·Dashboard·Research·Work·Archive
/
← wiki
ticker stocksemicap-equipment updated 2026-05-30

ONTO — Onto Innovation Inc.

Thesis

Onto Innovation is the highest-quality pure-play on advanced packaging process control: the #3 semiconductor process control equipment company globally (after KLA and Applied Materials), with a defensible niche in bump/packaging inspection (Dragonfly) and OCD metrology (Atlas) at the exact inflection point where AI infrastructure is forcing the industry's most complex packaging ramps in history. The thesis rests on AI infrastructure driving sustained demand for advanced packaging (CoWoS, HBM, EMIB/Foveros), where Onto's Dragonfly family is the inspection tool of choice for bump metrology and packaging QC. This is a capital-efficient, no-debt, $300M+ FCF generator priced at ~46x forward earnings — demanding, but not unreasonable for a semicap business with genuine defensible share in the fastest-growing segment of the supply chain.

Verdict / stance: SCALE BUY — initiate a starter position (30-40% of target size), scale in on pullback or Q1 2026 earnings confirmation. All five FundamentEdge gates pass. Management governance is clean with a B+ grade and no red flags. Conviction: Medium-High — the thesis is clear and well-anchored, but the valuation is demanding and the Q3 2025 step-down in margins raises questions about Semilab integration execution. Thesis-status across prior fragments was 'watching'; the Apr 2026 swarm output upgraded to a scale-in buy. Expected holding period: 18-36 months (through the AI packaging capex up-cycle; reassess when HBM4 ramp plateaus or KLA launches a credible Dragonfly equivalent).

What has to be true (single most important thing): advanced packaging revenue grows >30% in FY2026 as guided, confirming that the Q3 2025 step-down was transitory rather than structural. Supporting legs: (1) the SK Hynix $240M+ volume agreement through 2027 anchors near-term visibility; (2) Semilab adds $100-120M of incremental, accretive FY2026 revenue; (3) Atlas G6 provides a specific window of OCD metrology share gain at GAA/2nm nodes before KLA closes the gap.

Bull case: AI packaging super-cycle is structural, not cyclical. Every AI accelerator chip requires advanced packaging (CoWoS, HBM stacking, chiplet integration). SK Hynix's $240M commitment validates this. Semilab adds materials characterization at accretive margins. Atlas G6 positions Onto for the 2nm transition. Backlog doubled in the three months prior to Q4 2025 earnings to ~2 quarters of visibility.

Bear case: KLA is encroaching on Onto's packaging niche with superior analytics and scale. Onto's OCD market share has slipped (Paragon Intel and Dr. Castellano flagged 'strategic drift'). Valuation is demanding at ~46x forward earnings, leaving no room for execution miss. Semiconductor capex is cyclical and AI capex could moderate. The Q3 2025 margin/revenue step-down may not be fully transitory.

Snapshot

One-liner: Onto Innovation makes the optical/x-ray/acoustic measurement instruments that chip fabs use to control yield — the eyes of the fab — with a specialized, defensible position in advanced packaging inspection and OCD metrology for leading-edge logic and memory.

Ticker / Exchange: ONTO / NYSE. GICS: Information Technology / Semiconductor Equipment. HQ: Wilmington, Massachusetts. Founded: October 25, 2019 (merger of Rudolph Technologies + Nanometrics; Rudolph traces to 1940). CIK 704532.

Valuation snapshot (as of Apr 24, 2026):

  • Stock price: $307.86
  • Market cap: ~$15.3B (one fragment: $15.31B)
  • Enterprise value: ~$14.69B (net cash reduces EV vs. market cap)
  • P/E (TTM): 110.7x — heavily distorted by FY2025 GAAP margin compression from Semilab purchase accounting
  • Forward P/E (FY2026E): ~46x on consensus EPS $6.65
  • EV/EBITDA (TTM): 53.8x
  • P/FCF (TTM): 51.1x
  • EV/Revenue (TTM): 14.6x | P/Sales: 15.2x
  • FCF yield: ~2.0% (FY2025 FCF $300M / mkt cap $15.3B)
  • PEG: 1.45 | Beta (5Y): 1.47
  • Dividend: None
  • 52-week range: $85.88 – $316.00 (stock up ~135% over the trailing year; deep-dive cites +156%)

Key stats: Revenue $1.005B (FY2025). FCF $300M (FY2025). No debt; net cash ~$640M (one fragment: ~$622M LTM). Shares ~49M diluted, flat for 5 years. ~1,593 employees (pre-Semilab). Serves 240+ customers in 24+ countries.

