but the real finding is that two auto-parts manufacturers in the same SIC code face opposite tariff exposure based solely on USMCA compliance status. Teleflex is pouring $36.4 million into Mexico PP&E while disclosing its products are 'not currently compliant' with USMCA. Standard Motor Products, in the same industry, explicitly states its Mexico operations are 'mostly exempt.' The nearshoring advantage isn't about being in Mexico — it's about qualifying under the trade agreement that makes Mexico economically distinct from China.
PFAS — per- and polyfluoroalkyl substances — appear in 736 operational SEC filings across 22 industries, but the most important pattern isn't breadth. It's that the same class of chemicals creates fundamentally different financial exposures at each stage of the value chain: billion-dollar settlements for producers, dual-role plaintiff-defendant paradoxes for water utilities, product lawsuits for companies that never manufactured PFAS, and forward-looking supply chain risk for semiconductor equipment makers who warn there may be no replacement 'at similar costs, or at all.'
Dick's Sporting Goods grew its core business EPS 3.8% to $14.58 in FY2025 while consolidated GAAP EPS fell 29% to $9.97. The FY2025 10-K reveals a two-speed retailer: the DICK'S segment earned 11.1% margins while Foot Locker lost $52 million. At $202, the market assigns negative $1.4 billion to the FL acquisition — making this a binary bet on a $177 million turnaround pinned to one back-to-school season.
Williams-Sonoma reported 'record' diluted EPS of $8.84 while net income declined 3.3% and operating margins contracted 50 basis points. The 10-K reveals this paradox: a $49M multi-year freight correction inflated FY2024 margins by 70bp, making FY2025's 'decline' an accounting phantom. Normalize for the freight ghost and a calendar artifact, and operating margins actually improved 40bp — the opposite of consensus. Meanwhile, $854M in buybacks manufactured the EPS record from a shrinking denominator, and Q4 tariff costs accelerated 4.25x, setting up a H1 FY2026 margin valley the market may be mispricing.
Everpure generated $615.7 million in free cash flow in FY2026 — and kept $2.2 million of it. The FY2026 10-K reveals that 99.6% of FCF recycles into an equity compensation system where $482M in stock grants, $343M in buybacks, and $271M in tax withholdings consume everything the business produces. Meanwhile, two catalysts sit unpriced: an ~$838M valuation allowance release signaled within 12 months and RPO acceleration from 14% to 40%. The filing shows why headline EPS could surge 3-5x without improving shareholder economics.
Boeing reported $2.48 in earnings per share for FY 2025 — its first profitable year since 2019 on 600 commercial deliveries. But the 10-K reveals adjusted EPS of -$9.35 after stripping a $9.6 billion divestiture gain that even Boeing's own 'core' non-GAAP metric fails to remove. The 737 recovery is generating real value at 4.8% incremental margin, but $8.4 billion in 777X losses consumed every dollar. At $216, investors are pricing in a turnaround the filing confirms is underway — but slower and more expensive than any headline metric reveals.
Lowe's reported a 74-basis-point operating margin decline in its FY2025 10-K, and the stock dropped 17% on cautious guidance. But the segment footnote tells a different story: core retail gross margins improved 53 basis points — three times the headline gain. The real threat isn't acquisition dilution. It's employee compensation absorbing 78.5% of incremental retail revenue, producing a negative incremental operating margin of -12.0%. With $10.8 billion in buyback authorization sitting unused and ROIC declining for three consecutive years, the investment case hinges on whether three variables — acquisition margins, labor costs, and buyback timing — can all move favorably at once.
Disney reported Q1 FY2026 revenue of $26.0 billion (+5% Y/Y) and adjusted EPS of $1.63, beating consensus by $0.06 — but the 10-Q reveals adjusted EPS actually declined 7.4% from the year-ago quarter's $1.76. Organic streaming subscriber growth was just 1%, with higher prices (+4%) and the Fubo acquisition (+4%) doing the real lifting. Operating cash flow collapsed 77% to $735 million, but the filing attributes this to a multi-year California wildfire tax catch-up — a non-recurring event the 8-K press release failed to explain. Experiences remains the profit anchor at $3.3 billion operating income (33.1% margin), absorbing 76% of Disney's $3.0 billion Q1 capex.
AST SpaceMobile reported a $342 million net loss — and buried in the same filing is $1.2 billion in contractual revenue commitments from AT&T, Verizon, and Vodafone. The company that consensus calls 'pre-revenue' generated $70.9 million in actual revenue, but $298 million in annual fixed costs tick regardless of launch timing.
Intercontinental Exchange delivered 20.7% EPS growth on 7% revenue growth in FY2025 — a 95% incremental margin that screams compounding monopoly. But the 10-K reveals operating cash flow grew just 1.1%, the '51% recurring revenue' claim masks a largest segment that's 62% transactional, and $160 billion in clearinghouse assets make every screener metric wrong. At 28× earnings, the filing quietly contradicts the growth narrative.
Dollar General reported a 107-basis-point gross margin recovery in its FY2025 10-K — but the filing's custom XBRL disclosure reveals that 75% came from inventory shrink reduction, a one-time operational fix with a calculable ceiling. Meanwhile, the WOTC tax credit expiration creates a $0.29-$0.46 EPS headwind that $69M in interest savings from $1.65B in debt paydowns can only partially offset. With ROIIC at -137% and the stock at 20.9x earnings, the filing shows a bridge year — not the steady-state compounder the multiple implies.
Snowflake collected $755 million more from customers than it recognized as revenue in FY2026 — nearly doubling the prior year's haul. But management's own 10-K discloses a billing shift that could compress this cash engine right as SBC leverage posts its fastest improvement in company history. The filing reveals two competing clocks: one pushing toward GAAP profitability, the other pulling away from cash profitability.
Amphenol's FY2025 10-K contains a number that doesn't belong in a connector company's filing: 31% operating margins in Communications Solutions. The segment nearly doubled revenue to $12.1 billion, generating software-like profitability on physical connectors. But the same filing reveals that APH's $10.5 billion CommScope CCS acquisition — the second piece of a $12.6 billion systematic dismantlement — will dilute those margins by 150-300 basis points in 2026. The investment question: is the franchise expanding faster than margins are diluting?
Burlington Stores generated $1.2 billion in operating cash flow in FY2025 — yet retained only $172 million as free cash flow. The gap: a $1.06 billion capital expenditure program funding the largest supply chain build-out in the company's history, now 70%+ complete. The 10-K reveals that as capex declines $229 million in FY2026, Burlington's FCF yield — currently the lowest among all peers at 0.9% — could triple to ~2.3% within 12 months. But the thesis depends on whether 53-week-adjusted 6.7% revenue growth can sustain the 2x operating leverage that makes each new dollar of revenue twice as profitable as the average.
The $650B AI capex headline is real — three of four hyperscalers filed specific guidance with the SEC. But the filings reveal what the headlines miss: Meta's actual infrastructure commitment is ~$250B (not $125B), three of four face negative free cash flow at guided levels, and none disclosed ROI timelines for the largest corporate infrastructure bet in history.
Home Depot reported $164.7 billion in revenue for fiscal 2025 — a 3.2% increase that masks organic contraction. The 10-K reveals a 1,105 basis point operating margin gap between HD's core retail business and its newly acquired distribution arm, a 560 basis point ROIC collapse, and a 15-year base-case deleveraging timeline. All headline revenue growth was acquisition-driven; the core retail segment actually shrank 0.75%.
Aflac returned $4.7 billion to shareholders in FY2025 — 185% of its operating cash flow. For a 43-year Dividend Aristocrat trading at 14.7x adjusted earnings, this looks like disciplined capital allocation at a bargain. But the 10-K reveals the cash engine funding this generosity is shrinking: Japan profit remittances fell 6.1%, reported margins depend on $529 million in actuarial gains, and a $553 million annual gap between subsidiary cash generation and parent-level distributions is widening each year.
Walmart reported its first operating margin decline in three years — minus 13 basis points to 4.18%. But two line items buried in the MD&A, totaling $1.6 billion, explain the entire drop. Strip them out and margins actually improved. Meanwhile, US eCommerce reached $99.6 billion (20.6% of net sales), membership fees hit $4.4 billion, and Walmart Connect advertising grew 33%. The catch: ROI already declined from 15.5% to 15.1%, and $18.7 billion in construction-in-progress hasn't started earning returns yet.
Biogen reported GAAP EPS of $8.79 — a 21% decline — while simultaneously beating Non-GAAP guidance at $15.28 per share. The $957 million gap between these two numbers is the widest percentage divergence among large-cap biotechs. The 10-K reveals that 'core' operating income actually improved 1%, that the U.S. franchise grew 12.9% while Biogen physically retreated from European markets, and that anti-CD20 royalty margins at 56% are structurally compressing profitability even as headline revenue stabilizes. At 14.1x adjusted operating earnings with zero shareholder returns, the stock is a bet on whether $5.3 billion in pipeline commitments can replace what the MS franchise is losing.
Lululemon generated $515 million of new revenue in FY 2025 and destroyed $291 million of operating income doing it — a -57.4% incremental operating margin that looks like structural decline. But the segment data tells a different story: China Mainland produced a 48.6% incremental operating margin on $393 million of growth, meaning each new dollar of Chinese revenue produced nearly 50 cents of operating income. Meanwhile, the Supreme Court's IEEPA invalidation created a $216M contingent refund absent from guidance. At 13.1× trailing earnings — the same multiple as cyclical truck manufacturer PCAR — the market prices permanent decline when the filing shows a geographic earnings quality transformation already in motion.
Amgen reported $7.7 billion in net income for FY2025 — an 88.5% surge. But operating cash flow fell 13.3%, free cash flow collapsed 22%, and accounts receivable spiked 41% against just 10% revenue growth. Meanwhile, buried in the product revenue footnote: growth products overtook legacy at-risk drugs for the first time. The patent cliff narrative is partially obsolete — but a $12.3 billion debt maturity wall and negative ROIIC complicate everything.
Airbnb grew revenue 10% to $12.2 billion in FY2025, yet EPS declined. The explanation is not the investment cycle Wall Street cites — it is a double squeeze the filing reveals but earnings coverage misses. Interest income of $705M (22.5% of pretax profit) is eroding as rates fall, while field operations and policy costs are surging 34%, creating a structural expense floor that cannot be cut like advertising. At 33x earnings, the stock needs 17% annual EPS growth — the filing delivered -1.8%.
Block reported revenue growth of 0.3% in FY 2025 — the kind of number that makes growth investors look away. But the 10-K reveals financial solutions operating at 91.9% gross margins grew 28%, driving Cash App's margin from 32.2% to 41.1% in a single year. The catch: loan losses doubled to 19.7% of financial solutions revenue, and Block deployed $3.5B into lending — more than its entire free cash flow. With $8.3B in capital demands against $2.43B in FCF, and a 40% workforce reduction contingent on AI tools management admits may 'fail entirely,' Block's transformation is the highest-stakes bet in fintech.
CoreWeave grew revenue 168% to $5.1 billion in fiscal 2025 and sits on $60.7 billion in committed backlog — more than 11 years at the current run rate. But the 10-K reveals debt service of $4.4 billion already exceeds the $3.1 billion in operating cash flow, interest coverage has fallen below 2.0x, and the company must raise $18-20 billion in 2026 just to refinance and keep building. At $71.61 per share, the stock prices in the growth — but not the cost of funding it.
Wabtec reported $6.83 in GAAP EPS for FY2025 — but $461M of acquisition charges obscure a business generating 56% incremental gross margins and 17.5% adjusted operating margins. Meanwhile, $735M of off-balance-sheet working capital tools make the cash conversion cycle look 14 days better than it actually is. We apply a three-layer 'Acquisition Opacity Framework' to reveal the real economics behind the largest locomotive maker's most aggressive M&A year.
Dell Technologies reported record $113.5 billion in revenue and 36% EPS growth for fiscal 2026. But buried in the segment footnote of the annual report, ISG gross margin collapsed 550 basis points in a single year — even as the segment's operating income grew 30%. The company's $26 billion services layer and disciplined SG&A efficiency are masking a fundamental shift in earnings quality that the earnings release doesn't show.
Netflix generated $9.46 billion in free cash flow in FY2025 — more than Amazon, on one-sixteenth the revenue. The 10-K reveals a structural inflection: content payments decelerated from +29% to +4%, driving a 47% incremental operating margin. But $45 billion in total obligations and guild agreements expiring mid-2026 complicate the cash machine narrative.
Rocket Companies reported a $68 million GAAP loss for fiscal 2025 — on $6.7 billion in revenue. But the 10-K reveals two fundamentally different companies hiding in one set of financials: a pre-acquisition originator that burned $359M in EBITDA over nine months, and a post-acquisition servicing platform that generated $592M in Q4 EBITDA alone. At $76.1 billion maximum dilution, the stock prices in a company that doesn't yet fully exist in GAAP.
MongoDB's free cash flow exploded 315% to $500 million in FY2026, pushing FCF margin to 20.3%. Wall Street fixated on the 17% revenue growth guide and sent the stock down 22%. But the 10-K reveals something neither bulls nor bears are discussing: nearly 100% of that $505 million in operating cash flow is consumed by a $499 million equity compensation recycling system — buybacks plus a brand-new net share settlement policy that annualizes to ~$300M/year. The real question isn't whether MongoDB can grow — it's whether the S&M leverage engine (47% → 38% of revenue in 3 years) can outrun this treadmill.
Target Corporation reported its 57th consecutive year of dividend increases and called its SG&A rate 'flat.' But the FY2025 10-K reveals a $593 million interchange fee settlement masked a 30 basis point deterioration in the underlying cost base, adjusted operating income declined 14.2% — nearly twice the GAAP decline — and management's own $5 billion capex guide pushes projected free cash flow $538 million below the annual dividend. For the first time, Target must borrow to sustain the streak that defines it as an income stock.
