TechnipFMC's 10-K discloses a number that doesn't appear in any earnings summary: 61% of the company's $437 million Subsea gross profit improvement came from activity mix, not volume growth. For a $9.9 billion oilfield services company where 80% of subsea orders arrive without competitive bidding, this distinction between mix-driven and volume-driven margins changes the investment thesis. The same filing shows that every factor depressing FY2023's $56 million net income was non-recurring, that three investment-grade upgrades in 18 months unlocked $2.65 billion in financial flexibility, and that operators are shifting to portfolio-level procurement — widening FTI's moat through customer behavior, not just technology.
Karman Holdings' earnings release uses the word 'record' seven times. Nine days later, the 10-K filed April 3 tells a different story: $42.5 million in negative free cash flow, $156 million in unbilled contract assets growing 46% faster than revenue, and SG&A that surged 93%. The $60 million gap between reported net income and actual cash generation — and whether it closes — determines if KRMN's 562x P/E is a down payment on a defense platform or a mispriced bet on a company that hasn't yet learned to convert growth into money.
Snowflake's FY2026 10-K reveals that equity compensation costs shareholders $2.27 billion annually — 48.5% of revenue and 42% more than the reported SBC figure. The filing also discloses a shift from annual to quarterly billing that threatens the deferred revenue engine generating 68% of annual free cash flow in Q4. With $9.8B in contracted RPO providing a floor but litigation expanding into discovery, the equity-for-cash conversion system faces its first structural test.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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?
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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?
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.
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.
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.
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.
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.
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.
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.
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.
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%.
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.
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.
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.
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.
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.
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.