Discrepancy flag: Net cash is stated as ~$640M (FY2025) in ONTO.md, profile, and checklist, but ~$622M (LTM) in the deep-dive valuation/balance-sheet tables; EV is ~$14.69B (deep-dive) vs ~$14.7B (profile). The ~$640M is the FY2025 year-end figure; ~$622M is a later LTM read. Both retained.

Business

What it does: Designs, manufactures, and supports process control equipment used by semiconductor manufacturers to detect defects, measure film thickness, verify dimensional accuracy and overlay at every stage of chip fabrication and advanced packaging — the feedback loop that controls yield. Systems sit inline on the production floor: wafers or packaged substrates go in, the tool measures something that cannot be measured any other way (optical, x-ray, or acoustic), and a yield-control signal goes back to the fab. The equipment is not optional; it is a bottleneck gate in the production flow. Without process control, advanced-node and AI-packaging manufacturing is economically impossible.

Segments / business lines: Onto operates as a single reportable segment but tracks revenue by three end-market categories:

  • Advanced Packaging (~35-40% of revenue): Macro-inspection, 3D bump metrology, wafer-level packaging process control; Dragonfly family. The AI packaging story lives here.
  • Advanced Nodes (~35-40% of revenue): OCD (Optical Critical Dimension) and film metrology for leading-edge logic and memory; Atlas family. The GAA/2nm transition story lives here.
  • Specialty Devices & Other / Service (~20-30% of revenue): Compound semi (SiC/GaN), power devices, older nodes, plus aftermarket services/spares. Semilab product lines (FAaST/CnCV/MBIR) slot in here. Note: Onto does not break out segment revenue with full precision; the 48% cited for 'Specialty Devices + Advanced Packaging' in Q1 2025 is a combined figure.

Business model: Capital equipment (one-time sale per system, ~$1-3M+ ASP per tool, up to $5M for some OCD units), supplemented by aftermarket services/spares. Revenue is lumpy and order-driven; grows with fab capex cycles. No SaaS or recurring subscription line, though the installed base drives a growing service tail.

  • Equipment sales: ~75-80% of revenue
  • Services and spares: ~20-25%, recurring and less cyclical than equipment
  • Non-GAAP gross margin: ~54-55% (FY2025); GAAP lower due to amortization of acquired intangibles

Product portfolio

  • Atlas (OCD metrology): Measures gate heights, feature widths, film thicknesses, sidewall profiles via spectroscopic ellipsometry / OCD. Generations: Atlas III+ (FinFET), Atlas V (current HVM tool — FinFET, GAA logic, DRAM, 3D NAND), Atlas G6 (launched April 2025; next-gen for 2nm/1.6nm advanced logic and AI device process control). Competitive edge is the Ai Diffract and SpectraProbe electromagnetic-modeling software that extends spectral ellipsometry to 3D nanosheet (GAA) structures — KLA's spectral ellipsometry tools reportedly cannot handle the 3D modeling complexity for gate-all-around, giving Onto a window at 3nm/2nm.
  • Dragonfly (macro inspection + packaging metrology): Optical imaging (not e-beam) to scan large wafer/panel areas and flag anomalies; AI-assisted defect classification; 3D capability adds bump-height measurement. Dragonfly G3 is the current volume tool (the SK Hynix $240M agreement covers G3 for HBM3/HBM4 die-stack inspection; combines 2D defect inspection with 3D bump-height measurement). Dragonfly G5 launched Q1 2026 (early 2026): sub-150nm defect sensitivity, 5x throughput improvement.
  • NSX-330: 2D defect inspection + metrology for wafer-fab process control.
  • PULSE: Acoustic metrology for hybrid bonding, multilayer film measurement, thin-film Cu recess.
  • Semilab lines (acquired Nov 2025): FAaST (inline wafer contamination monitoring — rapid fluorescence/spectroscopic detection of metal contamination at ppb level); CnCV (materials characterization — composition/thickness of thin films, dopant profiling); MBIR (surface charge metrology — interface trap density and oxide charge; important for SiC/GaN). Together they let Onto offer a 'measure everything' suite — dimensional (Atlas) + defect (Dragonfly) + materials/electrical (Semilab) — that no single competitor can match.