McDonald's cut share buybacks by 27% in FY2025 while spending a record $3.365 billion on expansion — a paradox until you see the 10-K's franchise economics. The filing reveals a 90.3% incremental franchise margin, meaning every new restaurant dollar is more profitable than the last. This analysis unpacks the capital allocation pivot, the emerging $647M technology platform, and the real estate fortress financing it all through Euro debt at half the U.S. rate.
Marvell Technology grew revenue 42% to $8.2 billion in FY2026 — the fastest among major semiconductor companies. Free cash flow grew 0.2%. A $735.5 million receivables factoring program buried in an MD&A footnote inflates reported operating cash flow by up to 42%, while non-GAAP adjustments exceeding GAAP operating income create a dual-reality earnings picture. Five findings from the 10-K that aren't in any earnings summary.
First Solar reported $5.2 billion in revenue and 40.6% gross margins in FY2025, but the 10-K reveals $1,600 million in Section 45X credits constitute 75.5% of gross profit — and exactly offset the US production cost premium. With OBBBA 'severely limiting' these credits, we decompose whether a three-layer competitive barrier (AD/CVD duties up to 3,404%, TOPCon patent suits, FEOC restrictions) can convert subsidy dependence into pricing power.
Coinbase processed $5.2 trillion in trading volume in 2025 — a 156% increase that generated just 1.7% more transaction revenue. The blended take rate collapsed 60%, to 7.8 basis points, while consumer revenue declined 3.1% and stablecoin growth decelerated from 31.7% to 2.7% sequentially. This filing analysis reveals the closing window between Coinbase's revenue transition and cost inflation.
At year-end 2025, Strategy Inc was worth $8.85 billion less than its bitcoin. The 17.3% NAV discount, $883 million in annual fixed obligations against negative operating cash flow, and a BTC Yield metric that fell from 74.3% to 22.8% reveal a capital structure approaching a sustainability threshold. This is what Strategy's 10-K tells you that the Bitcoin headlines don't.
Circle Internet Group grew revenue 64% to $2.75 billion in FY2025 — and its margin didn't move. For every dollar of USDC revenue, $0.61 flows to distribution partners, leaving Circle with a 39-cent dollar unchanged for two consecutive years. Meanwhile, $653M in non-cash charges turned a $530M free cash flow machine into a GAAP loser. The 10-K reveals one escape: Circle's own platform share of USDC quintupled from 2.2% to 11.1%, and every percentage point migrated saves ~$18M per year in distribution costs.
CrowdStrike's Rule of 40 registers an impressive 49.2% — until you adjust for $1.1 billion in stock-based compensation that the metric treats as free. The adjusted score collapses to 26.4%, exposing a 38.5-percentage-point gap between cash returns (31.2%) and accrual returns (-7.2%). Meanwhile, DOJ and SEC probes have expanded beyond the July 19 outage into revenue recognition and ARR reporting, and three accounting tailwinds totaling $130-140M will inflate FY2027 margins. This is the real CrowdStrike hiding behind the headline numbers.
Autodesk grew revenue by $1,075 million in FY2026 — and spent $4 million less on sales and marketing than the year before. The FY2026 10-K reveals that the channel transition from 42% to 63% direct sales is recapturing $284 million in distributor margin while .DWG file format lock-in keeps customers in place despite the most aggressive go-to-market restructuring in enterprise software history. But half of revenue now comes from construction software, making this 'design company' increasingly a cyclical bet that the 9x revenue multiple may not adequately price.
UnitedHealth Group grew Medicare Advantage membership by 600,000 in FY2025 and revenue surpassed $447 billion. Operating income fell 41%. The FY2025 10-K reveals a medical cost ratio of 89.1% — the worst trajectory among five managed care peers we analyzed from raw filing data. UNH's +5.9 percentage point MCR deterioration is nearly double the peer average, exposing the vertical integration model as a cost amplifier, not a dampener.
BlackRock reported record $14 trillion in AUM, record $698 billion in net inflows, and tripled its private markets revenue to $653 million quarterly after the HPS acquisition. But the same 10-K reveals GAAP EPS fell 16% to $35.31, free cash flow dropped 24% to $3.55 billion, and dividends now consume 91% of remaining cash. A 5-component decomposition of the $12.78-per-share GAAP-adjusted gap reveals that only $3 is genuinely temporary — the other $10 represents real, recurring costs hidden behind management's preferred metric.
DoorDash reported its first full-year GAAP operating profit of $723 million on $13.7 billion in revenue — a milestone investors waited years to see. But the 10-K reveals three hidden subsidies: stock-based compensation 45% larger than operating income, $348 million in capitalized software that never hits the income statement, and a near-zero 0.75% tax rate from one-time acquisition benefits. True free cash flow was flat year-over-year at $1.83 billion, and $9.15 billion in real obligations don't appear on any debt line. The Profitability Gap Cascade — from management's $2.8B Adjusted EBITDA to $776M in commitment-adjusted free cash flow — is a 3.6:1 ratio, the widest among peers.
Blackstone reports $0 in debt and a 203% dividend payout ratio — both technically accurate, both completely misleading. The FY 2025 10-K reveals $12.3 billion in operating borrowings at the partnership level, a Private Equity segment within $10 million of overtaking Real Estate for the first time, and a compensation structure that absorbs 70 cents of every marginal fee dollar. At ~$100 (as of March 2026), the stock implies ~20x Distributable Earnings — a reasonable multiple, but only if the BCRED redemption crisis doesn't compress the fee-earning asset base.
Berkshire Hathaway's net income fell 24.8% to $67.0 billion in FY2025 — but operating cash flow surged 50.3% to $46.0 billion. The divergence reveals the most important insight in the filing: $39.1 billion of reported earnings are mark-to-market noise that management calls 'meaningless.' The real operating P/E is approximately 38x, not the 15.7x on every terminal. Inside the $176B float machine at -5.4% cost, Buffett's $369B farewell, and the stress points investors should watch.
Venture Global generated $6.6 billion in operating cash flow — nearly triple its $2.3 billion in GAAP net income — yet trades at 7.3x EV/EBITDA, a 32% discount to infrastructure peers. The 10-K reveals why: Calcasieu's post-COD quarterly margin collapsed 36.2 percentage points, CP2's cost estimate escalated $4 billion in one quarter with no explanation, and preferred shareholders capture 58% of the total dividend payout. The margin lifecycle is quantifiable — and the market appears to be pricing it correctly.
Chubb Limited reported a record 81.2% combined ratio in FY 2025 — the best in its history. But inside the 10-K, North America Commercial now earns more from investing premiums ($3,840M in NII) than from underwriting them ($3,783M). At 12x trailing P/E with 24.2% ROTCE, the dual-engine compounder looks mispriced — until you aggregate $2.5 billion in reserve sensitivity buried across three actuarial disclosures. The world's largest P&C insurer has quietly become an investment company, and the market is still pricing it as a cyclical underwriter.
Brookfield Asset Management reported a record 2025 — $112 billion fundraised, fee-related earnings up 28%, and a 15% dividend increase. But the 10-K filing reveals that BAM's FY2025 was produced by a fundamentally different entity than existed twelve months earlier, earning just 56 basis points on $603 billion in fee-bearing capital while peers like Blackstone earn 90bps. The February 2025 Arrangement restructured BAM so thoroughly that every data vendor generated fictional year-over-year metrics. Once corrected, the real story emerges: $110.9 billion in undeployed capital earning zero fees today, a 134% FCF payout ratio funded by BAM's first-ever $2.5 billion debt issuance, and the fastest-growing segment — Renewable Power & Transition, up 32.9% with a 66.1% margin — now led by the CEO who built it.
Charles Schwab delivered its most profitable year ever in FY2025 — $8.4 billion in net income, with 87 cents of every incremental revenue dollar flowing to profit. But the 10-K reveals a paradox: management returned $12.2 billion to shareholders while the held-to-maturity portfolio carried $7.9 billion in unrealized losses. Our analysis decomposes the one-time double tailwind behind record margins, quantifies a 2-3 quarter exhaustion horizon for the counter-cyclical NII advantage, and maps the dual-path regulatory scenario that determines whether the capital return was brilliant or reckless.
Marriott International reported 14.2% EPS growth in FY2025, but the 10-K reveals a structural divergence beneath the headline. Co-branded credit card fees grew $105 million — exceeding the $94 million from actual hotel rooms growth — and the credit card increment accelerated 78% year-over-year. Meanwhile, the debt-funded buyback machine that delivered one-third of EPS growth is getting more expensive: average buyback price rose from $273 to $318 per share. This analysis decomposes Marriott's three growth engines and tracks where the efficiency is breaking down.
AIG returned $6.8 billion to shareholders in 2025 — a 14.1% yield — while posting its first $2.3 billion underwriting profit since 2008 and earning a triple credit upgrade to AA-/A1. But the 10-K reveals that the accident year combined ratio is flat at 88.3-88.9, meaning all headline improvement comes from non-recurring reserve releases and catastrophe luck. Meanwhile, the Corebridge monetization funding those record buybacks has depleted from $3.8 billion to roughly $2.1 billion, giving the current pace 12-18 months of runway. Both pillars converge in mid-to-late 2026.
EchoStar reported a $14.5 billion net loss in FY 2025 while its stock rallied 308% to a $31.3 billion market cap. The 10-K reveals that the FCC compelled the $42.65B spectrum sales under threat of license revocation, that the company already missed interest payments in Q2 2025, and that $30 billion in creditor claims consume most of the deal proceeds before equity holders see an estimated $12.6-15.6 billion residual.
Ameriprise Financial reported a 'record' 53.2% adjusted operating ROE and 10.1% EPS growth in FY2025. But strip away three layers of earnings distortion — insurance fair value gains, buyback-driven share shrinkage, and management's adjusted metrics — and the company's core pretax income actually declined 3.8%. At $490 per share, the 10-K reveals an earnings engine running on borrowed altitude.
Goldman Sachs reported record FY2025 earnings of $51.32 per share — up 26.6% — on what Wall Street calls an M&A super-cycle. But the 10-K filing reveals that 73% of the flagship Global Banking & Markets segment's $6.4 billion revenue increase came from net interest income surging 182%, not deal-making fees. Core franchise fee revenue grew just 5.7%. At $879 and 2.21x tangible book, the market prices a deal-making franchise — the filing describes a rate-sensitive balance sheet with a wealth management cost problem.
Robinhood's pre-tax income nearly doubled in FY 2025, growing 98% to $2.1 billion on a 68.3% incremental operating margin. But a $572 million tax swing compressed headline net income growth to 33%, masking the operating breakout. Our novel revenue fragility scoring reveals only 24% of HOOD's $4.5 billion revenue is structurally durable — with $2 billion in PFOF-dependent trading and event contract revenue facing specific 2026 regulatory catalysts.
Capital One reported the fastest revenue growth (+37%) among large US banks and the highest P/E ratio (60x) — simultaneously. The 10-K reveals a $20.7 billion day-1 CECL provision, 92.3% concentrated in a single segment, that makes the $4.03 EPS the most misleading number in banking. Free cash flow tells a different story: $48.29 per share, a 20% yield. This analysis decomposes the four components of COF's earnings illusion and models when 60x becomes 15x.
Vertiv Holdings generated $10.2 billion in FY2025 revenue with 21.2% ROIC and doubled its backlog to $15 billion. But the 10-K contains no remaining performance obligation disclosure — the word 'backlog' appears only in narrative sections. Meanwhile, Q4 organic growth was Americas +46.2% against EMEA -14.1% and APAC -9.3%, and a $1.14 billion acquisition allocating 91% to intangibles accelerates amortization by $74 million per year. At ~73x trailing P/E — double the ~35x for Schneider Electric and Eaton — the valuation prices a global compounder. The filing shows a US pure-play with an unverifiable backlog.
International Flavors & Fragrances reported an operating loss of $382 million in FY 2025 — yet management called it 'solid performance.' The disconnect: 72% of IFF's $2.1 billion Credit Adjusted EBITDA comes from add-backs. At GAAP EBITDA of $580 million, leverage isn't 2.6x — it's 9.3x. With $3.84 billion in cumulative goodwill impairments across three reporting units, six divestitures in 18 months, and a dividend payout ratio of 161% at the already-cut level, IFF's 10-K documents a $26 billion acquisition being systematically dismantled to save the company it was supposed to transform.
Warner Bros. Discovery reported $728 million in net income for FY 2025 — its first profit since the WarnerMedia Merger. But the 10-K reveals a $2.9 billion debt extinguishment gain is the only item keeping that number positive. Cash taxes and interest consumed 97.7% of operating cash flow. A $15 billion bridge loan at 7.16% replaced the cheap debt that generated the 'gain.' And $16.5 billion in debt matures in 2027 behind a triple credit downgrade. Here's what the filing reveals about why the PSKY merger at $31/share may be less a strategic combination than a financial survival mechanism.
KKR reported a 27% decline in net income, a 93% collapse in operating cash flow, and a 49% drop in operating income in its FY 2025 10-K. And its core fee business grew 14%. Both statements are true — because 56% of KKR's GAAP pre-tax income belongs to third-party fund investors, the OCF 'collapse' was a $9.2 billion co-investment deployment, and insurance spread compression threatens the flywheel economics. This filing reveals two fundamentally different companies inside one set of financial statements.
Interactive Brokers grew revenue 20% to $6.2 billion with a 77% pretax margin — but this isn't really a trading platform. The 10-K reveals 57% of revenue is net interest income, and rate cuts paradoxically helped: NII grew 13% despite 27bps of margin compression because $29.6B in balance growth overwhelmed the rate impact. A structural 0.50% spread floor creates a ~$650M NII minimum. Meanwhile, 76% of net income flows to the Peterffy family through the LLC structure — public shareholders get 24% of a financial machine with zero debt and $203B in liquid assets.
Keurig Dr Pepper reported $3.6 billion in operating income for FY2025 — a 38% year-over-year surge. But $865 million of that growth came from lapping prior-year impairment charges, all three operating segments show gross margin compression, and operating cash flow actually declined 10.3%. The filing reveals a business where organic operating income grew approximately 3.4%, appliance volume collapsed 19.9%, and the coming JDE Peet's acquisition will push pro-forma FCF payout to 93.6%.
