KLA Corporation (NASDAQ: KLAC) — The Toll Booth on Yield, Priced for a Cycle That Never Turns

Institutional Research Memo — Semiconductor Process Control

An independent equity research note Report date: June 10, 2026 | Price (2026-06-09/10): ~$2,140 | Market cap: ~$279B | Enterprise value: ~$280B Fiscal year-end: June | Shares outstanding: ~130.6M | CIK: 0000319201

Note: The main analysis (Sections 1–15) is written as recommendation-free, evidence-first research and contains no price target. The single exception is the Author’s Take block immediately below, which is an explicitly subjective opinion.

⚡ Author’s Take

This block is the author’s own subjective opinion. It is general information and not investment advice. The analysis (Sections 1–15) that follows is written without a recommendation and without a price target.

Verdict: HOLD / best-in-class business — wrong price for fresh money. AVOID buying at ~$2,140 (≈60× trailing / ≈43× forward earnings). Accumulate only on a genuine cyclical washout — roughly the $1,250–1,500 zone (~28–33× a normalized ~$45 EPS, or ~17–20× EV/EBITDA). Not a short: the franchise is the strongest in the WFE complex and the up-cycle is still live. Tag: “The best toll booth in semiconductors, sold at a toll-free-road price.”

KLA is, by my reading, the single highest-quality business in the wafer-fab-equipment complex — better than Lam, better than Applied, on the metrics that matter most. It owns ~56.5% of semiconductor process control, runs at 6.5× the share of its nearest competitor (who sits below 10%), earns a 62% gross margin and a ~42% operating margin (the highest in WFE by a wide margin), converts ~31% of revenue to free cash flow, throws off ROIC in the 35–68% range, and has gained share for a decade while raising prices. Its moat is the most defensible kind in technology — it sells “learning rate,” the only durable competitive advantage in chip manufacturing — and process-control intensity is structurally rising (from ~5.3% to 7.4% of WFE, targeted at 9% by 2030) as chips get bigger, more varied, and 3D. I have essentially no quarrel with the company.

My entire problem is the price. At ~$2,140 the stock trades at ~60× trailing earnings, ~21× sales, and ~43× book — all three simultaneously at the 99.98th percentile of KLA’s own ten-year history. It has never been this expensive on any of these measures, against a decade-average P/E nearer 18–22×. That multiple is being paid on earnings running at a cyclical high (customers describe demand “beyond what supply can support,” fabs “not enough,” utilization maxed), at the top of an AI-capex super-cycle that is barely 24 months removed from KLA’s last down-year (FY2024 revenue −6.5%). That is the textbook peak-earnings × peak-multiple setup. The market is mispricing not the moat — that’s real and arguably under-appreciated — but the cyclicality: the price underwrites WFE marching to ~$215B by 2030 with no deep digestion, China declining only gently, and the multiple holding near record highs. The most telling fact in the whole report: the sell-side’s own mean price target (~$1,855) sits below the current price even as Cantor stretches to $2,500 — when the average bull is underwater, price has outrun fundamentals. Conviction: medium-high on the valuation call, lower on timing (momentum can run for quarters). Flips bullish on a real cyclical reset that resets the entry into the low-to-mid $1,000s with the franchise intact. Flips bearish (a more urgent avoid) if a memory/AI-capex digestion prints WFE down year-over-year while the multiple is still north of ~40×. Own the toll booth — at a toll worth paying. This is not it.

1. Executive Summary

KLA Corporation is the dominant supplier of process control to the semiconductor industry — the inspection, metrology, and data-analytics tools that find defects, measure critical dimensions, and accelerate the yield-learning rate that determines whether a $15–20B fab makes money. Founded in 1975 and celebrating its 50th year, KLA is organized into three segments — Semiconductor Process Control (~89% of revenue), Specialty Semiconductor Process (~5%, vacuum deposition/etch for MEMS, RF, power), and PCB & Component Inspection (~6%) — but it is, in practice, a process-control pure-play with two small adjacencies. FY2025 (June) revenue was $12.16B, up 23.9% off the FY2024 trough, with services a growing ~25% of the total off a vast, decades-lived installed base.

The business quality is, on the evidence, the best in WFE. Process-control share is ~56.5%, up ~2 points since 2021, and 6.5× the nearest competitor — a structurally widening lead. Gross margin is ~62% and operating margin ~42% (Q3 FY2026), the highest in the equipment group and well above Lam (~50%/35%) or Applied. Free cash flow ran ~$4.0B TTM at a 31% margin; ROIC sits at ~35% gross (including cash) to ~68% net of cash; and ROE is ~95% (buyback-amplified). The moat is genuine and, unusually, deepening: KLA’s advantage rests on “learning rate” — its tools generate the data (1s and 0s, not silicon) that lets customers ramp yield faster, the only sustainable edge in a world where every fab buys similar process equipment. That advantage compounds through four reinforcing legs (Section 4): a near-unmatched portfolio (optical, e-beam, X-ray, reticle, metrology, software), an installed-base/data flywheel (1,600+ applications engineers embedded in customer fabs), R&D scale (KLA spends roughly a nearest-peer’s entire revenue on R&D), and design-in/requalification lock-in. Critically, process-control intensity — process control as a share of WFE — has risen from ~5.3% to 7.4% over the cycle and is targeted at ~9% by 2030, because larger AI die, rising design variety (hyperscaler custom silicon), HBM/advanced-packaging complexity, and 3D architectures all demand disproportionately more inspection and metrology.

Capital allocation is exemplary: a 17-year unbroken dividend-growth record (just raised 21%), a target to return >90% of free cash flow, ~$11B of buyback authorization, and incentives tied to relative free-cash-flow margin and operating-margin dollars rather than revenue or size. Unlike Lam, KLA has used selective M&A (Orbotech, $3.4B, 2019) and carries modest net debt (~$0.95B), but the integration record is good and leverage is trivial (<0.2× EBITDA).

Two structural caveats temper the quality. First, cyclicality: WFE is a sharp cyclical, and KLA — though cushioned by ~25% recurring service revenue — swings with it (FY2024 revenue fell 6.5%; the 2023 memory downturn was severe industry-wide). Earnings today are at/near a cyclical high amid an AI-capex boom that management itself calls supply-constrained. Second, China: 33% of FY2025 revenue (down from a 43% peak), exposed to export controls and domestic-substitution pressure at the trailing edge. Neither breaks the franchise; both cap and cyclicalize it.

The entire tension sits in valuation. At ~$2,140 (EV ~$280B) KLA trades at ~60× trailing / ~43× forward earnings, ~21× sales, and ~43× book — all at the 99.98th percentile of its ten-year history, the most expensive the stock has ever been. Embedded-expectations analysis (Section 10) implies the market is underwriting the Investor-Day 2030 model (revenue ~$26B, EPS ~$84) substantially and a permanently elevated multiple — i.e., that the AI build-out has abolished the WFE cycle. Our scenario work is negatively skewed: even the base case (earnings roughly doubling by 2030) struggles to produce attractive returns once the multiple normalizes from a record extreme. The franchise is correctly understood as the best in WFE; the cyclicality and the starting multiple are, in our reading, mispriced.