Customers

Onto does not disclose customer names in SEC filings (reports concentration by tier; serves '240+ customers in 24+ countries'). Likely majors, from earnings calls / channel checks:

  • TSMC (TSM): ~25-30% est. — Atlas for GAA/OCD (co-qualification for 3nm/2nm), Dragonfly for advanced packaging.
  • SK Hynix (000660.KS): ~15-20% est. — $240M+ Dragonfly G3 volume agreement through 2027, $60M specifically for 3D bump metrology / HBM4 ramp.
  • Samsung Electronics (005930.KS): ~10-15% est. — Atlas for GAA; Dragonfly for packaging; hybrid bonding evals ongoing.
  • Intel (INTC): ~5-10% est. — EMIB bump inspection, Foveros hybrid bonding metrology.
  • Micron (MU): ~5% est. — DRAM/NAND, HBM exposure. No single named customer definitively disclosed >20%. Taiwan (~31% of revenue) likely corresponds substantially to TSMC. If TSMC cut orders 50%, it would cost ~$125-150M (~12-15% of revenue) — painful but not existential given Semilab and >30% packaging growth offsets.

Moat / competitive position

Onto is the #3 player in process control globally (after KLA and Applied Materials), with single-digit share in some sub-segments (non-metal thin-film OCD) but strong share in advanced packaging inspection (Dragonfly) and ~20-25% in OCD metrology with share gains at GAA nodes. Moat sources:

  • Switching costs (strongest): Process recipes are calibrated to specific tools; changing suppliers mid-generation requires 6-18 months of re-characterization. This locks in placements once installed.
  • IP / software: Ai Diffract 3D OCD modeling (genuine differentiation vs KLA at GAA, but time-limited as GAA recipes standardize); PULSE acoustic metrology; Dragonfly macro-inspection architecture.
  • Packaging specialization: Dragonfly's first-mover position in back-end packaging inspection — more durable than the OCD edge because the field is less crowded; 5-7 year head start.
  • Semilab materials characterization: Unique 'one-stop' wafer-quality monitoring (electrical + optical + materials) not previously available from any single competitor.

Competitive set: KLA (KLAC) — dominant ~50%+ process control share, ~4-5x Onto's size, ~$10.6B FY2025 revenue, the encroachment threat; Applied Materials (AMAT) — OCD via Centura/VeritySEM, fab-bundling breadth; Nova Measuring (NVMI) — secondary OCD, XPS/optical combo; Camtek (CAMT) — APAC packaging inspection at lower price points, growing ~30%+, the direct Dragonfly-turf threat; Rigaku Holdings (268A.TSE) — x-ray metrology, co-development signal. Porter snapshot: rivalry High; buyer power High (TSMC/Samsung/SK Hynix run competitive bake-offs); supplier power Low-moderate; threat of new entrants Low (18-36 month qualification cycles, PhD-intensive modeling software); threat of substitutes Low (process control is not optional).

Operations footprint

Asset-light relative to peers; outsources much mechanical assembly. Capex ~$20-32M/year (~3% of revenue). Key sites: Wilmington MA (HQ, R&D), Milpitas CA (thin-film/OCD R&D, legacy Nanometrics), Boise ID (mfg, legacy August Technology), Lehi UT (eng/mfg), Semilab USA (acquired Nov 2025), plus field service in Taiwan, South Korea, Japan, Singapore, Malaysia, Vietnam, China (sharply reduced), Europe.

Value-chain position: Onto sits at the process-control layer — not the 'main' fab tool (ASML/Lam/Applied/TEL) but the measurement/control layer wrapping every process step. Process-control TAM ~$7-12B; KLA captures the largest share (~$8B); Onto at ~$1B implies ~8-12% of TAM. Upstream optics/laser suppliers include Coherent/II-VI (COHR, ~$14B cap) and Rigaku for x-ray sources — no single-source critical bottleneck identified.

Financials

Revenue trajectory (GAAP): FY2022 $1,005M (+27%) → FY2023 $816M (-19%, cycle trough) → FY2024 $987M (+21%) → FY2025 $1,005M (+1.8%, just crossed $1B) → FY2026E $1,270M (+26% consensus). The 3-year CAGR FY2023-FY2026E is ~16% — solid for a capital-equipment company.

Income statement:

Metric FY2022 FY2023 FY2024 FY2025 FY2026E
Revenue $1,005M $816M $987M $1,005M $1,270M
Revenue growth +27% -19% +21% +1.8% +26%
Gross profit $539M $420M $515M $500M ~$690ME
Gross margin (GAAP) 53.6% 51.5% 52.2% 49.7% ~54%E
EBIT $237M $116M $187M $133M ~$290ME
EBIT margin 23.6% 14.2% 18.9% 13.2% ~23%E
Net income $223M $121M $202M $137M ~$325ME
Net margin 22.2% 14.8% 20.4% 13.6% ~26%E
EPS (diluted) $4.49 $2.46 $4.06 $2.78 $6.65E

Margins: Non-GAAP gross margin ~54-55% (consistent); GAAP 49.7% in FY2025, compressed by Semilab purchase-accounting amortization, expected to recover toward 54%+. Q4 2025 non-GAAP gross margin 54.6%, non-GAAP operating margin 25.2%. The FY2024 EBIT margin 18.9% → FY2025 13.2% step-down is a one-time purchase-accounting effect, not structural erosion.