Monolithic Power Systems sits on $1.3 billion in cash with zero debt and a $493 million buyback authorization — yet stock repurchases collapsed 99% from $635 million to $6.6 million in FY2025. The answer isn't in the headline numbers. It's buried in the income tax footnote: a foreign repatriation pipeline that dropped from $642 million to $275 million, barely enough to cover the dividend. Meanwhile, the 10-K discloses a material weakness in tax computation that required FY2024 restatement, Enterprise Data re-accelerated violently in Q4, and gross margins held within 10 basis points despite the largest revenue mix shift in company history.
Coterra Energy reported 40% revenue growth, 63% EBITDA margins, and a 20% free cash flow yield in FY2025 — numbers that suggest a company firing on all cylinders. But the 10-K reveals $354 million in derivative timing gains inflating the top line, per-BOE operating costs surging 35%, and $3.5 billion in hidden obligations that Devon Energy inherits in a merger neither party can exit. A 4-factor revenue decomposition shows three different growth stories with radically different durability, and the cheapest upstream FCF yield in the peer set may be the market's way of pricing in what the headlines don't show.
International Paper spent $7.2 billion to acquire DS Smith, took a $2.47 billion goodwill impairment within eleven months, then announced it would spin off the acquired business. But the impairment is only the first layer. Our 10-K analysis reveals $3.85 billion in total first-year value destruction — including a $518 million foregone tax shield and $867 million in integration cash drain — plus $1.23 billion in annual recurring charges that will weigh on earnings for 14+ years.
AvalonBay Communities grew FFO per share 3.8% to $11.41 and launched a $488 million buyback program — its first ever — yet the stock fell 18%. The 10-K filing reveals AVB is running three simultaneous capital allocation experiments through a single balance sheet that swung from net cash to net debt in one year. With $3.3B under construction, $1.5B+ in near-term refinancings, and ~$0.24/share of hidden 2026 headwinds, the math either compounds or collapses.
The iShares Silver Trust (SLV) reported $21.3 billion in net income for FY2025 — on paper, more profitable than 98% of the S&P 500. But 83.3% of that income is unrealized silver gains that never produced a dollar of cash. Meanwhile, each share's silver backing has eroded 9.3% since inception because the trust sells silver to pay BlackRock's 0.50% annual fee — a fee BlackRock can raise without shareholder approval. The filing reveals a structurally simple but economically powerful fee extraction mechanism operating behind one of America's most popular commodity investments.
Altria Group paid $4.16 per share in dividends on $4.12 per share in earnings — a 101% GAAP payout ratio that screams 'unsustainable.' But the 10-K tells a different story: free cash flow grew 5.4% to $9.1 billion, covering the dividend at 1.29x. Meanwhile, smokeable operating income grew for three consecutive years despite falling revenue, driven by a structural mechanism where settlement costs decline faster than volume. The paradox deepens with $2.1 billion in NJOY write-downs, on! losing the pouch race to ZYN, and Marlboro shedding 1.5 share points in a single quarter.
Moderna reported a -158% operating margin on $1.9 billion in FY2025 revenue. But the 10-K reveals something the earnings release doesn't: Moderna's mRNA manufacturing platform produces vaccines at a 78.2% core gross margin — 23 percentage points higher than reported. The gap? $472 million in transitional costs that have declined 87% since 2023. With $8.1 billion in total financial assets providing 4.3 years of runway, and two date-certain catalysts (Europe lockout expiry Dec 2026, FDA flu vaccine decision Aug 5, 2026) approaching, the central question shifts from survival to revenue inflection timing.
LPL Financial's FY 2025 10-K reveals a paradox: record $2.4 trillion in advisory assets and 37% revenue growth alongside an 18.5% decline in GAAP earnings, negative operating cash flow of -$411 million, and a leverage reality (4.74x) that management's preferred metrics (1.95x) obscure by 2.8 turns. We decompose the $740 million acquisition toll, expose a 30:1 asymmetric rate risk in the hidden profit engine, and identify the specific integration milestones that determine whether Commonwealth was worth $2.7 billion.
Mondelez International's FY 2025 10-K reveals a $729M gap between input cost inflation and pricing recovery — but that's the number everyone can find. What the filing buries across three separate disclosures is how the company funded $5.5B in shareholder returns on just $2.3B in free cash flow: a 37-fold commercial paper surge, $3.6B in bank-intermediated 'supplier' payables, and a Board that quietly doubled the company's debt authorization. Neither the 44.7% GAAP EPS decline nor management's 12.8% adjusted figure tells the truth.
Edison International reported $11.55 in earnings per share for FY2025 — a 249% surge that pushed its PE ratio to 5.2x. But $2,591 million of that earnings explosion was non-core wildfire settlement recoveries, and core EPS was just $6.55. Meanwhile, the Eaton Fire recovery framework absorbed 98.7% of initial claims at just $13 million pre-tax net cost to shareholders. The catch: customer bad debts tripled to $397 million — 4.1% of authorized revenue — threatening the ratepayer-funded mechanism that makes the framework work. At $60 per share, Edison isn't just a wildfire bet — it's a bet on whether the recovery framework survives its own funding source deteriorating.
iShares Gold Trust posted $24.3 billion in FY2025 'net income' with zero revenue, zero cash flow, and zero employees. Gold rose 65%. The trust's NAV doubled. But the 10-K filing reveals four different answers to the question every gold investor should ask: what did I actually earn? We decompose IAU's record year into four components — gold price appreciation, per-share value capture, flow amplification from 229.5 million new shares, and BlackRock's $124 million in guaranteed fee extraction.
Natera reported its first quarterly profit in company history — $47.2 million in Q4 2025. But the 10-K reveals $60.6 million came from a one-time tax benefit, $135 million in annual revenue reflects prior-period catch-ups rather than current tests, and $1.17 billion in identified future dilution sits in the footnotes. The profitability transition is real, but it has three layers — and the quality of each determines whether NTRA is a compounder or a value trap.
NVIDIA generated $216 billion in revenue with 75% gross margins — the most dominant operating performance in tech history. But the FY2026 10-K reveals $165 billion in total obligations growing 2-5x faster than revenue, a $17 billion Groq license that created $14.4 billion in goodwill on zero revenue, and a $40 billion equity portfolio entangling NVIDIA with its own customer ecosystem. These aren't reasons to sell — they're the specific conditions investors should monitor at 25x forward earnings.
Expedia reported 41.8% operating income growth on 7.6% revenue growth — 5.5x operating leverage that made FY2025 look like a transformation story. But the 10-K reveals that revenue margins were flat at 12.3% in both years, meaning every basis point of margin expansion came from cost cuts. The consumer-facing B2C business grew just 2.4% while B2B drove 18.2% growth. And the record $3.11B FCF is partially inflated by a $10B deferred-bookings tailwind that reverses if bookings slow. What replaces the restructuring story?
EQT Corporation reported a 67.7% EBITDA margin in FY2025 — the highest among major energy companies and 20 percentage points above EOG Resources. Yet the 10-K reveals a paradox: the Equitrans Merger that built the best cost structure in natural gas also created $22 billion in total economic obligations, compressed ROIC to the lowest in the peer group at 7.39%, and left shareholders with just 13.7% of $2.8 billion in free cash flow while every peer returned 47% or more. The filing shows 64% revenue growth was three non-repeatable tailwinds stacked on flat organic production, and 2026 guidance confirms it: flat output, 18% more capital, rising intensity.
Markel Group reported a 23% decline in net income for FY 2025 — the kind of headline that sends retail investors running. But the same filing shows free cash flow grew 9% to $2.55 billion, adjusted operating income grew 10% to $2.30 billion, and management's own intrinsic value framework says the stock is 20% undervalued at $2,150. The catch: 70% of Markel's underwriting profitability above breakeven depends on $484 million in annual reserve releases that surged from near-zero just two years ago.
Public Service Enterprise Group earned $26 million in free cash flow in FY2025. The company just committed to a $22.5-25.5 billion capital plan requiring $4.8 billion per year — a 47% increase from current levels. The 10-K reveals the debt-funded share of investment rises from 38% to 53% at plan-level spending, creating hidden leverage acceleration masked by the headline net debt/EBITDA improvement. Meanwhile, PSEG Power's genuine 4.7x earnings improvement is invisible behind GAAP volatility, and New Jersey's permanent summer shutoff moratorium has pushed bad debt reserves to 11.6% of gross receivables — 2-3x the utility industry norm.
AbbVie buried $15 billion in non-cash charges inside its fiscal 2025 10-K — more than triple its reported net income. The 96x GAAP P/E collapses to 21x on a cash basis, but $25.4 billion in contingent consideration obligations and a franchise concentration reboot (42.3% of revenue in two drugs, heading to 51.5%) reveal the truth sits between the extremes.
Xcel Energy reported 8.57% ongoing EPS growth in FY 2025 — right at the top of management's 6-8% target. But GAAP EPS actually fell 0.58%. The 10-K reveals a $10.9 billion capital machine operating at 3.68x depreciation, generating negative incremental returns (-0.86% ROIIC) while diluting shareholders 8.57% annually. Production tax credits don't benefit shareholders despite the -13.8% effective tax rate, AFUDC inflates 8.2% of net income, and wildfire charges classified as 'non-recurring' have now appeared in three of the past five years.
Ford Motor Company reported $12.5 billion in free cash flow for FY 2025 — an 85% surge that covers the dividend 4.2 times over. But the 10-K reveals management's own adjusted FCF is $3.5 billion, covering the dividend at just 1.17x. The $9 billion gap, driven by Ford Credit's $162 billion balance sheet, is the most important thing about Ford's financials that standard screeners miss. Add in $14.8 billion in EV write-offs, $2 billion in tariff drag, and a $17.2 billion warranty reserve growing 22% annually, and the 5.7% yield anchoring the bull case has virtually no margin of safety.
Ameren Corporation reported 21% EPS growth in FY2025 — its strongest year in a decade. But the 10-K reveals that $0.83 of every $5.35 in earnings per share came from a non-recurring 8.5% effective tax rate, and the $387M rate case that powered the revenue surge was only effective for seven months. Our analysis normalizes earnings to $4.52/share, reconstructs the transmission segment's true 67.3% operating margin, and maps the PISA cap and earnings sharing constraints that bound Ameren's $20.8B growth story.
Coupang reported $34.5 billion in revenue and a 1.4% operating margin — a number that prices it like a low-quality retailer. But the 10-K segment footnote reveals Product Commerce alone generated $2,485 million in EBITDA at an 8.4% margin, while Developing Offerings destroyed value on every incremental dollar with -3.7% marginal gross margins. Meanwhile, stock-based compensation consumed 100% of operating income, a 64.2% tax rate ate most of what remained, and a $1.2 billion data breach voucher program looms over Q1 2026. This is a profitable business voluntarily subsidizing a billion-dollar bet — wrapped in a geographic tax trap.
Canadian Pacific Kansas City reported 13% EPS growth and returned $4.7 billion to shareholders in FY2025 — the most aggressive capital return of any Class I railroad. But the 10-K reveals that $4.7 billion was 215% of free cash flow, funded by $3.1 billion in new debt and a one-time asset sale. The operating ratio improvement of 160 basis points was inflated by non-recurring tailwinds worth 80-100bps. Revenue per RTM was flat. And the Mexican concession's exclusivity expires in 2037, not 2047. This is a company betting its future earnings will prove today's cash generation was a trough.
NRG Energy returned $1.66 billion to shareholders on just $766 million in free cash flow — a 2.17x over-distribution funded by debt. Weeks later, it closed the largest independent power acquisition in U.S. history at $14 billion. But the filing reveals only 20% of NRG's depreciation reflects real asset wear, -16.8% returns on incremental capital, and $28 billion in total financial commitments at 8.6x EBITDA. This is a binary bet on an EBITDA near-doubling that hasn't happened yet.
Royal Caribbean improved its debt-to-equity ratio by 53% — from 4.54x to 2.13x — without paying down a single net dollar of its $21.3 billion debt. Instead, the cruise giant doubled shareholders' equity through retained earnings. Our deep dive into the FY2025 10-K reveals the 55% incremental operating margin engine, $7.2 billion in government-subsidized ship financing, and the $7.6 billion OCF threshold that determines whether this strategy compounds or collapses.
Ares Management reported a 1% GAAP operating margin on $4.76 billion in revenue — $46.9 million in operating income that makes a grocery store look profitable. But the same 10-K reveals $2,583 million in fee-related earnings hiding behind the GAAP facade, a 55:1 gap driven by $2.6 billion in partnership-structure consolidation costs. At 16.6x FRE with 30% growth and 61% margins, the stock looks cheap — until you account for $1.675 billion in contingent earnout liabilities due by June 2028, $1.38 billion in new floating-rate debt, and a $740 million SBC regime that persists for three more years.
Sempra reported a 36% earnings decline, negative $6 billion in free cash flow, and a 94% dividend payout ratio — yet the stock barely moved. The 10-K reveals why: $1.1 billion in non-recurring charges create the widest GAAP-to-adjusted earnings gap (73%) among major US utilities, producing two contradictory P/E ratios from the same filing. At $88 per share, Sempra trades at either 32x or 19x earnings. Meanwhile, the KKR SI Partners deal is not just $10 billion in cash — $7.9 billion in debt deconsolidation makes it an $18 billion balance sheet event. But California's regulatory disallowances have escalated from zero to $676 million in three years, and wildfire strict liability adds an unquantified overhang. One asset sale close decides which Sempra investors actually own.
Axon Enterprise grew revenue 33.5% to $2.78 billion in FY 2025 while reporting a GAAP operating loss of $62.1 million. The headline SBC figure is $634 million — but the 10-K reveals two additional cost channels totaling $366 million, bringing the total economic cost to approximately $1 billion, or 36% of revenue. When you adjust operating cash flow for financing-classified SBC tax payments, Axon's cash generation flips negative. This analysis unpacks the three-channel SBC model, the $481.7 million working capital paradox behind the $14.4 billion backlog, and what 15.9x EV/Sales actually assumes.