2. Business Overview

What KLA does. KLA builds the tools that measure and inspect semiconductors rather than the tools that build them. Where Lam etches and deposits, ASML prints, and Applied does a bit of everything, KLA owns the process-control layer: it finds defects (on wafers, reticles/masks, and unpatterned films), measures critical dimensions, film thickness, overlay/alignment, and topography (metrology), and — increasingly — turns the resulting petabytes of fab data into yield-improving analytics. The pitch, in CEO Rick Wallace’s words, is that “learning faster is the only sustainable advantage in semiconductor manufacturing” — because every chipmaker buys broadly the same deposition/etch/litho equipment, the differentiator is how fast they ramp yield, and that is what process control enables. KLA’s output is “1s and 0s … just information,” which the customer pairs with its own engineers to find and kill defects in a $15–20B fab that starts at zero yield. (Source: FY2025 10-K Business; Investor Day, 2026-03-12.)

The three segments.

The split is stable: process control is ~89–90% of revenue and the overwhelming majority of profit; the two adjacencies (both rooted in the 2019 Orbotech acquisition and earlier deals) are smaller, lower-margin, and more trailing-edge exposed. KLA exited the Display business in March 2024. For analytical purposes this is a process-control company with two modest satellites. (Source: FY2025 10-K Note 18 Segment Reporting.)

How it makes money — systems plus a growing service annuity. KLA sells capital systems (the inspection/metrology tools, often $1–10M+ each, with custom optics — a single high-end lens weighs 300kg — and ~$1M of custom compute inside) and then earns a high-margin, recurring service stream off the installed base. Service was ~$775M in the March-2026 quarter alone (+16% YoY) and is roughly a quarter of revenue, growing at a targeted 13–15% CAGR. Service is the durability engine: as tools become more advanced and stay productive in fabs for longer (average tool life has risen markedly), customers pay for performance, availability, and upgrades — a predictable cash-flow anchor that dampens the systems cycle and rises every year. (Source: Q3 FY2026 call, 2026-04-29; Investor Day.)

End markets and customers. KLA sells into every node of the manufacturing chain — reticle/mask shops, wafer manufacturers, leading-edge logic/foundry, DRAM/HBM, NAND, specialty, advanced packaging, and PCB. Customer concentration is high and structural to WFE: TSMC and Samsung were each >10% of revenue in FY2023–FY2025 (the two named >10% customers), reflecting that the entire leading-edge universe is a handful of names (TSMC, Samsung, SK Hynix, Micron, Intel, plus major Chinese fabs). International revenue is ~89% of the total. (Source: FY2025 10-K.)

Geography — the leading-edge shift in one table. The geographic mix has rotated hard toward the AI leading edge over two years:

The story is in the China-down / Taiwan-up rotation: as US/allied export controls and a softening trailing edge pull China from 43% to 33%, the AI-driven leading-edge build (TSMC up from 18% to 27% of revenue) more than fills the gap — and at higher process-control intensity and better mix. (Source: FY2025 10-K Note 18 geographic table.)

Verdict (Business Overview). KLA is a focused, exceptionally high-quality process-control franchise with a growing recurring-service annuity layered on a cyclical systems business. The model is structurally better than a commodity equipment vendor because it sells the one thing that scales with complexity rather than unit volume — yield-learning — and because each tool sold seeds a decades-long, higher-margin service relationship. Customer and geographic concentration keep it cyclical and geopolitically exposed, but the core is the best-positioned slice of WFE.

3. Industry Dynamics

Market structure and size. Wafer-fabrication equipment is a large, concentrated, structurally attractive industry. SEMI and the equipment makers peg total WFE (including advanced packaging) at roughly $120B in 2025 rising to >$140B in 2026, with KLA explicitly guiding “$140B plus” and — unusually — guiding 2027 to grow faster than 2026 on unprecedented forward visibility from customer fab-construction schedules. KLA’s 2030 framework assumes WFE reaches ~$215B (±$20B), growing ~1% faster than a semiconductor industry compounding at ~11%. The trajectory matters more than the level: WFE has moved from an ~8–10% historical trend to 15–20%+ AI-driven up-years. (Source: Q3 FY2026 call; Investor Day; SEMI 2025/2026.)

A rational, segmented oligopoly — and KLA owns the best niche. The top five vendors — Applied Materials, ASML, Lam, Tokyo Electron, and KLA — control the great majority of front-end equipment, and the structure is unusually good because the leaders are non-overlapping by tool type, each a near-monopolist in its lane: ASML owns EUV lithography; TEL dominates coat/develop; Lam leads etch and is #2 in deposition; Applied is the broad-line #1 in deposition; and KLA owns process control (~56.5% share, 6.5× its nearest rival). Because the players rarely attack each other’s strongholds, rivalry is rational, there is no price war, and industry gross margins have risen. Of these niches, process control is arguably the most defensible and the highest-margin — KLA’s 62% gross margin is the proof. Barriers to entry are formidable: multi-year qualification, decades of accumulated defect-library and recipe data, an installed-base service flywheel, and R&D scale (KLA’s R&D budget approximates a nearest-peer’s entire revenue). (Source: FY2025 10-K Competition; Investor Day; investment-research-frameworks skill.)

The intensity tailwind — the single most important industry dynamic for KLA. Unlike the rest of WFE, KLA’s served market grows faster than WFE because process-control intensity (PC spend ÷ WFE) is structurally rising. KLA’s own series: ~5.3% when WFE was ~$60B, then 5.8% → 6.4% → 6.7% → 7.0% → 7.1% → 7.4% today, targeted at ~9% by 2030. The mechanism is well-evidenced and not merely management narrative:

• Larger die. AI GPUs/accelerators are very large; on a fixed wafer, a given defect density destroys far more value (600 small die with 1% loss vs. 60 large die with 10% loss), so customers inspect/measure more.
• Rising design variety. Hyperscaler custom silicon means many more design starts (100+ designs on 3nm) and many more process flows, each needing variability control — process control scales with design diversity, not just wafer volume.
• Memory becoming logic-like. HBM/advanced DRAM add logic content, TSVs, and lose redundancy; KLA’s DRAM revenue rose ~3× while the DRAM equipment market rose ~2×.
• Advanced packaging. CoWoS/SoIC/hybrid-bonding need front-end-like, nanometer-level inspection; KLA’s packaging process-control revenue is growing from ~$635M (2025) to ~$1B (2026), +~58%.
• More inspection steps per node. KLA cites ~30% more inspection steps at each leading logic node transition.

This is the crux of the structural bull case: even if WFE merely grows with semis, KLA grows faster than WFE on rising intensity, and faster still on share gains — a layered compounding the rest of the group lacks. (Source: Investor Day; Q3 FY2026 call.)