Cash flow & balance sheet:

Metric FY2022 FY2023 FY2024 FY2025
Operating cash flow $137M $172M $246M $328M
Capex ($18M) ($23M) ($32M) ($29M)
Free cash flow $118M $149M $214M $300M
FCF margin 11.8% 18.3% 21.7% 29.8%
Cash & equivalents $176M $234M $213M $346M
Net cash (no debt) $548M $698M $852M $640M

FCF compounded from $118M (FY2022) to a record $300M (FY2025), ~26% annually — exceptional capital efficiency; 29.8% FCF margin is outstanding for capital equipment. Net cash declined from $852M (FY2024) to $640M (FY2025) — the $495M Semilab acquisition funded entirely from cash, offset by $300M FCF; the deep-dive shows a later LTM net cash of ~$622M (discrepancy retained). No long-term debt, no dividend, no ATM, no convertibles/warrants.

ROIC vs WACC: ROIC ~14% (FY2025 TTM, temporarily compressed by Semilab goodwill/intangibles entering the capital base; FY2024 peak ~18%; FY2022 ~20%+). WACC ~10-11% (no debt; beta 1.47; cost of equity via CAPM). Value-creating spread ~3-4pp (compressed from ~8-9pp at the FY2022 peak). FY2026E ROIC ~22% if Semilab revenue ramps and the intangibles base is more fully utilized.

Quarterly revenue and second-derivative

Quarter Revenue QoQ % YoY % 2nd derivative (YoY ∆)
Q1 2024 $228.9M — +15.7% —
Q2 2024 $242.3M +5.9% +21.0% +5.3pp
Q3 2024 $252.2M +4.1% +21.0% 0pp
Q4 2024 $263.9M +4.6% +25.5% +4.5pp
Q1 2025 $266.6M +1.0% +16.4% -9.1pp
Q2 2025 $253.6M -4.9% +4.7% -16.3pp
Q3 2025 $218.2M -13.9% -13.5% -34.5pp (trough)
Q4 2025 $266.9M +22.3% +1.1% +14.6pp (reversing up)
Q1 2026E ~$280M +4.9%E +5.0%E accelerating
Q2 2026E $300M+/$310M+ +5-10%E +22%+E accelerating

Note: Q1 2025 revenue appears as $266.6M (deep-dive/checklist quarterly tables) and the prior canonical noted it; the second derivative turned sharply negative in Q2-Q3 2025 and reversed positive in Q4 2025. The Q3 2025 trough was driven by Semilab integration digestion, customer ordering between Dragonfly G3 and G5 generations, and macro uncertainty. At Q2 2026 guidance of $300M+, annualized run rate = ~$1.2B+, consistent with the $1.27B FY2026 consensus.

Incremental margins (YoY)

2024 incremental gross margins ran 62-74% (strong operating leverage; every incremental revenue dollar drove ~$0.65-0.74 gross profit); incremental EBIT margins ~28-42%. 2025 incrementals deteriorated sharply Q2-Q4 (Semilab purchase-accounting amortization depressed reported GAAP); the Q3 2025 -$34M YoY revenue decline with -$25.8M gross-profit decline implies ~76% incremental GM on the downside (high deleverage, expected for a software-heavy fixed-cost structure). FY2026 restoration to the 60-70% gross / 35-45% EBIT range is the expectation if revenue accelerates +26% as guided.

R&D: FY2024 R&D expense $116.8M (~11.8% of revenue); ~370 of ~1,551 employees in R&D (~24% of workforce). SBC ~$60-70M (~6-7% of revenue, normal range).

Geographic revenue mix (FY2024): Taiwan 31.2%, South Korea 28.9%, China 11.8%, United States 10.5%, Southeast Asia 6.6%, Japan 5.8%, Europe 5.2%. Taiwan + Korea ≈ 60%. China was deliberately reduced to <3% of FY2025 revenue (the deep-dive narrative rounds the FY2024 figure to '12%'; ONTO.md's headline '<3%' refers to FY2025). Financial-health verdict across fragments: Strong / clean — no declining cash flow, no rising leverage, no aggressive accounting; the non-GAAP vs GAAP distinction is transparent, not deceptive.