United Airlines generated $59.1 billion in record revenue in FY2025, yet its 7.6% return on invested capital barely clears its 6.2% cost of debt. The 10-K reveals three businesses traveling in three different directions under one ticker: a base airline losing pricing power (PRASM -2.9%), ancillary fees growing at 6.7%, and MileagePlus — now fully unencumbered — surging at 10.3%. At 5.75x EV/EBITDA, the market prices all three as one.
Abbott Laboratories reported $44.3 billion in revenue and a 52nd consecutive dividend increase — the kind of headline that signals a diversified compounder. But the 10-K's segment footnotes reveal that Medical Devices generates 61% of segment operating income, the $23 billion Exact Sciences acquisition targets the lowest-return division, and 1,760 infant formula lawsuits carry zero reserves. Here's what the filing reveals about the real risk-reward behind a 33.6× P/E.
Johnson & Johnson reported $26.8 billion in net income for FY 2025 — a 90.5% surge driven almost entirely by a $7.0 billion talc reserve reversal. Strip that entry and normalized earnings grew just 5.5%, while free cash flow was flat and total debt surged 63% in two years. At $208, the stock trades at 23.4x normalized earnings — not the 18.9x the headline suggests. The 10-K reveals three concurrent risk clocks: a patent cliff hitting hardest in the U.S. (-7.6% STELARA erosion), a leveraged capital return strategy funded by $9.2 billion in new debt at 3-4x old rates, and a 93% FCF payout ratio that only works with continued borrowing.
3M reported adjusted earnings of $8.06 per share for FY 2025 — a 10% improvement. On the same page, GAAP earnings were $6.00 — a 17% decline. The $2.06 gap is the widest in 3M's 123-year history, driven by $3.2 billion in litigation payments filed as a discrete cash outflow for the first time. The filing reveals two fundamentally different companies: a recovering industrial generating 19% ROIC with expanding margins, or a declining litigant whose GAAP free cash flow is consumed by settlement obligations extending to 2036.
Republic Services grew free cash flow 15.8% to $2.4 billion in FY2025 — four and a half times faster than revenue. But the 10-K reveals the Q4 earnings beat was entirely tax-manufactured (3.4% ETR vs 21.5% Q1-Q3), revenue is decelerating on a 3-year arc (7.1% to 3.5% to a guided 3.0%), and acquisitions now account for 37% of total growth and rising. At 30.7x earnings, investors face a $6.1B 2029 refinancing wall and unquantified PFAS risk against $2.3B in closure obligations.
NextEra Energy reported adjusted earnings growth of 8.2% for FY2025 — and GAAP earnings decline of 2.1%. The 10-K reveals the difference: a $1.25 billion federal tax benefit that surged 8.3× in a single year, converting $4.5 billion in pretax income into $6.8 billion in net income. Strip it out, and investors at $80/share are paying 46× earnings for a company that burned $12 billion more cash than it generated. This is the story of America's largest clean energy company and the tax credit machine that holds its valuation together.
Pfizer reported $1.36 in GAAP earnings per share for FY2025 — while paying a negative effective tax rate for the third consecutive year. Strip the tax distortion and EPS falls to $1.04. But the Biopharma segment quietly earned $29.3 billion at a 47.9% margin, up 3.1 percentage points. The 10-K reveals two depletion clocks running simultaneously: $4.68B/year in intangible amortization mechanically lifting GAAP EPS, and $18.5B in patent cliff exposure destroying revenue — with the curves crossing in 2027-2028.
Union Pacific reported 'record-breaking' FY 2025 earnings of $11.98 per share, up 8%. But the 10-K reveals that only 18% of that growth came from actual railroad operations. The remaining 82% — industrial park land sales, share buybacks, and a one-time tax benefit — all disappear in FY 2026, creating a $0.55/share headwind that organic operations must replace. With the Norfolk Southern merger facing unprecedented regulatory scrutiny and no STB precedent, investors are paying ~$69/share for merger optionality they cannot handicap.
Edwards Lifesciences reported Q4 GAAP EPS of $0.11 — missing consensus by 82%. But the same filing reveals 60.5% segment operating margins, among the highest in medtech. The $534 million in litigation, impairments, and restructuring charges that crushed reported earnings mask an adjusted operating margin of 29.4%. The real question isn't whether Edwards is profitable — it's whether the TMTT franchise ($551M, +56.4%) can reach contribution-positive scale before TAVR decelerates, while a $920.8 million tax contingency looms over the fortress balance sheet.
Cigna Group added $27.8 billion in new revenue in FY 2025 — and generated just $220 million in additional operating income. The 10-K reveals Evernorth's PBS sub-segment margin collapsed 55 basis points to 2.65% as the rebate-free transition extracted ~$722 million in foregone profit, exceeding the company's own $500 million optimization savings target. Meanwhile, CI simultaneously ran a $749 million restructuring program and acquired $548 million in new specialty pharmacy goodwill, funded by a $4.5 billion debt issuance.
CME Group posted record revenue of $6.5 billion, record average daily volume of 28.1 million contracts, and record earnings of $11.16 EPS in FY2025. But the 10-K reveals that 62% of pretax income growth came from reinvesting $164 billion in customer collateral — not from the exchange fee business. With a 97.8% dividend payout ratio and the Fed signaling rate cuts, CME's most important profit engine is entirely outside management's control.
FedEx reported Q3 FY2026 adjusted EPS of $5.25, crushing the $4.14 consensus by 27% — but the 10-Q reveals $0.41 per share came from a non-recurring foreign tax loss carryforward, reducing the operational beat to 17%. Meanwhile, FedEx Freight's operating income didn't actually collapse 97% to $8M as GAAP suggests — $126M in spin-off costs allocated to the segment masked an adjusted margin of 6.7%. With the Freight spin-off 10 weeks away, the standalone entity will start life with $3.7B in debt while FedEx parent retains the cash proceeds. The real story of Q3 is a Federal Express package business delivering 7.9% adjusted margins on 6.1% yield growth, offset by labor costs growing 320 basis points faster than revenue.
Linde's Engineering segment reported a 3.1% revenue decline in FY2025 — a number that appears in every earnings recap. But buried in Note 18 of the 10-K: intersegment Engineering sales surged 38% to $2,702M, meaning total activity actually grew 15.7%. Linde is deliberately converting its Engineering segment from third-party EPC into a captive builder for its own $10B clean energy backlog, exchanging one-time fees for 15-20 year take-or-pay gas contracts. Meanwhile, EPS compounds at 25% annually on zero volume growth — powered by pricing, buybacks, and a 2.3% cost of debt against 10.8% ROIC.
Merck reported 8% EPS growth and tripled its buybacks in FY2025. But the 10-K reveals free cash flow collapsed 32% to $12.4 billion, funded by $14 billion in new debt that halved its debt coverage ratio. With Keytruda — 49% of total sales — facing a triple cliff of patent expiry, IRA pricing, and MFN constraints in 2028-2029, we analyze whether the pipeline's $4.4 billion in replacement revenue can scale fast enough to bridge a $31.7 billion gap.
Enbridge reported record adjusted EBITDA of C$20 billion in FY2025 and celebrated its 31st consecutive dividend increase. Revenue surged 22%. Net income jumped 40%. But the 10-K tells a different story: core pipeline toll revenue grew just 1.8%, more than half of revenue is zero-margin commodity pass-through, and the dividend consumed 278% of free cash flow.
The gap between these two narratives — management's DCF showing 1.5x coverage versus GAAP FCF showing 0.36x — comes down to C$7.8 billion in annual growth capex. Is it discretionary? The filing's C$18.3 billion in non-cancellable purchase commitments suggest much of it is not.
We decompose Enbridge's revenue quality, dividend mechanics, segment margins, and leverage trajectory using data from the 10-K, Q3 8-K, and Q2 10-Q to show what the earnings headline misses.
Thermo Fisher Scientific generated $44.6 billion in revenue in FY2025 and grew EPS 7.3% to $17.74. But free cash flow declined 13.4% to $6.29 billion. At 32.6× trailing earnings, the market embeds roughly $50 billion of enterprise value in TMO's ability to keep acquiring its way to growth. The 10-K reveals three hidden subsidies — $283 million in interest rate swap benefits, $16 billion in sub-3% legacy bonds repricing higher, and a declining amortization tailwind — propping up the earnings narrative while $14 billion in near-term obligations come due.
Duke Energy deployed $14 billion in capital in FY2025 — the most any US regulated utility has ever spent — earning just 108 basis points above its 4.41% borrowing cost. The 10-K reveals a $200-220 billion decade capital plan the earnings release never mentioned. Meanwhile, free cash flow collapsed to -$1.7 billion, O&M costs surged 24% hidden by a fuel cost decline, and $10 billion in planned equity issuance contradicts the 'avoiding dilution' narrative. Four funding sources. One self-cannibalizing growth plan.
Gold hit $5,000 and both mining companies and royalty companies are printing record cash. But the SEC filings reveal that the conventional wisdom about miners' 2x leverage is wrong — it's only 1.18x. We analyzed four companies' filings to build a framework that shows exactly when each model wins, anchored to a single number: $2,750.
PNC Financial Services reported record 21% EPS growth in FY2025 — and the stock trades at 10x forward earnings, pricing in zero growth. Four widely-cited red flags explain the market's skepticism: collapsing cash flow, inflated earnings, a stagnant core segment, and stretched valuation. The 10-K disproves all four. But two genuine risks survive the correction cascade: capital markets revenue cyclicality (47% of fee growth from a +24% surge) and a concentrated CRE office maturity wall (57.7% criticized, 38.7% maturing in 12 months).
Medline Inc. grew revenue 11.5% to $28.4 billion — the fastest among its med-surg peers. But the 10-K reveals that $2.9 billion in new revenue produced just $67 million in incremental operating income, a 2.3% capture rate below the company's own cost of debt. With ROIC at 5.96% versus a 6.50% cost of debt, a Q4 gross margin cliff to 24.6%, and a potential $11 billion Tax Receivable Agreement owed to PE sponsors, the filing documents a capital structure that demands more than operations can deliver.
Elevance Health grew revenue $22.1 billion to $199 billion in FY 2025 — and lost $663 million in operating income doing it. The filing reveals a three-layer earnings mask: a non-recurring 880bp tax rate collapse ($593M benefit), a $666M settlement tailwind absorbed by cost deterioration, and aggressive buybacks exceeding free cash flow. At 13.9x trailing earnings the stock looks cheap, but at 15.5x normalized earnings, investors are pricing in growth that the operating trajectory doesn't support. Meanwhile, a $3.9 billion shadow exposure to Mosaic Health — a PE-backed joint venture — dwarfs the $732M headline.
Alphabet reported $132.2 billion in net income for FY2025 — a record. But the 10-K reveals $149.1 billion in contractually locked purchase commitments, $24.1 billion in unrealized equity gains inflating headline growth, and a tax timing benefit that boosted cash flow by $13.5 billion. Adjusted earnings grew 15.6%, not 32%, and free cash flow was flat. The investment question: can Google Cloud's 956-basis-point margin expansion convert the largest corporate infrastructure buildout in history into returns?
Otis Worldwide spent $145 million on its UpLift restructuring program in FY2025 and realized $84 million in savings — a net cost of $61 million. The 10-K reveals the real margin engine isn't the restructuring investors have been told to watch: it's the mechanical decline of New Equipment, where Q4 margins hit a record-low 3.64%. With GAAP EPS down 14% and adjusted EPS up 6%, the $0.55 per-share gap between the two earnings realities creates fundamentally different valuations — 24.8x or ~29x — for the same company.
Prologis, the world's largest logistics REIT, generated an industry-best 79.5% EBITDA margin in FY2025 — yet FFO per share fell 6.0% while dividends grew 5.2%. The 10-K reveals $35.3 billion in debt at 5.06x EBITDA with a $10.68 billion maturity wall approaching at refinancing rates 110 basis points above the current portfolio. The 18% embedded lease mark-to-market provides ~3.1% annual organic rent growth, but that barely offsets the projected interest headwind. Meanwhile, the filing describes data centers as 'selective development' — three weeks before management announced a $25 billion commitment. The entire valuation at 19.9x FFO hinges on whether the data center pivot is optionality or necessity.
Broadcom reported Q1 FY2026 revenue of $19.3 billion (+28.7% Y/Y) and non-GAAP EPS of $2.05, but the 10-Q reveals a more complex picture. A single distributor now accounts for 42% of total revenue — up from 29% a year ago — as AI semiconductor revenue doubled to $8.4 billion. Meanwhile, infrastructure software grew just 1.4% Y/Y, GAAP EPS declined 14.3% Q/Q due to tax normalization, and capital returns of $10.9 billion exceeded operating cash flow by 32%, requiring new borrowings to bridge the gap.
American Tower reported 5.1% revenue growth in FY2025, but the 10-K reveals the US tower business — 49% of revenue — generated just $0.6M in incremental revenue. All growth came from Africa (+17.8%), data centers (+13.9%), and Europe. At 18.0x P/FFO, the market is pricing a growth transition story funded by an 8.4x capital intensity cross-subsidy from towers to data centers — while $12.2B in debt matures in 2027-2028 at higher rates.
Roblox — the gaming platform with 85 million daily active users — reported $6.8 billion in bookings and $4.9 billion in revenue for FY2025, with $1.36 billion in free cash flow. But the 10-K reveals a $1.07 billion GAAP loss bridged almost entirely by $1.13 billion in stock-based compensation. Meanwhile, 63% of that revenue came from prior-year bookings, recognized through a $6.5 billion deferred revenue liability controlled by a single 27-month accounting estimate. The structural cost floor consumes 66.6% of revenue before any personnel costs, $1.6 billion in new lease commitments will double infrastructure expenses, and FY2026 GAAP losses are guided to widen even as revenue grows 25%.