Value chain and profit pools. The equipment layer captures a disproportionate, well-defended share of the manufacturing profit pool because each niche is a near-monopoly. KLA’s position is doubly attractive: it is diversified across all customers and end-applications (it “wins no matter who wins” — every hyperscaler, foundry, or memory maker needs process control), and it sits at the highest-margin point in the chain. The risk is the flip side: equipment sits downstream of the customer capex decision, so when manufacturers collectively pause (a memory glut, an AI digestion), the equipment layer absorbs the full amplitude with a lag.

Regulation — the structural double-edge. US/Dutch/Japanese export controls bar leading-edge tool sales to restricted Chinese fabs. This protects the leading-edge franchise (freezing Chinese rivals out globally) but accelerates domestic substitution at the trailing edge, where China is strongest. KLA frames China as “more or less flat” in spending and growing slower than WFE — a managed headwind, not a cliff. A recent Hua Hong-related restriction was characterized as “fairly immaterial.” (Source: Q3 FY2026 call.)

Capital-cycle read (Marathon). The equipment supply side is disciplined, concentrated, and capex-light (KLA capex ~2.7% of revenue) — a genuine, durable positive that justifies a structurally higher multiple than a commodity cyclical. But the customer supply side — fab construction — is where the capital cycle lives, and the warning signs Marathon watches are emerging there: record fab capex, 15–20%+ WFE growth, management “I’ve not seen this before” optimism, capacity justified by AI-demand forecasts, “Terafab” rhetoric. Because the process-control oligopoly is nearly impossible to enter, mean reversion will not come from new competitors but from the demand cycle turning — historically a sharp event. The disciplined equipment structure does not immunize KLA against a customer-side overbuild; conflating the two is the central error the bull case risks. The favorable supply structure and the elevated demand-cycle position are both true at once.

Verdict (Industry): STRUCTURALLY EXCELLENT — among the best industries in technology — and KLA sits in its best niche, but the cycle is at an elevated, AI-driven high. Genuine barriers, rational non-overlapping competition, rising margins, and — uniquely for KLA — a rising-intensity tailwind that grows the served market faster than WFE. The non-negotiable caveats are deep cyclicality and China trailing-edge erosion. Underwrite the quality with explicit respect for where the demand cycle sits.

4. Competitive Position

The moat, named. KLA’s advantage is the most defensible structure in technology under Greenwald’s taxonomy: demand-side customer captivity (qualified, design-in tools whose requalification risks yield and schedule) reinforced by economies of scale (R&D and a 1,600-engineer service base amortized over the largest installed footprint) and proprietary intangibles (decades of accumulated defect-library and recipe data — the “learning rate” asset). It is not a network-effect or brand moat. It rests on five reinforcing legs:

1. “Learning rate” / data advantage — the dominant leg. Competitive advantage in chip manufacturing is yield-learning speed, and KLA’s tools generate the data that drives it. Because KLA inspects across the entire chain (reticle → wafer → logic/memory → packaging → PCB), it sees failure modes no point competitor sees, and feeds that back into better detection. This is the leg that ties directly to the financial outcome: 62% gross margins do not survive if the tools are commoditized.

2. Portfolio breadth. KLA brings a suite — optical (BBP laser-scanning, Gen 4/Gen 5), e-beam (TeraBeam multi-column), X-ray, reticle inspection, metrology, and data analytics — and its 1,600 applications engineers design each customer’s sampling/inspection strategy. A single-product competitor cannot match the cost-of-ownership of a portfolio that offloads work to the cheapest adequate tool. Common core technologies (custom lenses, sensors, compute) are reused across platforms — leverage a sub-scale rival cannot fund.

3. Installed-base / service flywheel. A vast, decades-lived installed base generates the recurring service annuity and the data/relationship that makes the next sale easier. Average tool life is rising, lengthening the annuity.

4. R&D scale within a narrow market. KLA invests roughly a nearest-competitor’s entire revenue in R&D, sustaining custom development “in every single area” of the tool — “our competitors cannot afford it.” In a market this narrow (a specific inspection step at a specific node at a specific customer), that scale is decisive.

5. Design-in / requalification lock-in. As across WFE, once a tool is qualified for a production application and node, customers maintain that selection so long as it performs — requalifying a rival mid-node risks yield. This is the standard WFE switching cost, and KLA enjoys it in full.

Share is stable-to-rising — the moat is widening. KLA committed at its 2019 Investor Day to grow process-control share ~0.5pt/year and has delivered: 56.5% today, +2pts since 2021, 6.5× the nearest competitor (up from 4× seven years ago). Recent share gains: #1 in advanced-packaging process control (from <10% four years ago, +14pp in 2025, ~70% revenue growth); gains in optical pattern-wafer inspection, e-beam (revenue $75–100M → $400M), and mask/reticle inspection. KLA passes Greenwald’s market-share-stability test decisively — it has led process control for decades with a widening lead. (Source: Investor Day; Q3 FY2026 call.)

The moat passes the financial test. The discipline is to ask: if you removed the moat, what deteriorates? Strip the learning-rate/portfolio/scale advantages and the 62% gross margin, ~42% operating margin, and 35–68% ROIC collapse toward commodity-equipment economics. They have not — and KLA’s margins are the highest in WFE, ~12 points of gross margin above Lam. That margin premium is the moat made visible. KLA also passes Greenwald’s high-and-persistent-ROIC test — returns stayed well above cost of capital even through the FY2024 trough. A business earning 35%+ ROIC for years without successful entry is, by definition, barricaded.

Head-to-head. KLA’s “nearest competitor” in process control is Applied Materials (which has an inspection/metrology business) plus Hitachi High-Tech (e-beam) and a handful of metrology specialists (Onto Innovation, Nova, Camtek in niches) — but none approaches KLA’s breadth or share; the nearest sits below 10%. Against the broader WFE leaders the relationship is non-overlapping: Lam (etch/dep), ASML (litho), and TEL (coat/develop) are not process-control competitors. The strategic point is that KLA’s lane is the least contested of the major WFE niches and the one with the strongest intensity tailwind.

Durability risk — the one genuine erosion vector. China indigenization is the quantifiable threat, concentrated at the trailing edge (the Specialty and lower-end process-control segments) where local champions are strongest and export controls hand them share. This caps the China third of revenue (already falling 43%→33%) but does not touch the leading-edge process-control near-monopoly, which Chinese firms cannot replicate — KLA’s hardest e-beam/optical/reticle tools embed decades of optics and algorithm IP. Secondary risks: customer concentration (TSMC + Samsung) and WFE cyclicality.

Verdict (Competitive Position): DURABLE, WIDE — the widest and best-defended moat in WFE. Learning-rate/data advantage plus portfolio breadth plus R&D scale plus an installed-base annuity, with a widening 56.5% share lead and a rising-intensity tailwind. The only material erosion vector is China trailing-edge substitution, which caps but does not break the franchise. If anything, KLA’s moat is under-appreciated relative to Lam’s and Applied’s.