Industry landscape

Onto sits in the consolidated process control (metrology + inspection) niche of semiconductor capital equipment — the measurement/control layer wrapping every process step. KLA holds ~50%+ share (~$10.6B FY2025 revenue); Applied Materials, Nova, Onto, and Camtek share most of the rest. Process-control TAM is ~$7-12B (Onto cites a $12.3B 'AI Era' TAM vs a $4.2B prior 'Mobility Era' TAM); the advanced-packaging subset (~$2-3B growing to $5B+ by 2030) compounds faster (~15-20% CAGR) than front-end capex. Barriers to entry are extremely high (PhD-intensive electromagnetic modeling, sub-nm optical hardware, 18-36 month fab qualification, global service network). The sector is both cyclical (fab capex cycles) and secular (rising chip complexity adds process steps); the AI infrastructure cycle is creating a secular demand shift that partially decouples from the traditional semiconductor cycle. Current cycle position: coming out of the FY2023 trough (-19%), with an H2 2025 air-pocket (Q3 2025 -13.5% YoY) and Q4 2025 re-acceleration to a record $267M with backlog doubling. See sector page: semicap-equipment

Management

CEO — Michael P. Plisinski (President & CEO): Age ~55. CEO since 2015 (Rudolph Technologies); led Onto since the merger inception October 2019 — the longest-serving CEO in the process-control metrology space (~11-year tenure). B.S. Computer Science, University of Massachusetts; Advanced Management Program, Harvard Business School. Career: founded Counterpoint Solutions (1999-2003, sold to August Technology 2003 for ~$1.45M); VP Engineering then VP/GM Data Analysis & Review at August Technology and Rudolph; EVP/COO Rudolph (2014-2015); CEO Rudolph (2015-2019). >25 years in semicap; a genuine domain expert, not a parachuted-in serial CEO. Orchestrated the 2019 Rudolph/Nanometrics merger ('merger for survival' vs KLA dominance); revenue has since grown from ~$450M to $1B+. Clean regulatory record — no SEC enforcement, bankruptcies, or fiduciary-duty/fraud judgments. Paragon Intel published a CEO analysis flagging OCD market-share loss to KLA and customer-concentration concerns (sell-side skepticism, not wrongdoing).

CFO — Brian K. Roberts: Appointed June 16, 2025 (replacing Mark Sefcik). 20 years CFO experience (11 at public companies); most recently CFO of Sensata Technologies (NYSE: ST, ~$4B revenue). Credible hire for the larger post-Semilab business; no adverse history.

SVP Customer Success — Shirley Chen: Appointed June 16, 2025. 25 years semiconductor experience; 18 years progressive sales leadership at KLA (KLAC) and Thermo-Fisher. The KLA background brings direct competitive intelligence and customer relationships — signals intent to pursue share at leading-edge customers.

SVP & General Counsel — Yoon Ah Oh: Multi-year tenure; no adverse history.

Note: CFO and SVP Customer Success were both new appointments in June 2025 — a partial refreshing of the C-suite as the company crossed $1B in revenue and absorbed Semilab.

Insider ownership & skin in the game

Aggregate insider ownership ~0.86% (very low, but typical for a post-merger mid-large cap where transaction equity went to legacy Rudolph/Nanometrics shareholders). Plisinski holds ~150,492 shares (~0.31%, ~$46M at $307-308). CEO insider selling: sold ~85,000 shares in Jan-Feb 2025 for ~$18M total (Jan 16: 30,000 @ ~$211.24 = $6.3M; Jan 21: 35,000 @ ~$219.19 = $7.7M; plus additional 10b5-1 transactions) — an ~18.87% reduction in his position, executed under pre-established 10b5-1 plans, at prices far below the current $308 (he diversified near the cycle bottom). Open-market purchases in last 12 months: zero — the key yellow flag (insider conviction signal absent). PSU vesting (March 2026): CEO 5,971 PSUs granted Mar 2024 settled into 4,597 shares (77% of target — FY2025 performance slightly below target); CFO 3,096 shares + PSU grants; SVP GC 724 shares.

Capital allocation

  • Rudolph + Nanometrics merger (2019): ~$700M+ combined EV; revenue grew ~$450M → $1B+, stock 10x+ from merger lows. Grade: A.
  • Inspectrology (bolt-on, date unclear): Added overlay metrology; cleanly integrated. Grade: B+.
  • Semilab product lines (Nov 17, 2025): ~$495M cash (later slightly revised; EIR business excluded from final scope). FY2026 contribution guided $100-120M revenue, H2-weighted, accretive per management. ~4x revenue multiple — slightly expensive but not egregious for a strategic capability extension. Grade: B (pending FY2026 execution).
  • Buybacks: $100M authorization (Nov 2020); modest usage; share count essentially flat at ~49M for 5 years. Notably did NOT buy back stock at FY2023 lows (~$100-130, ~26x trough P/E) — a missed opportunity, though cash was conserved for Semilab.
  • Equity issuance: None in 3+ years; no dilution. Overall capital-allocation grade: B+.