Diamondback Energy reported a 63% EPS collapse in FY 2025 while generating record free cash flow of $8.8 billion at a 58% margin — best in its peer set. The 10-K reveals the entire earnings decline was manufactured by a $3.4 billion full-cost ceiling impairment; ex-impairment EPS was virtually flat at ~$15.00. But the same accounting method that destroyed GAAP earnings also hides $837 million in annual interest — 3.4 times what appears on the income statement — while management's $603 million buyback headline masks just $305 million in net repurchases, all from a single related-party transaction.
Truist Financial returned 104% of its FY2025 net income to shareholders — $5.2 billion through dividends and buybacks — while CET1 capital declined 70 basis points. But the 10-K reveals a bank running at two speeds: Wholesale Banking already operates at 49% efficiency while Consumer Banking's earnings collapsed 17.8%. With 30-89 day delinquencies surging 22.3% and the filing's own stress test showing a $2.4 billion reserve hit would halve EPS, the $10 billion buyback authorization is a bet on a consumer franchise that's weakening faster than buybacks can compensate.
Keysight Technologies reported record Q1 FY2026 revenue of $1.6 billion, beating consensus by $60 million with non-GAAP EPS of $2.17 topping estimates by 9%. But the 10-Q reveals a $101 million tax benefit from an audit settlement drove the effective tax rate to -42%, inflating GAAP EPS by approximately 40%. Beneath the tax windfall, 14% organic growth — double the FY2025 pace — and a 96% Q/Q cash flow surge to $441M with 1.57x cash conversion signal a genuine inflection in Keysight's demand trajectory, driven by AI data center infrastructure buildout.
Bank of New York Mellon reported record revenue of $20.1 billion and 28.4% ROTCE in FY 2025 — numbers that suggest an efficiently run financial institution earning outsized returns. But the 10-K reveals a paradox: one segment generates 49% operating margins on infrastructure that moves $614 billion in securities, while another runs at 17% margins with declining revenue and $16.8 billion in goodwill from an 18-year-old acquisition. At 15.5x earnings, the market prices one company. The filing shows it's two.
Morgan Stanley posted record FY 2025 results: $70.6B revenue, $10.21 EPS, and 21.6% ROTCE — the highest return among large bank peers. But the 10-K reveals a paradox at the core of those results. The $139 billion sweep deposit base driving record net interest income growth is the same asset class-action plaintiffs allege was unfairly compensated. Meanwhile, the 300bp efficiency improvement blends a durable WM non-comp leverage engine with a cyclical IS comp ratio tailwind. At 17.4x earnings — the richest multiple in the large bank peer group — investors are paying for a stability narrative that the filing's own numbers complicate.
Realty Income reported 9% revenue growth and 98.9% occupancy in its FY2025 10-K — but the revenue bridge tells a different story. Same-store rent grew just 1.27%, while acquisitions funded by $2.4 billion in new equity contributed more than half of all growth. We decompose the acquisition treadmill, quantify the 5.4 percentage point dilution drag that halves per-share returns, and assess whether the private capital pivot can break the cycle before a $10.3 billion debt wall arrives.
Fortinet's 80.5% gross margin is one of the highest in enterprise technology. But the FY 2025 10-K reveals a cost inflection that blended margins conceal: service COGS grew 1.5x faster than service revenue, flipping from tailwind to headwind in a single year. Meanwhile, the company sits on $7.05 billion in deferred revenue — more than a full year of sales already paid for but not yet recognized — generating cash 26% ahead of GAAP earnings. With 98.1% of free cash flow consumed by buybacks and a 37-76% CapEx increase guided for FY2026, Fortinet is simultaneously more valuable than its P/E suggests and more vulnerable than its gross margin implies.
Welltower generated $10.8 billion in revenue in FY2025 and lost $480 million at the operating line. The gap between EPS ($1.39) and FFO per share ($4.45) — a 220% spread — is the widest among major REITs, driven by $1.5 billion in immediately-vested stock compensation. Meanwhile, $8.9 billion in equity issuance created a 12.1 percentage-point dilution drag, and UK revenue tripled overnight to 22.3% of the senior housing segment. At 42× FFO, the market is betting demographics will overcome all of it.
Grayscale Bitcoin Trust held 619,526 Bitcoin in December 2023. Two years later, it holds 165,692 — a 73.3% collapse that cost the Trust three-quarters of its assets while its sponsor still collects $281 million a year in fees. The 10-K filing reveals three compounding mechanisms hollowing out GBTC: a fee that actually costs 1.67% (not the stated 1.5%), redemptions that accelerate when Bitcoin falls, and insiders who dumped 92.3% of their own holdings. For remaining holders, GBTC isn't a Bitcoin investment — it's a subscription to a fee machine where the exit is blocked by a capital gains wall.
Simon Property Group reported $14.17 in EPS for FY2025 — a 95% increase that masks a ~25% decline in normalized earnings after stripping the $2.9B Taubman consolidation gain. The filing reveals the acquired TRG portfolio operates at roughly 37% NOI margin versus 75% for the legacy portfolio, compressing segment margins by 137 basis points. Meanwhile, The Mills format delivered 8.7% rent growth at 99.2% occupancy, challenging the monolithic 'dead mall' narrative. With $5.91B in 2026 debt maturities refinancing at a 73 basis point premium, SPG's 4.7% rent growth must outrun three convergent headwinds.
MercadoLibre generated $10.8 billion in free cash flow in FY 2025 — or $1.5 billion, depending on which line of its own 10-K you read. That $9.3 billion gap reveals a fintech business consuming 86% of cash flow for customer deposits and loan originations. Meanwhile, $1.1 billion in credit card receivable gains — classified as 'non-recurring' both years — represent 56% of net income, pushing the effective P/E from 51x to 117x. And Argentina, generating 3.6 times Brazil's adjusted operating margin in a hyperinflationary economy, supplies the profitability that makes the story work.
Marsh generated $5 billion in free cash flow in FY2025 — 25% more than the prior year — yet earnings per share grew just 3.1%. The 10-K reveals that the company's $7.75B McGriff acquisition compressed R&I segment margins by 160 basis points, with acquired revenue arriving at 14.1% incremental margin — half the existing 28.4% base. But beneath the GAAP noise, Marsh's cash conversion ratio of 1.27x is the highest among its financial services peers, and the company returned 89% of net income to shareholders while running a $400M efficiency program that the filing itself warns may not deliver.
Apollo Global Management reported $32 billion in revenue and declared 'record' earnings — while the 10-K shows net income fell 24.2% to $3.4 billion. The filing reveals that 42.5% of Asset Management fees are captive intersegment payments from its own Athene insurance subsidiary, SRE missed 10% growth guidance at just 4.2% as net investment spread compressed to 1.61%, and a $755 million non-operating insurance liability swing drove 70% of the GAAP earnings decline. At $145/share, APO trades at 25x GAAP P/E (expensive) and 9x EV/FCF (cheap) — the answer depends on whether Athene's spread compression stays below 25bp annually.
Robinhood reported $4.47 billion in revenue — up 52% — with $1.88 billion in net income and a 42% margin. Then you read footnote 13: 55% of that revenue flows through market maker counterparties.
Vertex Pharmaceuticals reported $12.0 billion in FY 2025 revenue with an 86.2% gross margin — numbers that suggest a well-oiled biotech machine. But newly disclosed segment data reveals that $1.05 billion of Vertex's $1.65 billion COGS is royalty payments, not manufacturing costs. Strip out the royalties and the real product margin is 95.0%. The problem: Royalty Pharma is now seeking to double the royalty rate on ALYFTREK, Vertex's next-gen CF product — putting $200 million or more in annual EBIT at stake in a company where 98.5% of product revenue comes from cystic fibrosis.
Boston Scientific has written off $9.9 billion in cumulative goodwill impairments — 35% of every dollar ever booked. Yet in FY2025, the serial acquirer's ROIC crossed 9.1% for the first time in 20 quarters. Then BSX announced $14.5 billion for Penumbra, which will spike leverage from 1.95× to ~4.1× net debt/EBITDA — landing 0.65× below the covenant ceiling. The 10-K reveals a company at a binary inflection: the acquisition lifecycle model either compounds from here, or a history of goodwill destruction repeats at unprecedented scale.
Bristol-Myers Squibb just reported the largest earnings swing in its history — net income reversed from a $8.9 billion loss to a $7.1 billion profit. Wall Street sees a recovery story. But the 10-K reveals free cash flow declined 7.9% to $12.8 billion, and three compression vectors — IRA pricing, an unprecedented government trade deal, and EU generics already launching — are converging on BMY's revenue faster than the market's 2028 patent cliff timeline implies. At 11.7% FCF yield, BMY is either the most undervalued cash generator in large-cap pharma or a controlled decline the market has priced correctly.
Williams Companies reported record Modified EBITDA of $7.7 billion in FY2025 while free cash flow collapsed 58% to $1 billion. The 10-K reveals the company's data center power contracts are 10-12.5 years — half the 20-year duration management cited — with zero quantified return metrics in the filing. Meanwhile, $965 million in equity-method JV income inflates the preferred EBITDA by 15%, creating a 17% valuation gap. At a 243% FCF payout ratio, WMB's 3.3% dividend yield is funded by debt, not cash flow.
Ferrari returned €1.3 billion to shareholders in FY2025 while carrying just €72 million of industrial net debt — a 0.03x leverage ratio on 38.8% EBITDA margins. But the 20-F reveals that widely-cited free cash flow of €1.9 billion excludes €458 million in capitalized BEV development costs. True FCF is €1.4 billion, and capital returns consume 93.5% of it. Meanwhile, a €89 million Patent Box tax reversal explains why net income grew only 4.8% despite EBIT growing 11.8%. The fortress is real. The margin of safety is thinner than it appears.
Fifth Third Bancorp delivered record FY2025 profitability — $6.0B in net interest income, NIM of 3.11%, EPS growth of 12.4%. But the 10-K filing reveals that nearly two-thirds of the NIM expansion came from an $875M interest expense reduction, not asset yield improvement. With the Comerica merger creating the 9th largest US bank and zero integration costs disclosed in the filing, the investment question becomes whether $400M in synergies can be captured before the rate tailwind turns.
Gilead Sciences reported a 1,673% net income surge in its FY 2025 10-K — from $480 million to $8.5 billion. But buried in the filing's geographic revenue tables is a $636 million pricing gap no analyst has calculated: Biktarvy U.S. revenue grew 5.6% while international grew 11.5%. That gap is the IRA's real cost, derived from Gilead's own data. Meanwhile, the adjusted operating business grew 20%, normalized EPS is $6.42 not $6.78, and three regulatory threats are converging on the company's HIV franchise.
Stryker just posted its fourth consecutive year of double-digit revenue growth, crossing $25 billion. Operating margins expanded 314 basis points. But the 10-K reveals that $807 million of that expansion is an impairment swing — not operational efficiency. The annual cost of being a serial acquirer totals $1,659 million, exceeding Stryker's entire R&D budget, and all of it is excluded from the adjusted earnings that underpin a 41x multiple. Meanwhile, 68% of pre-tax income flows through international operations generating just 24% of revenue, and the German tax authority has assessed $754 million.
Spotify nearly doubled net income to €2.2 billion in FY2025 — the best year in the company's history — and Wall Street responded with 33 Buy ratings. But the 20-F reveals that €510 million of the profitability improvement came from a one-time deferred tax asset recognition that will never recur. With unrecognized DTAs exhausted from €818 million to €41 million, FY2026 net income may be flat or decline even as operating income grows 20%. The trailing P/E of 46.6x is tax-inflated; the tax-normalized P/E is 50-57x. Meanwhile, €1.5 billion in exchangeable notes mature in March 2026, the first-ever buyback program is accelerating, and a €358 million royalty lawsuit threatens the audiobook bundling strategy.
Danaher reported its worst operating margin decline in years — down 130 basis points to 19.1%, with Life Sciences collapsing to 7.1%. But the 10-K reveals that $562 million in non-cash impairments explain 92% of the decline. Adjusted margins were flat. FCF per share rose 2.2% while EPS fell 4.5%. And exactly 8 days before filing, Danaher announced a $9.9 billion Masimo acquisition that will push leverage to ~4.4x — its highest since the financial crisis. This is not a margin story. It's a capital allocation inflection point.
Imperial Oil reported a 32% earnings collapse in FY 2025 — its steepest since the pandemic. But the 10-K's reconciliation tables reveal $1.031 billion in one-time charges (Norman Wells end-of-life and restructuring) explain 68% of the decline. Strip those out, and you find a company producing at a 30-year high of 438,000 boe/d, returning 142% of net income to shareholders, and funding the highest total shareholder yield (10.8%) among major energy peers — all while borrowing from ExxonMobil at 2.7%.
Bloom Energy's stock surged 527% in eight months on a $20 billion backlog and 37% revenue growth. But the FY2025 10-K reveals that only $419 million — or 2.1% — of that backlog qualifies as a GAAP Remaining Performance Obligation. The other 97.9% includes $14 billion in service contracts terminable annually for convenience, product backlog inflated by anticipated tax credits, and 43% concentration on a single related party. Meanwhile, the company issued $2.5 billion in 0% convertible notes and posted a GAAP net loss of $88.4 million in its record year.
Palo Alto Networks delivered $2.594 billion in Q2 FY2026 revenue (+15% Y/Y) and $1.03 non-GAAP EPS, beating consensus by 10%. But the stock dropped 5% on Q3 EPS guidance of $0.78–$0.80 versus the Street's $0.92. The 10-Q reveals why: stripping CyberArk's ~$280 million quarterly revenue contribution from Q3 guidance implies organic sequential growth of just 2.7%, while 73% headline incremental operating margins collapse to 16% when stock-based compensation timing is removed. With $6.45 billion in non-cancelable cloud commitments and four new CyberArk-related risk factors, the platformization thesis faces its first real integration test.
Caterpillar grew revenue 4.3% to $67.6 billion in FY 2025, but every new dollar destroyed $0.69 of operating profit — an incremental margin of -69.1%. Buried inside the collapse: a $3.5 billion capex concentration in Power & Energy while Construction Industries hemorrhages pricing power. The filing reveals two fundamentally different businesses sharing one ticker, one balance sheet, and a 30x P/E that demands a $2.17 billion pricing reversal to justify.