5. Growth History and Forward Opportunities

Historical growth — cyclical, but with a rising secular floor. Revenue over five years:

The FY2021–FY2025 revenue CAGR is ~15% — meaningfully higher than Lam’s ~6% over the same window, because KLA layers share gains and rising intensity on top of the WFE cycle, and because its FY2024 downturn was shallower (−6.5% vs. Lam’s −14.5%), cushioned by service and process-control resilience. The trailing run-rate is accelerating: Q3 FY2026 revenue was $3.415B (+11% YoY), the June quarter is guided to ~$3.575B, and management guides the total company to high-teens calendar-2026 growth with semi process-control systems growing >20% — then 2027 faster still. (Source: EDGAR; Q3 FY2026 call.)

Quality of growth — high on economics, cycle-amplified on durability. The growth is exceptional on margin (40–50% target incremental operating leverage), broadening across end markets (leading-edge logic/foundry, HBM/DRAM, advanced packaging all contributing), and almost entirely organic in recent years. It is lower-quality on durability: it rides an AI-capex up-leg off a recent trough, is concentrated in TSMC/Samsung, and carries China trailing-edge risk. Extrapolating the current ~20%+ systems growth as a baseline would be a serious error — the through-cycle rate is the ~13–17% Investor-Day CAGR, itself dependent on the AI build-out persisting.

Forward opportunities — the layered growth algorithm. KLA’s 2030 model is unusually explicit, and worth decomposing because the thesis is only as good as its drivers:

• Baseline: semiconductors ~11% CAGR (2025–2030); WFE ~1% faster, to ~$215B (±$20B).
• + Intensity: process control rising from 7.4% to ~9% of WFE.
• + Share: KLA targeting another ~150bps of WFE share (to ~9% of WFE), worth ~4.5% growth above the WFE baseline — consistent with the ~160bps/6.5% over-market growth delivered the prior five years.
• + Service: 13–15% CAGR off the installed base, an accelerant as today’s record shipments enter the service base in 2027+.
• = ~$26B revenue and ~$84 EPS by 2030, a 13–17% revenue CAGR and a faster EPS CAGR via operating leverage and buybacks.

Concrete near-term inflections corroborate the algorithm: advanced packaging (+58% to ~$1B in 2026), e-beam scaling ($400M and rising), DRAM/HBM intensity (KLA DRAM revenue +3× vs. market +2×), data-analytics software ($250M), and ~30% more inspection steps per logic node. Treat the specific quantified targets ($26B, $84, 9% intensity) as management hypotheses to validate over time (this management has hit its 2019 and 2022 Investor-Day targets, which earns some benefit of the doubt); the directional thesis — intensity and share both rising — is independently corroborated by the die-size and design-variety mechanics. (Source: Investor Day; Q3 FY2026 call.)

Verdict (Growth): HIGH-QUALITY ON ECONOMICS AND ON THE SECULAR ALGORITHM, BUT CYCLE-AMPLIFIED NEAR-TERM. KLA has the best structural growth algorithm in WFE — the only major name where served-market expansion (intensity) compounds on top of share gains and the WFE cycle. But the current ~20% pace is riding an AI up-leg off a trough; the honest baseline is the low-to-mid-teens through-cycle CAGR, large cyclical amplitude included.

6. Financial Quality

Margins — the highest in WFE, and structurally improving.

KLA’s ~62% gross margin sits roughly 12 points above Lam (~50%) and well above Applied — the clearest single financial signal of the process-control moat. Gross margin is being defended even against a ~100bps headwind from elevated DRAM chip costs (DRAM is an input to the image-processing computers shipped inside KLA tools) and tariffs, which makes the underlying gross-margin expansion (mix toward leading-edge, manufacturing scale, service) more impressive. Notably, management is explicit that it does not price on scarcity (“we don’t price based on scarcity at KLA”) despite customers fighting for slots — pricing is value/cost-of-ownership based, a discipline that protects long-run customer relationships at the cost of leaving some near-term cyclical pricing upside on the table. Incremental operating leverage targets 40–50%. (Source: Q3 FY2026 call.)

Returns — exceptional, read with care. Headline ROE is ~95% (TTM) — but this is heavily buyback-amplified: years of repurchases have compressed book equity (just $5.83B against ~$13B of revenue), mechanically inflating ROE. The cleaner lens is ROIC: with TTM operating income ~$5.4B, NOPAT (at a normalized 15% tax) ~$4.6B against invested capital of ~$11.8B (debt $5.95B + equity $5.83B) gives ~39% gross ROIC, or ~68% net of the ~$5B cash. Even normalizing the low GAAP tax rate, returns stay extraordinary. These are genuine, durable economics — oligopoly pricing power, an asset-light model, and the high-margin service annuity. The DuPont read: the real signal is the ~33% net margin (moat-driven and durable); the equity multiplier is manufactured by buybacks; so anchor forward expectations on the ~33% margin and ~39%+ ROIC, not the 95% ROE. (Source: EDGAR XBRL; company filings.)

Cash flow and capital intensity. KLA is asset-light for a capital-equipment maker: capex runs ~2.7% of revenue ($335M FY2025) and D&A is modest. Free cash flow: FY2023 ~$3.33B, FY2024 ~$3.03B (positive and robust even at the trough — the service-annuity hallmark), FY2025 ~$3.75B, and ~$4.0B TTM at a 31% FCF margin. Over the last five calendar years FCF compounded ~20%, above the ~16% revenue CAGR — operating leverage made visible in cash. That KLA throws off ~$3–4B of FCF even in a down-year is the single most important financial-quality fact for a cyclical: it never has to raise capital at the bottom and can keep buying stock and raising the dividend through a downturn. (Source: EDGAR; Q3 FY2026 call.)

Balance sheet — strong, modestly levered (unlike Lam). As of Q3 FY2026: ~$5.0B total cash/marketable securities vs. ~$5.95B debt = ~$0.95B net debt — investment-grade across all three agencies, with a well-laddered bond maturity profile. This is the one place KLA is less conservative than Lam (which is net cash): KLA has used debt to fund buybacks and the Orbotech deal. But net debt/EBITDA is ~0.16× — trivial — and the modest leverage is a deliberate, low-risk efficiency choice, not a vulnerability. (Source: Q3 FY2026 call/10-Q.)

Quality of earnings — high, with two watch-items. SBC is low for tech (~$265M, ~2.2% of revenue) and more than offset by buybacks (net share-count reduction). Net income is cash-backed (CFO ≈ or > NI across FY2023–FY2025), no goodwill impairments, no channel-stuffing tells. Two watch-items: (1) the GAAP/non-GAAP gap — KLA guides on non-GAAP (Q3 FY2026 non-GAAP EPS $9.40 vs. GAAP $9.12, a ~3% gap, mostly SBC and intangible amortization from past M&A) — modest and clean, but use non-GAAP with eyes open; and (2) the DRAM-cost gross-margin headwind (~100bps) persisting through at least CY2026, which management has secured supply against but which caps near-term margin upside. Neither is a red flag. (Source: Q3 FY2026 call.)

Verdict (Financial Quality): BEST-IN-CLASS — economics clearly improve with scale, and are the strongest in WFE. 62% gross margin, ~42% operating margin, ~39%+ ROIC, ~31% FCF margin, FCF positive through the trough, low SBC, trivial net leverage. The moat surfaces directly and unambiguously in the returns. The principal financial-quality risk is not accounting but position-in-cycle: earnings are at/near a cyclical high.