Compensation & alignment

CEO total target comp ~$7M (est.; ~89.5% at-risk / 10.5% salary, base ~$730K est.). Annual cash MBO tied to revenue + non-GAAP operating income; long-term equity = time-based RSUs + relative-TSR-vs-SOX PSUs. PSU history: FY2022 tranche vested 200% (3-year TSR 121%, 93rd percentile SOX); FY2023 tranche vested 200% (2-year TSR 146%, 90th percentile SOX); FY2024 tranche settled 77% of target (Mar 2026, below median — soft FY2025). Double-trigger change-of-control (both a CoC event AND qualifying termination required). Anti-hedging and anti-pledging confirmed in the 2026 proxy. Say-on-pay 96.4% support (2025 meeting). No spring-loading evidence. Key misalignment flag: management comp is tied primarily to non-GAAP operating income, which excludes Semilab purchase-accounting amortization — so the GAAP margin compression depressing reported FY2025 earnings doesn't hurt their comp. This is a disclosure-transparency issue, not a governance red flag.

Governance

Board of 7 members, ~86% independent (one fragment phrases this as 86% independent):

  • Christopher A. Seams — Independent Chair (M&A Committee Chair, Nom/Gov). 30+ years semi; ex-CEO Deca Technologies (a packaging company — directly relevant); EVP Cypress; AMD/Philips prior. NACD Certified. Chair since the 2019 merger. (One fragment's holdings table abbreviates as 'Seams (Board Chair)'.)
  • Michael P. Plisinski — CEO/Director (not independent).
  • Stephen D. Kelley — Independent (Audit, Compensation). Current CEO of Advanced Energy Industries (AEIS); prior CEO Amkor Technology (2013-2020). Dual-employer situation noted (AEIS sells power-delivery systems to the same customer base) — not a red flag, but monitored; abstains from conflicting discussions.
  • Susan D. Lynch — Independent (Audit Chair); financial leadership at software/hardware tech firms; added March 2024.
  • David B. Miller — Independent (Nom/Gov Chair).
  • Stephen S. Schwartz — Independent (Audit, M&A); current CEO of Azenta (formerly Brooks Automation); dual-employer noted, no direct overlap; added July 2024.
  • May Su — Independent (Compensation Chair). No dual-class shares, no poison pill, no staggered board (annual elections), majority voting, independent chair separate from CEO. Clean 2026 proxy: 'no related person transactions since the beginning of fiscal year 2025.' Standard corporate structure (Delaware parent + international service subsidiaries + Semilab USA LLC); no shell entities, no asset migration, no related-party self-dealing.

Credibility / follow-through

~80% follow-through rate (8 of 10 tracked commitments delivered or on-trajectory). Guidance tendency: Conservative / Straight Shooter — consistent revenue beats through 2024; FY2025 mixed (Q2 EPS $1.25 slight miss; Q3 2025 revenue $218.2M slightly below guide midpoint though they had flagged it as 'the lowest revenue point for the year'; Q3 EPS $0.92 beat; Q4 2025 non-GAAP EPS $1.26 vs ~$1.27 est., slight miss). Low weasel-language frequency; never said 'no plans to raise capital' then raised. The one communication gap: the Q2-Q3 2025 Semilab purchase-accounting margin impact was not clearly pre-communicated before Q2 results (GAAP-watchers caught off-guard).

Ownership

Institutional ~93-95% of outstanding shares. Top holders (13F): Vanguard 5,576,305 (~11.3%), William Blair Inv. Mgmt 1,726,587 (~3.5%, the most notable active conviction holder), Wellington 1,218,548 (~2.5%), Geode 1,127,532 (~2.3%, +45% recent), Paradigm Capital 1,125,838 (~2.3%), Fisher Asset Mgmt 701,999 (~1.4%), T. Rowe Price 680,266 (~1.4%), AQR 615,651 (~1.2%, +315% recent — quant factor signal), D.E. Shaw 517,189 (~1.0%, -48% reduction), Millennium 495,000 (~1.0%). No activist 13D filers. Short interest ~9.0% of float; 2.88 days to cover (Dec 2025); below peer-group average ~9.01%; modest.