Medtronic delivered Q3 FY2026 revenue of $9.02 billion — its highest growth rate in 10 quarters at 8.7% year-over-year — with non-GAAP EPS of $1.36 beating consensus by $0.02. The stock fell 3.2% anyway. The 10-Q filing reveals why: $306-356M in MiniMed separation costs are front-loaded into the income statement, a $1.146B antitrust verdict went unmentioned in the press release, and Medical Surgical's $19.8B goodwill sits on a cushion of just 12%. Cardiovascular acceleration — up 13.8% with PFA capturing 80% of the EP market — is the real story, but the separation repricing makes it harder to see.
Equinix generated $4,530M in Adjusted EBITDA and negative $400M in free cash flow in FY2025 — a $4.9 billion definitional chasm. At $766 per share, the world's largest data center REIT trades at 22× FFO while running FCF-negative for the first time, funding $1.86B in dividends entirely from capital markets while betting $4.3B that AI demand will fill 52 data centers under construction. The entire bull/bear debate collapses into one number: $284M in 'recurring capex.'
IBM reported a 76% net income surge in FY2025 — its best earnings growth in over a decade. Two weeks later, the stock crashed 13%, erasing $31 billion, when Anthropic announced Claude Code could automate COBOL modernization. The 10-K reveals both reactions were wrong: two non-recurring events inflated earnings by $3.1 billion, true growth was 8-9%, and consulting — the supposed AI victim — is only 17.1% of IBM's profit. The real story is a quiet software takeover generating 61.7% of segment profit at 31.1% margins.
Chevron's FY2025 10-K reports a 106% dividend payout ratio — the company paid more in dividends than it earned. But operating cash flow tells the opposite story, covering the dividend 2.6 times. The $73.5 billion Hess PP&E step-up created the widest cash-vs-earnings divergence among supermajors, collapsed ROIC from 9.5% to 5.9%, and transformed Chevron from the least leveraged to the most leveraged in its peer group. Here's what the filing reveals about dividend safety, balance sheet risk, and when the Hess bet might pay off.
Booking Holdings generated $9.1 billion in free cash flow in FY2025 — a record — while GAAP earnings fell 8%. The $26.9 billion online travel agency's 10-K reveals why: $2.7 billion in accounting artifacts (FX losses on EUR debt, a KAYAK impairment, and convertible note charges) consumed the entire $1.67 billion operating improvement. Underneath the GAAP noise, adjusted EBITDA grew 20.5%, the merchant moat deepened to 70% of bookings, and the self-funding cash flow flywheel returned 84% of FCF to shareholders. The complication: the AI disruption that triggered a $457M write-down of KAYAK is the same force BKNG is betting $700M to harness.
Shopify's gross margins have declined from 50% to 46% over six quarters. Most investors see this as a red flag. But Shopify's FY2025 10-K reveals a counter-intuitive dynamic: as lower-margin Merchant Solutions grows to dominate the revenue mix, operating margins are actually expanding — because it needs dramatically less incremental S&M and R&D. The question is whether fintech risk changes this equation.
Palantir paid $23 million in taxes on $1.66 billion in pretax income — a 1.37% effective rate. The FY2025 10-K reveals exactly why: a single line item in the tax footnote worth $720 million that rises and falls with the stock price. That mechanism boosted reported net income by 19.4%, putting the real P/E closer to 247x than the reported 199x. The filing also shows that only 37% of the $11.2 billion remaining deal value is contractually binding, and that international commercial revenue grew just 2.4% in a year the company grew 56%. The business transformation is real — but so is the amplification.
Citigroup delivered the highest total shareholder yield (8.7%) among US megabanks in FY 2025, returning $18.6 billion through buybacks and dividends. The stock re-rated 63% from its 8-quarter P/B median. But the 10-K reveals 42% of buybacks were funded by depleting the CET1 buffer to just 90 basis points above regulatory minimums — a one-time capital maneuver, not sustainable earnings power. This analysis decomposes what's organic, what's borrowed, and what has to go right in 2026.
Kenvue reported 41% EPS growth in FY2025 — its best result since separating from Johnson & Johnson. But the metric management actually uses to evaluate the business declined 4.6%. The entire recovery traces to $860 million in non-recurring accounting items: impairment cessation, separation cost winddown, and stock compensation forfeitures. Meanwhile, the company's Skin Health & Beauty segment suffered a 30% two-year profit collapse, three product ingredient categories face simultaneous FDA and litigation scrutiny, and 96% of the company's cash sits overseas — while Kenvue pays out 107.6% of net income in dividends. This is the final independent 10-K of a $15.1 billion consumer health company that has already agreed to be absorbed by Kimberly-Clark for $48.1 billion.
Humana grew revenue 10.1% to $129.7 billion in FY2025 — and saw operating cash flow collapse 69% to $921 million. Management calls it 'timing,' but the filing's own receivables numbers explain only 28% of the $2.05 billion decline. Meanwhile, Q4 quietly ran at an estimated 106.6% combined ratio, Star Ratings plunged from 94% to 20% of members in 4+ star plans, and a $3.5 billion revenue headwind looms for 2026. The only bright spot: CenterWell's 5.96% margin engine generates 44.6% of segment profits from 17.3% of revenue.
Cadence Design Systems reported a 93 basis point operating margin decline in its FY2025 10-K — but the same filing reveals that adjusted margins expanded 200+ basis points to 31.1%. A four-surface earnings decomposition shows the widest GAAP/non-GAAP gap in company history (16.4 pp), a $151M legislative tax windfall inflating FCF, and a China 'recovery' that merely returned to FY2023 levels. With $7.8B in backlog and 6.8% ROIIC, the Hexagon acquisition is the swing factor.
MetLife earned $3.2 billion in FY 2025 — and also earned $6.1 billion. Both numbers are in the same 10-K filing. The $2,964M gap between GAAP net income and adjusted earnings isn't an error; it's the defining feature of the largest U.S. life insurer. Our Adjustment Bridge Persistence Analysis reveals that 78% of the gap is structural accounting noise, while 22% represents genuine earnings quality risk — and 26% of the adjusted earnings that produce the 'cheap' 8.9x P/E depend on a single assumption: 9% private equity returns.
Sherwin-Williams used the word 'record' nine times in its Q4 2025 earnings release. The 10-K tells a different story: GAAP income before taxes declined 3.3% to $3,338 million, the 27% FCF surge was nearly half capex normalization, and the $1.17-per-share gap between adjusted and GAAP EPS reflects restructuring and impairment charges entirely absent in FY2024. At 31× earnings, management's own 2–4% growth guidance leaves a 12-percentage-point gap between what the price assumes and what the filing supports. Meanwhile, the most aggressive PSG price increase in recent history tests whether contractor loyalty can absorb a 7% hike in a 'softer-for-longer' demand environment.
NXP Semiconductors reported $2.28 billion in free cash flow for FY2025 — up 20% while net income fell 20%. Wall Street cheered the cash generation. But the 10-K reveals the FCF surge has a $14 billion shadow: a VSMC foundry purchase commitment equal to 115% of annual revenue that doesn't appear in any standard cash flow metric. When you add annualized VSMC obligations back to reported capex, NXP's effective FCF yield drops from 4.2% to 1.6% — below the risk-free rate. Meanwhile, restructuring charges have tripled in two years ($98M to $261M), the new CEO's kitchen-sink Q4 included $100M in R&D cuts during a $1.27B SDV acquisition spree, and incremental returns on new capital are -58%.
Teradyne reported $3.19 billion in FY2025 revenue, up 13.1% — but Taiwan alone generated 149% of that growth. The 10-K reveals $1.8 billion in non-cancelable supply chain commitments, $231 million in cumulative Robotics losses, and deteriorating cash quality masked by reduced AR factoring. At 55.6x earnings, the market prices secular AI growth from a business with a 0.4% five-year revenue CAGR.
AppLovin reported 70% revenue growth in FY 2025 — but the 10-K reveals nearly all of it came from AI pricing optimization (+72% revenue per install), not reaching more users (+3% install volume). With 99.3% incremental operating margins, Cash ROIC of 108.5%, and a $272 million Singapore tax benefit expiring June 2028, the filing paints a picture of extraordinary economics at an extraordinary price. Our 4-Component Growth Sustainability Model separates the $228 billion valuation into testable monitoring metrics.
Arista Networks reported $9 billion in revenue growing 29% with 43% operating margins, zero debt, and a $10.7 billion cash fortress. But the FY2025 10-K reveals something the earnings call doesn't say: 42% of that revenue comes from two customers (Microsoft surging to 26%), $6.8 billion in non-cancellable purchase commitments just jumped 42% in a single quarter, and the headline cash flow quality flatters the income statement. ANET trades at nearly the same P/E as NVIDIA. The filing shows why that comparison deserves scrutiny.
Pinterest reported that net income fell 78% in FY 2025 — and it doesn't matter. Behind a $1.57B tax phantom, $880M in stock-based compensation, and extreme Q4 seasonal concentration, the real story is $1.25 billion in free cash flow growing 33% annually. At 12× trailing FCF, the market is pricing zero growth. The filing reveals whether that skepticism is warranted — or whether investors are being offered a $4.2B revenue platform with 30% FCF margins at a mid-teen cash flow multiple.
SoFi reported $481 million in net income for FY2025 — but 73% came from unrealized fair value gains on loans valued using management's models. The same 10-K reveals a $576 million fee business growing 234% that generates cash without touching the balance sheet. SoFi is transforming from a lender that risks its own capital to a platform that earns fees on other people's. The 3.62x book value — 2.5x LendingClub's — is a bet on which story wins.
Norfolk Southern reported 10.2% EPS growth and 200 basis points of operating ratio improvement in FY 2025. Revenue grew $57 million — less than half a percent. The 10-K reveals that 100% of the earnings improvement came from cost cuts, while the pending $85B Union Pacific merger has inflated compensation by $154M, suppressed capex by 20%, and triggered customer defection. At $289, investors are buying a binary merger bet with asymmetric downside.
S&P Global posted a 42.2% operating margin in FY2025 — the highest since the IHS Markit merger. Two weeks later, the stock dropped 18%. The 10-K reveals that two definitively non-recurring tailwinds — $133M in merger cost elimination and a $273M disposition gain — account for approximately 270 of the 290-basis-point margin expansion, while $1,064M in annual deal amortization suppresses a 50% adjusted margin by 8 percentage points. Ratings generates 45.6% of segment profit at a 63.8% margin, powered by a 28% billed issuance surge that may not sustain. The market repriced growth from 14% to 9-10% — directionally correct, but the underlying quality may be mispriced from both directions.
Vertiv Holdings reported 168.8% net income growth in FY 2025, but our earnings quality decomposition reveals that $203.5 million — nearly one-quarter — came from a non-repeating tax rate correction. The remaining 76% is durable operating leverage on $10.2 billion in revenue, backed by a $15 billion backlog with SaaS-like deferred revenue visibility. But the filing also reveals asymmetric risks: no disclosed cancellation penalties on that backlog, a widening GAAP-to-adjusted earnings gap, and percentage-of-completion estimation risk that grows with project scale.
United Parcel Service eliminated 48,000 positions, closed 93 buildings, and claimed $3.5 billion in cost savings in FY2025. Yet operating margin fell to 8.9%, free cash flow dropped 23% to $4.8 billion, and the company borrowed $4.2 billion while paying out 113% of FCF in dividends. Our analysis of the 10-K filing traces where the $3.5 billion went through a savings absorption waterfall and reveals why the 6.6% dividend yield is a leveraged bet on future margin expansion.
Philip Morris trades at 22x earnings — double Altria's multiple — on the promise that ZYN and IQOS justify a consumer tech premium. But MetricDuck's ROIC data shows Altria earns 44.8% on invested capital versus PM's 34.5%. The FY2025 10-K reveals why: $28 billion in goodwill and intangibles from the Swedish Match acquisition, $11.7 billion in trade receivable factoring inflating operating cash flow, and an Americas segment earning just 10.4% margins while East Asia delivers 47%. The transformation is real. The valuation premium may not be.
Amphenol posted 52% revenue growth, 450bps of margin expansion, and closed its largest acquisition in history. The stock dropped. DuPont decomposition explains why: 86% of the balance sheet growth came from a single quarter, and incremental returns fell below the cost of debt.
Rivian Automotive celebrated its first positive annual gross profit in FY2025 — $144 million after years of losses. But the 10-K reveals a structural trap: the automotive segment still loses $432 million (-11.3% margin) while VW-funded software payments provide the entire surplus ($576M at 37% margin). With 301 million antidilutive shares poised to dilute investors by 25% the moment profitability arrives, and R2 launching into a market stripped of EV tax credits, the path from milestone to sustainable value creation is narrower than headlines suggest.
We screened 1,752 SEC filings for recession-resistance metrics — ROIC, FCF margin, interest coverage, leverage, and consistency. The results were counterintuitive: nearly 80% of the strongest companies are in sectors most investors consider cyclical. Defensive sectors scored lower on every single metric. Here are the 15 most recession-resistant operations in the S&P 500, ranked by the numbers.
86% of Amazon's ROIC decline is capital turnover compression — not margin deterioration. The 10-K reveals why: $59B in new servers, a depreciation reversal on AI hardware, and $440B in irrevocable commitments. At $200B capex, the margin of safety is compressing.
Howmet Aerospace earns 34% gross margins with sole-source pricing power in jet engine castings — yet GAAP ROIC is only 18.6%. The FY2025 10-K reveals why: $4.5 billion in legacy goodwill from the Alcoa-Arconic spinoff chain sits on the balance sheet generating zero revenue. Tangible ROIC is 35%. Cash ROIC hit 28% in Q4. The market prices HWM at 59x earnings anyway — and a $1.8B acquisition is about to add more goodwill.
Screen 50 stocks by ROIC, FCF Yield, and 3 other metrics yfinance cannot provide. Percentile-ranked composite score: Quality (60%) + Value (40%). One script, one API call, no signup.