7. Capital Allocation

Among the cleanest records in large-cap semicap — with one real distinction from Lam. KLA allocates capital across internal R&D, a 17-year-growing dividend, buybacks, and selective M&A.

Capital return — assertive and disciplined. KLA targets returning >90% of free cash flow (recently raised from ~70%), and lives it: TTM capital returned was ~$3.2B; the company just raised the dividend 21% (to $2.30/quarter, $9.20/year) — its 17th consecutive annual increase — and added a $7B buyback authorization on top of ~$3.9B remaining (~$11B total). FY2023–FY2025 dividends were $733M / $773M / $905M (a steady grower at a conservative ~22% payout, leaving room to keep raising through a downturn), and buybacks were $1.31B / $1.74B / $2.15B (plus an outsized $3.97B in FY2022). Management calls itself a believer in “assertive capital allocation … allocating every dollar of cash.” (Source: Investor Day; Q3 FY2026 call; EDGAR.)

Did they buy well? Diluted share count fell from 151.6M (FY2022) to ~133.75M (FY2025) to ~131.4M guided — a ~13% reduction over four years, funded from FCF and more than offsetting SBC. The heavy FY2022 buyback ($3.97B) was well-timed into weakness; the steady pace since is sensible policy. The mild critique — identical to Lam’s — is that the program is steady-state rather than counter-cyclical, so the current pace continues at the most expensive valuations in KLA’s history (99.98th percentile). Buying back stock at ~60× earnings is lower-octane per-share value creation than buying at the trough; it is defensible as a consistent policy but it is not opportunistic, and at these multiples the buyback yield is thin (~1% of market cap). (Source: EDGAR share series.)

M&A — selective, and a genuine point of difference from Lam. Unlike Lam (famously organic, having walked away from the KLA-Tencor merger in 2016), KLA has grown partly by acquisition: the original KLA + Tencor merger (1997) created the company, and Orbotech ($3.4B, 2019) created the PCB & Component and Specialty segments and seeded the advanced-packaging position now compounding at ~58%. The integration record is good — management explicitly “absorbed a company with lower margins … and improved it using the KLA operating model” — and there has been no large, value-destructive deal. The primary “investment” remains internal R&D ($1.36B FY2025, 11.2% of revenue, rising). This is disciplined, Marathon-friendly behavior — but the M&A track record is a thing to monitor, not a clean “purely organic” record like Lam’s. (Source: FY2025 10-K; Investor Day.)

Incentives — well-aligned to returns, not size. The 2025 proxy keys annual incentives and PRSUs to relative free-cash-flow margin, non-GAAP operating-margin dollars, and non-GAAP EPS — explicitly not revenue or headcount, which aligns management with profitable growth and capital discipline rather than empire-building. The pay-for-performance read is strong: a $100 investment tracked to ~$489 (KLA TSR) vs. ~$278 (peer group) over the disclosed window. Hedging/pledging prohibited; ownership guidelines in force. CEO Richard Wallace. The mild gap (as at Lam) is the absence of an explicit ROIC hurdle, though FCF-margin and operating-margin metrics proxy for it well. (Source: 2025 DEF 14A.)

Insider read — neutral. Form 4 activity across the trailing window is routine — grants, tax-withholding on vesting, and 10b5-1-planned sales — with no discretionary open-market purchases by named officers/directors. The standard mature-mega-cap profile: no red-flag dumping, but no bullish accumulation signal either. (Source: EDGAR Form 4 corpus, CIK 0000319201.)

Verdict (Capital Allocation): INTELLIGENT — a genuine strength. Returns-oriented incentives, a 17-year dividend-growth record, >90%-of-FCF return target, a real ~13% share-count reduction, and good M&A integration. Two nuances: the buyback is steady-state rather than counter-cyclical (and is now running at all-time-high valuations), and the record is selectively acquisitive rather than purely organic — both minor, neither thesis-breaking.

8. Changes and Headwinds — Last Two Years

Strategic and operational. The dominant change is the AI-capex inflection reshaping demand and mix: the geographic rotation from China (43%→33%) to Taiwan/leading-edge (18%→27%); advanced packaging emerging as a ~$1B, +58% business with KLA now #1; e-beam scaling 4–5× to $400M; DRAM/HBM intensity rising (KLA DRAM revenue +3× vs. market +2×); and a structural margin step-up (gross margin to ~62%, operating margin to ~42%). The recovery from the FY2024 trough (−6.5%) to record quarters is the headline financial change. (Source: Section 3/Section 5/Section 6 evidence.)

Corporate actions. The March-2026 Investor Day introduced a new 2030 model ($26B revenue / $84 EPS), raised the capital-return target to >90% of FCF, lifted the dividend 21% (17th straight year), and added $7B of buyback authorization. The Display business was exited (March 2024). None of these alter the franchise; they reinforce it.

Headwinds. Four stand out: (1) Cyclical position — demand is “beyond what supply can support,” fabs and clean-room space are the binding constraints, and earnings sit at/near a cyclical high; the AI-capex durability question (some hyperscalers’ quarterly capex came in “a little light”) is the macro swing factor. (2) China — 33% of revenue and declining, with export-control and domestic-substitution pressure at the trailing edge. (3) DRAM-cost margin headwind — ~100bps through at least CY2026. (4) Valuation/expectations — the re-rating to the 99.98th percentile is itself a headwind, leaving no room for disappointment (Section 10).

Verdict (Changes): NET POSITIVE FOR THE BUSINESS, NET NEUTRAL-TO-NEGATIVE FOR THE STOCK. The operational changes (intensity, share, packaging, margin) genuinely strengthen the franchise. But the same period pushed earnings to a cyclical high and the multiple to an all-time extreme, so the thesis-relevant change for an investor is that the easy money has been made.

9. Risk Analysis

Catastrophic-loss assessment. The probability of a permanent total/near-total loss is low: KLA is investment-grade, deeply profitable through-cycle (FCF positive even at the FY2024 trough), and protected by the widest moat in WFE. The realistic severe-downside scenario is not insolvency but a large drawdown — a cyclical earnings decline (perhaps −20–30%) compounded by multiple compression from a record-high starting point, which the bear case quantifies at roughly −50–60% peak-to-trough (Section 10). The tail risk that would be catastrophic is exogenous (a Taiwan/Korea supply-chain disruption), impairing the entire semiconductor complex, not KLA specifically.

Verdict (Risk): The dominant risks are valuation and cyclicality, not solvency or franchise quality. This is the best business in WFE whose stock carries elevated risk purely because of where the price and the cycle sit.

10. Valuation Discussion (Embedded Expectations)

No price target and no recommendation appear in this section, by design. The analysis is framed as embedded expectations and scenarios.