Overall management grade: B+. Competent, experienced, domain-credible management with clean governance and no red flags. The main yellow flag is low insider ownership plus systematic CEO selling (no open-market buys). For Pink's purposes: a 'trust and deploy' team — you are betting on domain expertise and strategic execution, not founder-level personal conviction. Semilab is the thesis-defining bet for FY2026-2027; Plisinski's reputation is staked on it.

Catalysts & risks

Catalysts (bull)

  • SK Hynix $240M+ volume agreement through 2027 — Dragonfly G3 for HBM3/HBM4 die-stack inspection; $60M specifically for 3D bump metrology / HBM4 ramp; ~$80-100M/year implied run-rate over the remaining term. The clearest forward revenue signal in the business.
  • Semilab integration contributing $100-120M revenue in FY2026 (H2-weighted), purely incremental and accretive.
  • Atlas G6 (launched April 2025) OCD platform for 2nm/1.6nm/1.4nm and AI device process control — a specific window of OCD share gain at GAA nodes before KLA closes the gap.
  • Dragonfly G5 (launched Q1 2026) — sub-150nm defect sensitivity, 5x throughput; initial production shipments / revenue recognition.
  • Four active hybrid-bonding next-gen inspection system evaluations (as of Feb 2026) with 'significant improvement' performance feedback from customers.
  • Rigaku co-development on a hybrid x-ray/optical metrology engine — flagged April 2026 (Collyer Bridge channel check), likely 3D DRAM application; not yet formally disclosed by Onto. Watch the Q1 2026 earnings call. Potential entirely new product category with no direct competition. See 268A for Rigaku profile.
  • Advanced packaging revenue guidance >30% growth in FY2026.
  • Q2 2026 revenue guidance >$300M (acceleration from $266.9M in Q4 2025) — first time quarterly revenue exceeds $300M would confirm re-acceleration.
  • GAA / 2nm transition (Atlas G6 cycle ramp); AI datacenter buildout and CoWoS/HBM packaging spend acceleration.
  • Intel Foveros/EMIB scale-up — Intel projected ~$4B advanced-packaging capex (highest among foundries; deep-dive cites TSMC ~$3.6B, Samsung ~$2B).
  • Estimate revisions upward: Stifel upgraded from Hold to Buy/Strong Buy on April 17, 2026, raising PT from $220 to $350; Jefferies raised PT to $325 (from $300); Evercore ISI also positive. Consensus shifted to 5 Strong Buy + 3 Buy, 0 Hold, 0 Sell.
  • Medium-term (1-3 yr): HBM4 ramp (2026-2027, the single largest identified revenue driver); 3D DRAM adoption (TSMC 3D DRAM, SK Hynix HBM5+).

Risks (bear)

  • Valuation execution dependency (High impact if missed): At ~46x forward P/E there is near-zero room for a miss. The market is pricing near-perfection. Base-bear math: if FY2026 EPS of $6.65 comes in at $5.50 (-17%) and the multiple compresses to 35x (KLA-level), the stock trades ~$193 (a ~35% drawdown from $300). 20% EPS miss ($5.32) at 35x → ~$186 (~37-40% drawdown). 30% miss ($4.65) at 28x (Camtek-like) → ~$130 (~57% drawdown, extreme bear).
  • KLA competitive displacement (Medium): KLA could develop a Dragonfly equivalent in 3-5 years with $1B+ investment; any KLA product announcement targeting bump/packaging inspection at TSMC or SK Hynix is a thesis-watch event. Onto's OCD share has slipped (Paragon Intel; Dr. Castellano 'strategic drift').
  • Camtek acceleration (Medium): CAMT growing ~30%+ at lower price points in APAC; pressures Onto pricing at cost-sensitive OSATs.
  • Customer / geographic concentration (Medium): Taiwan + Korea ~60% of revenue; structural given semicap geography.
  • Semilab integration underperformance (Medium): ~$495M paid for $100-120M projected FY2026 revenue (~4-7x revenue at the low end). If revenue comes in at $70-80M (slow H2-weighted ramp), the multiple looks expensive; cross-selling FAaST/CnCV/MBIR into Onto's base is unproven.
  • Q3 2025 deterioration not fully transitory (Low-Medium): If Q1 2026 reveals advanced packaging is softer than guidance, the thesis needs re-evaluation — the >30% packaging guidance may bake in already-contracted SK Hynix volume, so any incremental customer softness is incremental miss.
  • Intel execution risk (Medium): Intel <10% of revenue; not single-point-of-failure, but a stall in Intel's EMIB/Foveros ramp removes a growth customer.
  • GAAP margin never recovers from Semilab amortization to the level consensus expects.
  • China export-control escalation (Low, largely resolved): China <3% of FY2025 revenue; no strategic reliance. Longer-term TAM risk if Chinese alternatives (Naura, AMEC) qualify domestically.
  • AI-enabled process-control disruption (10+ year, not near-term): virtual metrology could theoretically reduce hardware-metrology demand.
  • CEO key-person risk (Low-Medium): 11-year tenure, no public succession plan; some depth below via Roberts/Chen.
  • Dilution risk: Low. Shares flat at ~49M for 5 years; no ATM, shelf, convertibles, or warrants; $300M+ FCF self-funds; Semilab was cash-funded.