Micron Technology reported Q2 FY2026 revenue of $23.9 billion, nearly tripling year-over-year as the AI-driven memory supercycle delivered unprecedented pricing power. Gross margins hit 74.4% — double the year-ago level — while the balance sheet flipped from net debt to $6.5 billion net cash after $4.6 billion in debt prepayments. But the 10-Q reveals a deteriorating risk landscape with new tariff and trade investigation risks, a widening patent litigation front, and $27 billion in FY2026 capital commitments that test whether peak-cycle returns can fund the next generation of capacity.
Meta's Family of Apps earns a 51.5% operating margin. Its ROIC is declining at -6.3 points per quarter. Both facts are true simultaneously. The gap between them reveals everything about the largest AI infrastructure bet in advertising history — and the accounting policies designed to make it look cheaper than it is.
Cummins destroyed $770 million on its hydrogen bet in just 14 months — yet its most profitable segment is growing faster than it can build capacity. The 10-K reveals a structural shift Wall Street overlooked: every dollar moving from CMI's declining truck business to power generation earns 2.1× the EBITDA margin. But Accelera's remaining battery business still burns ~$500M annually, roughly offsetting the margin multiplier's benefit. At 14.6× trailing EBITDA, the stock prices in flat earnings — the question is whether the power gen transformation breaks through before the truck trough deepens further.
A stock with a low PE looks cheap. But if ROIC is declining, you're buying a deteriorating business at a discount that keeps getting bigger. Stock Pulse compares any stock to its own 2-year baseline and diagnoses VALUE TRAP or OPPORTUNITY. One API call, no signup.
At $2,800+ gold, all three major gold miners are profitable. The question is what each does with the windfall. Newmont (NEM) is shrinking — divesting six non-core mines for $3.4 billion, retiring $3.9 billion in debt — and its quarterly ROIC peaked at 19.1% in Q2 2025 before declining to 16.9% in Q3. But NEM's 20-quarter median ROIC is 5.6%, and $1.1 billion in divestiture gains inflate current earnings. Agnico Eagle (AEM) produces gold at the industry's lowest all-in sustaining cost — $1,339/oz in FY2025, up $100/oz from FY2024 due to higher royalties — from mines concentrated in Canada, Australia, and Finland. AEM generated record free cash flow of $4.4 billion and grew reserves to a record 55.4 million ounces. Barrick (B) holds 85 million ounces in reserves but its AISC rose to $1,637/oz in FY2025 — making it unprofitable below $1,637 gold while AEM breaks even at $1,339. Barrick's Mali crisis was resolved in December 2025 after costing an estimated $1.25 billion. This analysis compares per-ounce economics, capital allocation strategies, and jurisdiction risk using SEC filing data.
We computed ROIC from XBRL financial data for four agricultural equipment companies. Deere reports 5.9% consolidated ROIC — nearly identical to AGCO's 5.3%. But Deere's three equipment segments earn 23.8% return on $20.6 billion in segment assets. Financial Services earns 1.6% on $70 billion. The same ROIC number simultaneously hides world-class equipment economics and a $70 billion finance arm with deteriorating credit quality, interest coverage below 2x, and $47.5 billion in commitments. AGCO's FY2025 10-K resolved key uncertainties: the $350M revolver drawdown was fully repaid ($0 at year-end), the PTx Trimble goodwill impairment was a FY2024 charge (FY2025 assessment: clean), and interest coverage recovered to ~4.9x. Replacement parts revenue grew 3.2% to $1,873M while tractor sales fell 23% — the balance sheet signature of an aging fleet. Tariff exposure is wildly asymmetric: Deere faces $1.2B in costs, AGCO just $65M. Titan Machinery has three of four segments at a loss.
Five US-listed rare earth and critical minerals companies hold a combined ~$16 billion market cap. We computed cash runway, dilution velocity, and project funding gaps from their latest 10-Q XBRL data — then cross-referenced every government funding headline against what the SEC filings actually disclose. Key findings: NioCorp is 27% funded for the $1.14B Elk Creek Project with an EXIM timeline they 'cannot estimate.' Perpetua is 32% funded for a $2.2B project facing two federal lawsuits. NB shareholders have been diluted 210% from baseline. MP Materials is the only company with binding customer contracts and government price protection — but it's still burning $230M per year in free cash flow.
We computed ROIC from XBRL financial data for every NYSE-listed copper miner with processed filings. Freeport-McMoRan (FCX) — the world's largest publicly traded copper producer — earns the lowest return on invested capital at 9.2%, while Southern Copper (SCCO) generates 24.2% on half the revenue. The 10-K filings reveal why: FCX's Indonesia operations depend on $2.82/lb gold credits to achieve negative cash costs, while SCCO's $0.89/lb cost structure is built on vertical integration and 60-year mine lives.
Our analysis of 2,100+ SEC filers reveals that cost of revenue reporting in XBRL is even more fragmented than revenue. Only 38% of filers use the most common element (CostOfGoodsAndServicesSold), 36% have no standard COGS element at all, and ExxonMobil's $199.5 billion in crude oil purchases is invisible to standard extraction — producing a misleading 100% gross margin.
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.
Accenture beat Q2 FY2026 estimates with $18.04 billion in revenue (+8% Y/Y) and $2.93 EPS — but the 10-Q tells a different story. Below-the-line headwinds from a 390-basis-point tax rate spike and an unexplained $84.5 million non-operating swing consumed $0.25 per share of the $0.30 operating improvement investors didn't see. Meanwhile, EMEA's operating margin collapsed to 10.30% in a quarter with zero restructuring costs — worse than Q1's 12.99% which included $170 million in business optimization charges. The headline 'beat and raise' masks a business where 85% of revenue is growing at low-single-digit local currency rates and the post-restructuring margin recovery thesis faces its first real test.
Regeneron Pharmaceuticals reported $4,505 million in net income for FY2025 — a 2.1% increase that suggests stability through the EYLEA biosimilar storm. But the 10-K reveals operating income fell 10.4%, a $581.5 million equity securities windfall inflated Q3 earnings, the OBBBA tax law nearly doubled the effective rate, and $3.7 billion in share buybacks reduced diluted shares by 6.9%. A four-component earnings bridge shows that none of REGN's headline growth came from selling more drugs at better margins.
Compare any two stocks across Valuation and Quality panels — ROIC, FCF Margin, EV/EBIT, FCF Yield, and Total Shareholder Yield that yfinance cannot provide. One API call, no signup. Clone the repo and run it in 60 seconds.
Developer tutorials that ship working code. Each lab walks you through building a real financial tool with the MetricDuck API — from stock comparison tools to valuation alerts. Clone the repo, run the code, extend it for your own projects.
Data centers need contractors. But which contractor stocks are worth buying? This practical framework covers the 5 metrics that matter: ROIC, backlog, operating margin, DSO, and cash conversion—with real benchmarks and SEC filing examples.
FIX has 40.8% ROIC. PWR has 9.7%—despite 4x the revenue. The 'data center exposure' explanation is incomplete. The full story: FIX earns 2.5x higher margins AND turns assets 30% faster. Here's the DuPont analysis of six infrastructure contractors building America's data centers.
TMUS has the strongest dividend safety metrics—4.29x coverage, 23% FCF payout, 0.5x leverage—yet our analysis flags it as highest execution risk. AT&T, often dismissed for its 2.2x coverage, has the clearest path to dividend sustainability through 2028.
Which MLP offers the safest distribution for income investors? MPLX generates 20.4% ROIC with 45% operating margins. Energy Transfer's 2.74x interest coverage is declining 0.65 points per quarter. EPD's 27-year streak masks deteriorating capital efficiency. The data tells a clear story.
Intuitive Surgical generates 3x the returns on capital of Medtronic and Stryker. DuPont decomposition reveals ISRG wins on both NOPAT margin (25.6% vs 15%) AND capital turnover (0.84x vs 0.49x) — a rare double advantage. The counter-intuitive finding: high stock compensation actually understates their operating efficiency.
Both Jabil (29% ROIC) and Celestica (39% ROIC) rank among the best capital allocators in EMS. But DuPont decomposition reveals they get there through opposite mechanisms — JBL wins on turnover (13x), CLS wins on margin (10.2%). The counter-intuitive finding: CLS has better metrics AND faster growth, yet trades at an apparent discount. Here's what the data reveals.
Boston Scientific has 69% gross margins. Abbott has 56%. Yet Abbott's ROIC is 2.6x higher. This paradox reveals something fundamental about capital efficiency in medical devices—and why screening for 'high margin' stocks can lead you astray.
High BDC yields attract income investors, but most analysis stops at yield percentage. This 5-point framework using dividend coverage, cash conversion, and AFFO trajectory reveals the critical difference between ARCC's negative dividend coverage (-2.42) and MAIN's sustainable payout (+1.06).
The conventional wisdom says Visa is the safer, higher-quality payment stock. Our ROIC analysis tells a different story: Mastercard's 78% ROIC dwarfs Visa's 36%, and the reason reveals something important about how to read these metrics.
Dividend investing isn't just about yield—it's about sustainability. Our dividend research library covers payout ratio analysis, hidden liability screening, and sector-specific safety rankings across insurance, utilities, consumer staples, and more.
CL has the highest ROIC (34.4%), yet PG—with the lowest ROIC (23.9%)—shows the strongest improvement trajectory (+21.6% over 8 quarters). Meanwhile, KO's superior cash generation masks a $12 billion tax liability. Here's how to read beyond the headline metrics.
All three companies raised dividends in 2025. All three beat Q3 earnings estimates. But one has a $950 million hidden liability that most investors don't see. Progressive's 26.8% payout looks sustainable until you examine what's buried in the filing footnotes.
Oracle's -$10B Q2 free cash flow triggered alarm bells. Decomposition reveals the collapse is 70% working capital timing, not structural deterioration. The real story: $248B in off-balance sheet lease commitments dwarfs their $100B balance sheet debt. This is the largest infrastructure bet in enterprise software history.
Intel's Q4 2025 reveals three diverging businesses: CCG's 4.7x operating leverage makes it vulnerable, DCAI's 227% incremental margin on new revenue is the hidden turnaround story, and Foundry's loss-per-dollar worsened despite revenue growth.
Netflix's Q4 2025 earnings revealed a 4-6 percentage point growth deceleration for 2026. We analyze the structural factors behind this shift and decode the $83B Warner Bros acquisition strategy.
Adobe's ROIC didn't just beat peers—it doubled in 8 quarters. The story isn't 'Adobe has high ROIC.' It's WHY it improved from 50% to 105% while Salesforce stayed at 12%. DuPont decomposition reveals the answer.
Investors often equate bigger capex plans with growth. But our cash flow analysis reveals the opposite: Entergy's smaller, focused $7.6B program has better OCF coverage than AEP's $72B plan. When all three utilities are burning cash, execution discipline matters more than scale.
Three fuel cell companies target AI data center power, but execution quality varies dramatically. Bloom Energy's +4.5% ROIC vs Plug Power's -104% isn't a small difference—it's a fundamental divide between a functioning business and a value destroyer. Our SEC filing analysis reveals warning signs even for the winner.
GE Vernova's Electrification segment grew +32% YoY, but the company trades at 98.9x earnings with a 3.57% ROIC. Eaton's Electrical Americas grew +15% with 30.3% operating margins—twice GEV's margin at half the P/E. MetricDuck data shows why quality beats hype in AI infrastructure.
At $25B market cap and 41x revenue, Rocket Lab prices in Neutron success. Our 5-pass filing analysis reveals what the narrative misses: Launch Services growth inflated 32% by catch-up accounting, backlog concentration (69% top 5) exceeds revenue concentration, and gross margins compressed from 55% to 37%. Here's what to watch before the rocket flies.
PSX 21% ROIC vs VLO 4.5%? The numbers lie. VLO's Cash ROIC is 23.3%—actually the highest. One-time California impairments distort the picture. Here's the analytical framework that matters.
Yum Brands' shareholder equity is -$7.5 billion. Yet its ROIC is 50.8% - 2.2x McDonald's 22.8%. This isn't financial engineering. It's the byproduct of franchise economics: when a business doesn't need capital, it returns excess to shareholders. We break down the mechanism behind negative equity in asset-light models.
All three mega-banks maintain similar capital adequacy (TCE ratios 6.7-6.9%), but reserve quality tells a divergent story: JPM 5.43x vs BAC 2.48x coverage, BAC's flat ALLL despite highest NPLs, JPM's hidden deterioration (NPAs up 23%), WFC turnaround validated (lowest NCOs). The 4-metric framework reveals which banks are prepared for credit stress.
The market bet on CVX after Maduro's arrest, but the data reveals COP's pure E&P model (29.9% ROIC) plus $8.5B in claims creates asymmetric upside. Chevron books zero proved reserves in Venezuela despite being the only US operator. Which exposure would you rather have?
Four hospital operators, four business models, 14x ROIC spread. HCA's scale generates 19.2% returns. THC's ambulatory surgery centers deliver 16.8% margins. UHS's behavioral health focus shows improving trends—but deteriorating litigation risk. CYH's rural model is structurally broken. Static ROIC analysis misses the risk velocity layer that determines which returns are sustainable.
Palantir (14.6% SBC) and AppLovin (2.4% SBC) both grew revenue ~65% in Q3 2025. So why does one require 6x more equity compensation? Our SEC filing analysis reveals diverging trajectories that matter more than static spreads: APP's SBC is declining 38% YoY while PLTR's is accelerating 42% YoY. The difference: human-intensive vs AI-engine scaling economics.
CrowdStrike's 10-Q tells a rare story: risk factors that are actively materializing. Two risks escalated, one is new, and zero have been resolved. The July 19 incident created $101M+ in expenses, litigation with no disclosed maximum exposure, and management admissions that read like warnings, not disclaimers. This is a test case for reading risk factors seriously.