Where the multiple sits. At ~$2,140 (EV ~$280B), KLA trades at:

The single most important fact in this report: KLA’s P/E (~60.6), P/B (~48), and P/S (~21.6) are all simultaneously at the 99.98th percentile of its own ~10-year history (composite percentile 99.98). It has never been this expensive on any of these metrics — more extreme even than Lam’s 99.86th-percentile reading. The decade-average P/E is ~18–22×; the current trailing multiple is roughly 3× that mean. The multiple is pricing a regime change — process control as a permanently faster, less-cyclical secular grower — not a normal up-cycle. (Source: company filings; own-history valuation percentiles, 2026-06-09.)

Versus peers. KLA sits at the expensive end of an already-rich cohort — trailing P/E KLAC ~60.6 vs. LRCX ~61.7 / ASML ~59.5 / AMAT ~47; forward KLAC ~43 vs. LRCX ~42 / ASML ~37 / AMAT ~31. KLA’s premium to AMAT is partly justified by its superior margins, share gains, and intensity tailwind — but the entire group is at multi-year-high multiples, so “in line with peers” is not “cheap”; it is a sector-wide AI re-rating that would mean-revert together. KLA arguably deserves the highest multiple in the group on quality — but “deserves the highest multiple” and “is attractively priced” are different statements.

Embedded expectations. Reverse-engineering the ~$280B EV: to justify it on a ~25× exit P/E (still a premium multiple) requires ~$11.2B of net income, i.e., roughly the Investor-Day 2030 target ($26B revenue × ~33% net margin ≈ $8.6B, plus buyback-driven EPS accretion getting EPS to ~$84 on a shrinking share count). In other words, the current price requires the 2030 plan to be substantially delivered and the multiple to stay well above the historical mean. The market is underwriting (a) WFE compounding to ~$215B with no deep digestion, (b) intensity rising to ~9% on schedule, © China not deteriorating sharply, and (d) the multiple remaining permanently elevated. Each is plausible; all four together, with no cyclical interruption over five years, is an aggressive joint probability.

Scenarios (3-year, illustrative; not targets).

• Bull (super-cycle uninterrupted): revenue compounds ~16% to ~$19B by FY2028, EPS to ~$55–60; multiple holds ~40×; stock compounds modestly positive (~+8–12%/yr). Requires the cycle to not turn.
• Base (plan tracks, multiple normalizes): revenue ~$18B, EPS ~$50 by FY2028; multiple de-rates from ~60× toward ~35× as growth normalizes; stock roughly flat to modestly negative despite earnings rising ~40% — the peak-multiple math. This is the central case and it is unattractive at today’s entry.
• Bear (AI-capex digestion / memory glut): a WFE down-year takes revenue −15–20% and EPS −25–30%, and the multiple compresses to ~25–30×; stock −50–60% peak-to-trough — the peak-earnings × peak-multiple double-count.

The skew is negatively asymmetric: the bull case earns a modest positive return that requires the cycle never to turn; the base case loses despite earnings growing; the bear case is a large drawdown. That asymmetry — not any doubt about the business — is the valuation conclusion.

Verdict (Valuation): the franchise is the best in WFE; the price embeds near-flawless execution of a multi-year plan plus a permanently elevated multiple, at a cyclical earnings high. The return distribution from ~$2,140 is negatively skewed.

11. Variant Perception

Consensus belief. KLA is a wide-moat, share-gaining, best-margin compounder riding a secular AI/process-control-intensity tailwind that has structurally raised its growth rate; the cycle is “different this time” because AI demand is supply-constrained with multi-year visibility, so the premium multiple is warranted. (Reflected in Cantor’s $2,500 target and a string of upgrades.)

Strongest bull case. The intensity-plus-share algorithm is real and underappreciated: KLA grows faster than WFE structurally (intensity 7.4%→9%) and faster than that on share gains (+150bps to ~9% of WFE), and layers a 13–15% service annuity on top — a triple-compounding the rest of WFE lacks. Management has hit its 2019 and 2022 Investor-Day targets, earning credibility on the 2030 plan. Margins are the highest in the group and still rising. If AI capex sustains, earnings nearly double by 2030 and the stock can grow into its multiple.

Strongest bear case. This is a cyclical trading at the 99.98th percentile of its own valuation history, on peak-ish earnings, at the top of an AI-capex super-cycle barely 24 months removed from a down-year. The sell-side’s own mean target sits below the price. Process-control intensity gains and share gains are real but incremental (basis points a year); they cannot offset a WFE down-cycle or a multiple that mean-reverts from ~60× toward ~20×. China is a third of revenue and structurally eroding. “This time is different” — which management explicitly disavows saying while effectively arguing — is the most expensive phrase in investing.

The 3–5 assumptions that matter most:

1. Does the WFE cycle still exist? (Bull: AI has abolished it / muted it; bear: it is merely stretched and will turn sharply.) — The master variable.
2. Does process-control intensity actually reach ~9% by 2030? (Falsifiable: track the published intensity figure each year — it must keep stair-stepping up.)
3. Does KLA keep gaining ~0.5pt of share per year? (Falsifiable: process-control share and the “× nearest competitor” multiple.)
4. Does China deteriorate gradually or sharply? (Falsifiable: China revenue % and commentary; a sharp step-down vs. the managed decline.)
5. Does the multiple hold? (The dominant return driver from here — and the least controllable.)

What would falsify each side. Bull falsified: a WFE down-year (KLA revenue YoY negative) with the intensity gains failing to cushion it, or a multiple de-rate toward the historical mean. Bear falsified: two-plus more years of 15–20% growth with intensity hitting 8%+ and the multiple holding ~40×+, demonstrating a genuine regime change.

Verdict (Variant Perception): The bull and bear agree on the business and disagree only on the cycle and the multiple. The variant perception that matters is not “is KLA a great company” (it is) but “has AI abolished the semiconductor capital cycle” — and betting that it has, at the 99.98th valuation percentile, is the wager embedded in the price.

12. Fact vs. Interpretation

13. Open Questions

1. Cycle durability. How much of the 2026–2027 “unprecedented visibility” is true end-demand vs. customers queuing for slots to avoid shortages (double-ordering)? A pullback would expose it.
2. AI-capex sustainability. Several hyperscalers’ quarterly capex came in “a little light.” Does the WFE forecast implicitly require AI capex to keep compounding — and what hyperscaler spend level does “>$140B WFE in 2026 / faster in 2027” actually require?
3. Intensity ceiling. Is 9% process-control intensity by 2030 a hard target or a soft aspiration? What happens to the growth algorithm if intensity plateaus at ~7.5–8%?
4. China floor. Where does China revenue stabilize, and how much of the remaining 33% is exposed to further controls/substitution?
5. Margin trajectory. Does the ~100bps DRAM-cost headwind reverse in CY2027, and does mix/scale push gross margin above ~62%, or is ~62% the ceiling?
6. Buyback discipline. Will management moderate buybacks at 99.98th-percentile valuations, or keep the steady pace regardless of price?

14. What Must Be True

Bull case — what must be true:

• WFE compounds toward ~$215B by 2030 with no deep cyclical drawdown (AI capex sustains).
• Process-control intensity keeps stair-stepping toward ~9%; KLA keeps gaining ~0.5pt of share/year.
• Service compounds 13–15%; margins hold ~62%/~42%+.
• The multiple stays well above the historical mean (≥~40×) as the market accepts a “regime change.”
• Falsification test: A single fiscal year of negative KLA revenue growth, OR the published process-control intensity figure failing to rise, OR a multiple de-rate below ~35×, falsifies the bull case. Watch the annual intensity figure and the YoY revenue line.

Bear case — what must be true:

• The WFE cycle is merely stretched, not abolished; an AI-capex digestion or memory glut prints a WFE down-year within ~2 years.
• Earnings fall ~20–30% from peak; the multiple mean-reverts from ~60× toward ~25–30×.
• Intensity/share gains prove too incremental to offset the cyclical and multiple drag.
• Falsification test: Two-plus more years of 15–20% revenue growth with intensity reaching ~8%+ and the multiple holding ~40×+ falsifies the bear case — that would be genuine evidence of a regime change rather than a stretched cycle.

The two cases agree on the franchise and diverge entirely on the cycle and the multiple. That is the bet.

15. Source Appendix

See Appendix B — Source Appendix below for the full, categorized source list with URLs and access dates. Primary sources: KLA FY2021–FY2025 10-Ks (EDGAR CIK 0000319201), FY2026 10-Qs, 2021–2025 DEF 14A proxies, EDGAR XBRL company-facts (financial series), the Q3 FY2026 earnings call (2026-04-29) and March-2026 Investor Day transcripts, SEMI/industry data, and market-data and transcript sources. Peer context: public filings of LRCX, AMAT, and ASML.

APPENDIX A — Standard Diligence Questionnaire

KLA Corporation (NASDAQ: KLAC) — Standard Diligence Questionnaire

Supplemental to the main note. Fact / Interpretation / Assumption labeled where it matters. Sector analogs substituted where a question does not map.

General

What thoughtful questions have other investors asked about this company? The recurring institutional questions (from the Q3 FY2026 call and conference appearances): (1) How real is the 2027 visibility — true end-demand or customers queuing for slots/double-ordering? (2) Can you outgrow WFE indefinitely — is the intensity-plus-share algorithm durable or are the gains incremental? (3) Why isn’t the 2030 model higher given current momentum (management: WFE numbers haven’t moved as fast as semi-revenue, which is pricing-driven)? (4) China trajectory and export-control exposure. (5) Gross-margin path given the DRAM-cost headwind. (6) Will you pass through scarcity pricing (management: no — value-based pricing only). The meta-question is whether the AI build-out has muted the WFE cycle enough to justify a record multiple.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: at/near a cyclical high. Management describes demand “beyond what supply can support,” fabs and clean-room space as the binding constraints, and 2027 growing faster than 2026 — classic late-expansion. KLA’s FY2024 down-year (−6.5%) was only ~24 months ago.

Driven by the external environment or internal actions? Both. External: the AI-capex super-cycle and WFE recovery. Internal: genuine, durable share gains (+2pts since 2021) and rising process-control intensity, which raise the baseline through the cycle even as the level is cycle-amplified.

How stable are revenues? Less cyclical than the WFE average — ~25% recurring service revenue and the resilience of process control (yield matters most when fabs are stressed) made KLA’s FY2024 trough shallower (−6.5%) than Lam’s (−14.5%). But still a cyclical: a WFE down-year would take systems revenue down materially.

Outlook for products/services? Strong near-term (high-teens CY2026 company growth, systems >20%; 2027 faster). Long-term: 13–17% revenue CAGR to ~$26B by 2030 (management assumption), service 13–15%.

How big will this market be? WFE ~$120B (2025) → >$140B (2026) → ~$215B ±$20B by 2030 (management/industry estimate); KLA’s served process-control slice grows faster via rising intensity (7.4%→9%). Growing, global (~89% international), AI-driven.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Less in the leading-edge process-control niche — KLA’s share is rising (56.5%, 6.5× nearest competitor) and the lead is widening. More at the China trailing edge, where domestic substitution is encroaching.

How profitable is the business (ROIC, ROE)? Exceptional: ROE ~95% (buyback-amplified), ROIC ~39% gross / ~68% net of cash, net margin ~33%, gross margin ~62% (highest in WFE).

How profitable is the industry — competitors, barriers? Highly profitable oligopoly; KLA’s process-control niche is the most defensible and highest-margin. Barriers: decades of defect-library/recipe data, portfolio breadth, R&D scale (KLA’s R&D ≈ a nearest-peer’s revenue), multi-year qualification, installed-base service flywheel. Nearest process-control competitor is below 10% share.

Can the business be easily understood? Moderately. The franchise economics are clear (toll on yield, rising intensity); the technology is deep, and the cycle dynamics + China + multiple require real work to underwrite.

Can it be undermined by foreign low-cost labor? No at the leading edge (the IP is in optics/algorithms/data, not labor). Partially at the China trailing edge via state-subsidized domestic vendors — a managed, capped erosion.

Do brands matter? Not consumer brand — but reputation for tool performance and learning-rate enablement is a powerful B2B credential that drives requalification lock-in.

Nature of competition? Capability/cost-of-ownership competition within a narrow, qualified niche; rational, non-overlapping vs. the other WFE majors (Lam/ASML/TEL/AMAT do not compete in process control at scale).

Customers’ switching costs? High — design-in/requalification risk to yield and schedule; portfolio integration; embedded applications engineers. Two-sided lock-in with TSMC/Samsung.

Financial Condition & Balance Sheet

Assets not fully on the balance sheet? Yes — the accumulated defect-library/recipe data asset and the installed-base relationship/learning-rate franchise are the core value and are unrecognized; R&D is expensed.

Off-balance-sheet liabilities? None material flagged; standard operating leases and purchase commitments.

How conservative is the accounting? Reasonably conservative. Revenue recognition is tool-acceptance-based; the GAAP/non-GAAP gap is modest (~3%, mostly SBC + acquired-intangible amortization). SBC is low (~2.2% of revenue). No impairment or channel-stuffing tells.

How CapEx-hungry? Light — capex ~2.7% of revenue ($335M FY2025). KLA designs/integrates/services; it is not a heavy manufacturer.

Capital Allocation & Management

How much FCF, and how is it used? ~$4.0B TTM (31% margin). Target: return >90% of FCF — dividend (17 straight annual increases, just +21% to $9.20/yr) + buybacks (~$11B authorization). Philosophy: “assertive capital allocation … allocate every dollar of cash.”

Significant acquisitions recently? No large recent deal. History: Orbotech ($3.4B, 2019) created the PCB/Specialty segments and seeded advanced packaging; KLA-Tencor merger (1997). Integration record good. Selectively acquisitive (unlike Lam’s purely organic record) — a monitor item.

Buying back shares? Yes — diluted shares fell ~13% (151.6M FY2022 → ~131.4M). Critique: steady-state rather than counter-cyclical; currently buying at the 99.98th valuation percentile.