Wall Street treats homebuilders as a monolithic 'housing play.' But our analysis reveals a 4x ROIC spread (NVR 35% vs LEN 9%) that cannot be explained by margins alone. The divergence stems from three structural factors: NVR's asset-light lot model vs LEN's Millrose spin-off creating structural margin compression, DHI's $26B in specific performance land contracts creating downside asymmetry, and NVR's paradoxically 'cautious' guidance signaling genuine competitive strength.
The semiconductor equipment ROIC spread (LRCX 54% vs AMAT 27%) is observable in any screener. But SEC filings reveal four hidden signals: AMAT faces TWO government investigations plus $181M restructuring, KLAC's AI packaging segment grew 37% with 401bps margin expansion (proving AI demand is real), LRCX's cash conversion is declining -23.7% (hidden stress beneath high ROIC), and ASML's 4-customer concentration creates feast-or-famine cycles.
SanDisk's +569% stock surge looks disconnected from -17% TTM ROIC, but the real story is in the cycle fundamentals: +31% bit shipment growth overwhelming -9% ASP normalization, gross margins inflecting at +7.4pp/quarter, and positive FCF despite transition losses. This is textbook early-cycle recovery, not a bubble.
American Electric Power's dividend growth is accelerating to 7.10% YOY while maintaining the safest payout ratio (57%) among major utilities. Our 8-quarter payout trend analysis reveals AEP's payout improving at -70pp/quarter while NextEra Energy deteriorates at +7.5pp/quarter—an early warning invisible to traditional analysis.
The average 10-K filing contains over 500,000 characters of text. Most investors don't know what sections exist, let alone where to find specific information. This guide breaks down every section and explains what investors actually find in each.
AI infrastructure spending is approaching $500B annually across hyperscalers alone. Our research library covers capex efficiency metrics, depreciation manipulation detection, concentration risk screening, nuclear power for data centers, and supply chain analysis—everything you need to evaluate AI infrastructure investments.
The conventional wisdom says NuScale (SMR) is safer because it has NRC approval. But filing data reveals SMR paid $495M—$6.9M per reactor—to trigger a non-binding customer agreement. OKLO has regulatory uncertainty but uncommitted cash.
Centrus Energy (LEU) returned 264% in 2025 on the HALEU monopoly thesis. But SEC filings reveal a 69% collapse in core SWU pricing and negative gross profit. Meanwhile, Cameco (CCJ) delivered +88% gross profit growth with 530bps margin expansion.
VST's +321% appreciation prices in nuclear optionality, but ignores a $510M battery fire disaster and antitrust lawsuit. CEG's 'capacity price boom' masks 970bps margin spread between regions. TLN's 0.9x interest coverage means debt service consumes nearly ALL operating income. We analyzed 9 filings to surface what the market overlooks.
KLA Corporation delivers 43% ROIC—highest in semiconductor equipment. But SEC filings reveal 33% China revenue concentration, DOJ export investigations, and rare earth supply risks. Our DuPont analysis shows KLAC's zero-debt, asset-light model drives returns, while ASML's 75% invested capital growth explains its ROIC decline. Lam Research emerges as the 'boring winner' with stable returns and minimal risk.
Wall Street loves CVNA's +30% ROIC turnaround story. But our 8-quarter trajectory analysis reveals CarMax is in crisis mode: CEO fired December 2025, forced margin cuts, -8% volume decline. The data shows this isn't online vs traditional—it's about execution quality.
Novo Nordisk outearns Eli Lilly on every margin metric: 84.7% gross margin vs 82.9%, 44.2% operating margin vs 44.4%, and 41.7% FCF margin vs ~27%. Yet LLY's 52% ROIC far exceeds NVO's—which appears as an anomalous -3.6%. This isn't a data error: NVO's negative invested capital reveals an extraordinarily asset-light business model where the standard ROIC formula breaks down. For margin-focused investors, NVO wins. For capital efficiency purists, LLY dominates.
Reddit is playing both sides of the AI data war: licensing to Google and OpenAI for $130M/year while suing Anthropic and Perplexity. SEC filings reveal 'content licensing agreements' driving 'Other revenue' growth. We analyze the dual strategy, quantify the revenue, and assess the legal risk.
Two enterprise AI giants, two very different earnings quality profiles. Palantir (6/10) wins on cash conversion, accounting practices, and litigation risk. Snowflake (4/10) struggles with negative cash conversion (-0.47x) and aggressive software capitalization ($228M). Our 5-pass SEC filing analysis reveals what standard screeners miss.
Eli Lilly's ROIC doubled from 23% to 52% in 2.5 years as Mounjaro and Zepbound drove 54% revenue growth. Meanwhile, AbbVie's Humira declined 56% to biosimilar competition, testing whether Skyrizi and Rinvoq can fill a $20B revenue gap. This pharma divergence reveals how drug pipelines translate to capital efficiency.
LYFT has higher gross margin (42.6%) AND better earnings quality (7/10) than UBER. So why is LYFT barely profitable? The answer: OpEx. LYFT spends 42.2% of revenue on operating expenses vs UBER's 30.6%. Our 5-pass filing intelligence reveals how scale economics beat unit economics in rideshare.
AppLovin's ROIC went from 4.5% to 75% in 2.5 years while revenue growth accelerated from +17% to +86%. Meanwhile, Unity's ironSource merger resulted in -9% ROIC and declining revenue. The data reveals one of the most dramatic capital efficiency divergences in tech history.
Five cloud security platforms, five different approaches to shareholder value. Elastic (7/10) stands out with neutral accounting and sustainable SBC. CrowdStrike (4/10) raises red flags with 22.9% SBC/revenue and multi-front July 19 litigation. Our 5-pass filing intelligence reveals what standard screeners miss.
ARM Holdings generates 97% gross margin—the highest in semiconductors. Yet only 4% reaches free cash flow. Taiwan Semiconductor earns 56% gross margin but converts 63% to FCF. Intel, despite $5.7 billion in CHIPS Act disbursements, still burns cash. We analyzed 20-F and 10-Q filings to explain these capital efficiency gaps.
While Caterpillar is up +60% YTD, most see it as construction/mining. But CAT's Energy & Transportation segment—which powers AI data centers—is the real growth story at +16.8% YoY with stable 20% margins.
Devon Energy generated 48% ROIC in Q3 2025. ExxonMobil generated 11%. Both produce oil and gas. Why the 4.5x difference? The answer lies in capital allocation: XOM returns 73% of FCF as dividends while pure E&P operators reinvest in high-return wells.
Digital Realty has $2.3 billion in construction commitments plus an active SEC investigation. Welltower faces $337 million in put-call acquisition obligations. Standard FFO/AFFO screens miss these entirely. Here's what filing intelligence reveals about 8 major REITs.
Walmart's net margin is actually higher than Costco's (3.96% vs 2.97%). But Costco's ROIC is 2x higher (41.2% vs 20.9%). The difference isn't profit margin—it's asset turnover. Here's the DuPont decomposition of retail capital efficiency.
Honeywell is spinning into THREE companies by 2026. But while management focuses on restructuring, ROIC has quietly declined from 26.3% to 19.4% over 8 quarters. Here's what's dragging down returns and which segments might escape the trap.
Morgan Stanley estimates 99% of software FCF goes to stock-based compensation. But the impact varies dramatically: Snowflake's SBC is 14x higher than Amazon's as a percentage of revenue. Our Filing Intelligence analysis reveals which companies offset dilution - and which leave shareholders holding the bill.
AVGO's top 5 customers account for 40% of revenue. NVDA's largest customer is 22%. ANET depends on Microsoft (20%) and Meta (15%). Standard AI stock screens show none of this. Here's our 2-signal framework for stress-testing AI beneficiaries.
Qualcomm's consolidated margins hide a secret: the QTL licensing segment earns 72% operating margin vs QCT hardware's 30%. When Apple moves fully to in-house modems, what happens to QCOM's profitability? Our semiconductor ROIC comparison reveals the answer.
EQIX's capex intensity jumped +8.7pp in 8 quarters. CCI's ROIC collapsed to -12%. ANET's customer concentration is HIGH (Microsoft + Meta). Standard screeners show none of this. Here's our 4-signal framework for screening AI data center infrastructure stocks.
E&P companies properly account for $35.7B in asset retirement obligations. Pure refiners? Nearly zero—until they're forced to close. Valero's $337M Benicia ARO surprise in April 2025 reveals what investors miss in refinery financials.
Most investors buy stocks based on tips, news, or gut feeling. The edge comes from going to the source: the company's own financial reports. Here's how to research any stock like a professional, starting from zero.
Earnings can be manipulated through accounting choices. Cash flow cannot. Our earnings quality research library covers OCF/NI ratios, FCF consistency, stock-based compensation dilution, depreciation divergence, and buyback effectiveness—everything you need to separate real profits from accounting fiction.
Learn how to monitor $370B in AI infrastructure spending quarterly with a 3-metric framework. Track capex/revenue trends, depreciation manipulation signals, and growth alignment across Google, Microsoft, Amazon & Meta. Updated December 2025 with Meta's $600B commitment and Michael Burry's depreciation thesis.
Return on Invested Capital (ROIC) measures how efficiently a company turns invested capital into profit. Our ROIC research library covers sector benchmarks from 938 companies and detailed peer comparisons across 8 industries—everything you need to identify high-quality stocks.
Michael Burry warns Big Tech will understate depreciation by $176B through 2028. Amazon is taking a contrarian approach—shortening useful lives and accepting a $700M profit hit while peers extend to boost earnings. Learn how to screen for earnings quality using the capex/depreciation ratio.
Earnings quality measures how sustainable and real a company's profits are. High-quality earnings convert to cash; low-quality earnings are accounting tricks. This guide shows you exactly how to evaluate any company using our 3-metric framework with original data.
Headline revenue growth hides segment-level divergence. ServiceNow has BOTH segments growing 20%+, while Salesforce professional services declined -5.7%. Which SaaS company has the healthiest segment mix?
The Magnificent 7 make up 35% of the S&P 500. But comparing them all using P/E ratios is methodologically flawed—they span 5 distinct business models. Our sector-adjusted scorecard reveals GOOGL as best value (17.4x P/E), TSLA as most overvalued (85% optionality premium), and which stocks actually EARN their premiums.
ROIC measures how efficiently a company turns capital into profits. A good ROIC is 15%+ for most sectors, but utilities median is 5.7% while retail is 15.9%. This guide shows you exactly how to calculate, interpret, and screen for ROIC using original data from 938 companies.
Companies spent $2.9 trillion on buybacks since 2019. But are they actually reducing shares? Our analysis reveals the uncomfortable truth that 95% of programs are ineffective.
Three major fintechs, three very different approaches to earnings quality. PayPal (8/10) represents conservative accounting, Block (6/10) shows aggressive tendencies with regulatory exposure, and Coinbase (4/10) raises red flags with 0.10x cash conversion. Our 5-pass filing intelligence reveals what standard screeners miss.
Earnings can be manipulated through accounting choices. Cash flow cannot. Learn the cash conversion framework—synthesizing Damodaran's multi-period analysis, Greenwald's earnings power value, and Sloan's accruals anomaly—to separate real profits from accounting fiction.
Wall Street ranks Lockheed Martin's 30% ROIC as best-in-class. But our 8-quarter trajectory analysis reveals LMT's capital efficiency is actually declining (-0.018 trend) while RTX improves (+0.004). The data suggests the market may be mispricing defense contractor quality.
Learn to calculate free cash flow (FCF = OCF - CapEx) with real 2025 SEC filing data from Adobe, Zoom, Snowflake, and Intuit. Discover why ZM's 41.7% FCF margin beats Adobe, how Snowflake generates positive FCF despite $1.35B net loss, and why Intuit's quarterly FCF swings 7x between tax seasons.
Compare any S&P 500 company against 10 competitors using 20+ metrics with 5-7 temporal dimensions each. MetricDuck's free peer comparison tool automates 3-5 hours of manual Excel work in 2 minutes with 8-quarter trends, YoY analysis, and SEC-backed data.
AMD generates $1.72 in cash per $1 of profit—the best in semiconductors—while NVIDIA manages only $0.84. This signals AMD's earnings are higher quality and more sustainable despite NVIDIA's explosive growth.
Master the 3-metric cash flow quality framework used by institutional investors. Analyze OCF/NI Ratio, FCF Consistency, and Cash Conversion Cycle with validated 2025 data for NVIDIA, AMD, Intel, and Broadcom. Includes specific thresholds and step-by-step methodology.
Reading 10-Q quarterly filings takes 3+ hours per company. For a 20-stock portfolio, that is 240+ hours per year. MetricDuck structures the key insights with QoQ/YoY comparisons so you can analyze any 10-Q in 5 minutes.
Forward guidance is a company official forecast of future revenue, EPS, and margins. Learn where to find it in SEC filings and how to track guidance revisions for 500+ public companies.
Traditional dividend screening misses deterioration. This 4-metric framework with 8-quarter payout trends predicted 3M's 54% cut 6 months early, Intel's 2024 suspension, and Leggett & Platt's 89% reduction. Screen 938 dividend stocks for sustainability in under 30 seconds.
Manual ROIC screening takes 4+ hours per session. Learn our 5-step framework with sector-specific benchmarks from 938 companies showing why retail median ROIC (15.9%) differs from utilities (5.7%). Screen 500+ companies in 30 minutes with sector-adjusted thresholds.
Every guide to Form 20-F is written for companies filing it—not investors reading it. This guide shows you which sections to focus on, what to skip, and how to spot red flags in foreign company annual reports.
If you've searched EDGAR for ASML or Novo Nordisk quarterly earnings and found nothing, here's why: foreign companies use Form 6-K instead of 10-Q. This guide shows you exactly how to find quarterly results for any foreign stock.
Foreign companies file Form 6-K for everything from quarterly earnings to routine governance notices. Understanding this catch-all form is essential for analyzing international stocks—here's what you need to know.
If you've ever searched for a 10-Q from ASML, Novo Nordisk, or Alibaba and come up empty, you're not alone. Foreign companies have completely different SEC filing requirements—and understanding them is essential for investing in international stocks.