Issuing shares to insiders? SBC is low (~2.2% of revenue) and net share count falls — buybacks more than offset.

Compensation / incentives? Annual incentive + PRSUs on relative FCF margin, non-GAAP operating-margin dollars, non-GAAP EPS — returns/cash-oriented, not revenue/size. Hedging/pledging prohibited; ownership guidelines. CEO Richard Wallace. Well-aligned; mild gap is no explicit ROIC hurdle.

Motivations of management? Long-tenured team that hit its 2019 and 2022 Investor-Day targets; incentives reward profitable, capital-efficient growth and TSR. Credible operators.

Valuation & Market Data

ADR, MLP, or K-1? No — ordinary US common stock (NASDAQ: KLAC), 1099 reporting.

Dividend policy? ~$9.20/yr (after +21% raise), ~0.43% yield, ~22% payout — 17 consecutive annual increases; token yield, buyback is the primary lever.

How profitable? Best-in-class — see ROIC/ROE/margins above.

Net income vs. cash from operations? Closely tracking — CFO ≈ or > net income across FY2023–FY2025; earnings are cash-backed.

Risks & Downside

What would cause the stock to decline? A WFE down-cycle / AI-capex digestion taking earnings down; multiple compression from the 99.98th percentile (the dominant driver); a sharp China step-down; a disappointing quarter against priced-in perfection.

Risk of catastrophic loss? Low for permanent impairment — investment-grade, FCF-positive through the trough, widest moat in WFE. The realistic severe case is a large drawdown (~−50–60% peak-to-trough in a bear scenario), not insolvency. The genuine catastrophic tail is exogenous (Taiwan/Korea supply-chain disruption).

Chance of total loss? Negligible barring an exogenous geopolitical catastrophe.

Recent News & Events

Has the business environment changed recently? Yes — favorably for the business: AI-capex inflection, geographic rotation to leading-edge Taiwan (China 43%→33%, Taiwan 18%→27%), advanced packaging to ~$1B (+58%), e-beam to $400M, and a new, higher 2030 model. Unfavorably for the stock: the multiple re-rated to an all-time extreme.

Significant acquisitions / accounting changes? None recent. Display business exited March 2024. Routine accounting; no policy changes flagged.

Recent changes — markets, facilities, management? March-2026 Investor Day (new 2030 model, +21% dividend, +$7B buyback, >90% FCF-return target). Capacity/hiring ramp to support the 2027 build-out. Sell-side targets raised (Cantor $2,500, UBS $2,180) — though the mean target (~$1,855) sits below the current price.

APPENDIX B — Source Appendix

KLA Corporation (NASDAQ: KLAC) — Source Appendix

Primary sources first. Access date 2026-06-10 unless noted. Third-party AI/sentiment scores treated as signal, not evidence (validated against primary sources before use).

1. SEC Filings — Primary (EDGAR, CIK 0000319201)

• FY2025 Form 10-K (filed 2025-08-08, klac-20250630) — business, three-segment reporting (Note 18), geographic revenue table, competition, risk factors, customer concentration (TSMC, Samsung each >10%). https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000319201
• FY2021–FY2024 Form 10-Ks (klac-20210630 … klac-20240630) — multi-year revenue/segment/geographic history.
• FY2026 Form 10-Qs (Q1 klac-20250930 2025-10-31; Q2 klac-20251231 2026-01-30; Q3 klac-20260331 2026-04-30) — quarterly revenue, margins, balance sheet (cash, debt, equity).
• DEF 14A proxies 2021–2025 (FY2025 proxy filed 2025-09-23) — executive compensation, incentive metrics (relative FCF margin, non-GAAP operating-margin dollars, non-GAAP EPS), TSR pay-vs-performance, governance.
• Form 4 corpus (insider transactions, trailing window) — routine grants/vesting/10b5-1 sales; no discretionary open-market purchases by named insiders.
• EDGAR XBRL company-facts (EDGAR company-facts API) — revenue (RevenueFromContractWithCustomerExcludingAssessedTax), net income, operating cash flow, capex, buybacks, dividends, diluted shares, R&D, SBC, stockholders’ equity series (FY2021–Q3 FY2026).

2. Earnings Calls, Investor Day & Company Communications — Primary

• Q3 FY2026 (March-quarter) earnings call, 2026-04-29 (KLA IR webcast/transcript) — Q3 results (rev $3.415B, GM 62.2%, OM 42.6%, non-GAAP EPS $9.40), June-quarter guide ($3.575B), CY2026 WFE >$140B / 2027 faster, China commentary, advanced packaging $635M→$1B, service $775M (+16%), pricing discipline, gross-margin/DRAM-cost headwind.
• KLA Analyst/Investor Day, 2026-03-12 (KLA IR webcast/transcript) — 2030 model ($26B revenue ±$2.5B, ~$84 EPS ±$8), WFE ~$215B by 2030, process-control intensity 5.3%→7.4%→9%, share 56.5% / 6.5× nearest competitor, share-of-WFE 7.4%→9%, e-beam $75–100M→$400M, data-analytics SW $250M, +21% dividend (17th straight), +$7B buyback, >90%-of-FCF return target, segment/intensity mechanics (large die, design variety, HBM/memory, advanced packaging).
• KLA conference presentations (BofA, J.P. Morgan, UBS, Goldman, Citi, 2025–2026; company IR) — supplementary forward commentary.
• KLA IR shareholder letters / supplemental materials (kla.com/investors).

3. Quantitative Data Helpers

• Market data — snapshot (sector/GICS, description, revenue TTM $13.1B, margins, ROE/ROA, EV, shares, short interest, analyst data) and own-history valuation percentiles (: P/E ~60.6 / P/B ~48 / P/S ~21.6, all 99.98th percentile, composite 99.98, 2026-06-09). Note: financials are sourced from EDGAR XBRL and the 10-K/10-Q filings.
• yfinance (2026-06-10) — price ~$2,136, market cap ~$279B, EV ~$280.6B, total debt $6.15B, total cash $4.96B, 52-wk $832–$2,304.

4. News & Sentiment (market data, validated against primary filings)

• News (2026-06-10): Cantor Fitzgerald maintains Overweight, raises PT $2,000→$2,500 (2026-06-10); UBS maintains Neutral, raises PT $1,770→$2,180 (2026-06-09). Wall Street mean target ~$1,855 (below current price).

5. Industry & Framework Context

• SEMI equipment-market data (WFE ~$120B 2025 → >$140B 2026; total equipment record 2026) — SEMI press releases 2025/2026.
• Analytical frameworks (Greenwald Competition Demystified: demand-side captivity + economies of scale + intangibles in a narrow relevant market; market-share-stability and persistent-ROIC tests; Marathon Capital Returns: supply-side capital-cycle, asset-growth anomaly — applied to the customer-fab supply side).
• Peer context: publicly disclosed financials and filings of Lam Research (LRCX), Applied Materials (AMAT), and ASML — used for industry framing and comparative multiples; KLA analyzed independently from its own primary filings.

6. Key Figures Quick-Reference (all reconciled to EDGAR unless noted)