Near-term event risk

  • Q1 2026 earnings (est. May 2026): the highest-probability catalyst and primary event risk; a miss vs the $275-285M guidance would be severely punishing given the valuation. This is the explicit confirmation gate for the thesis.
  • SK Hynix HBM order cadence: any disclosure of reduced HBM capex pressures the thesis directly.
  • KLA earnings / product announcements (quarterly): any mention of a Dragonfly competitor.

What would make the thesis wrong

(1) KLA announces a macro-inspection tool directly competing with Dragonfly at a major AI packaging customer; (2) AI capex cycle moderates (hyperscalers slow datacenter build-out, broad semicap capex cuts >20%); (3) Semilab FY2026 revenue below $80M; (4) Intel's advanced-packaging business stalls due to broader restructuring; (5) GAAP gross margin never recovers above 50%.

Valuation / DCF

Verdict: not cheap; priced for execution. The thesis is emphatically NOT 'the P/E is low' — at ~46x forward earnings and ~15x EV/Revenue, ONTO is expensive on near-term multiples. The thesis is growth + structural demand + competitive position + Semilab accretion; valuation is a constraint on upside and a source of downside risk.

Current multiples (at ~$300-308):

Metric ONTO KLA (KLAC) Nova (NVMI) Camtek (CAMT) Applied (AMAT)
Forward P/E ~46x ~30x ~35x ~28x ~22x
EV/EBITDA (TTM) 53.8x ~22x ~28x ~20x ~18x
P/FCF 51x ~28x ~32x ~25x —
EV/Revenue (TTM) 14.6x ~10x ~10x ~7x ~6x
FCF yield 2.0% 3.5% 2.8% 4.0% 4.0%
Revenue growth (NTM) +26%E +12-15% +15-20% +25-30% +8-12%

ONTO trades at a premium to all peers on every multiple — ~53% premium to KLA on forward P/E. Partially justified by higher NTM growth (26% vs ~13%), a purer play on AI packaging vs KLA's diversified portfolio, and small/mid-cap re-rating potential.

Implied expectations: At $300-308 and $6.65 forward EPS, the market prices ~45-46x forward earnings for 26% revenue growth and ~139% EPS growth. PEG ~1.45 — not cheap but not egregious for the growth profile. The EPS re-rating from $2.78 (FY2025 GAAP) to $6.65 (FY2026E) is largely normalization from depressed FY2025 GAAP levels (Semilab amortization normalizes over 3-7 years); the revenue-growth assumption is more conservative. 5-year historical P/E range: ~15x (trough 2023) to ~55x (prior peak); ~46x forward is toward the high end but not bubble territory.

DCF sensibilities: $300M FCF base (FY2025), growing 20% for 3 years, then 10% for 3 years, then 5% terminal, discounted at 10% → ~$280-320 intrinsic range (the deep-dive frames the same exercise as 20% for 3 years decelerating to a 10% terminal → $280-320). The current price ~$300 is roughly at intrinsic value on conservative assumptions; upside comes from above-consensus execution (Semilab outperformance, Atlas G6 wins, HBM4 acceleration). If advanced packaging grows >30% AND Semilab tracks toward $120M+, FY2026 revenue could exceed $1.3-1.4B and the stock could be worth $350-400+.

Analyst targets: 8 analysts — 5 Strong Buy, 3 Buy, 0 Hold, 0 Sell. Average PT $290-317 (range by source). High target $350 (Stifel, Apr 2026); low target $160. FY2026E revenue consensus $1.27B; EPS $6.65. 100% buy-side consensus is unusual (small 8-analyst sample); the stock at $308 trades roughly at or slightly above the average target.

Scenarios:

  • Bull: $6.65 EPS × 53x (current multiple) = ~$352; or a $7.50E FY2026 beat × 48x = ~$360; target exit band $350-420 (≈53-63x forward FY2026E EPS, or ~30-35x normalized FY2027E EPS).
  • Base: ~$280-320 intrinsic; ~at current price.
  • Bear: 20% EPS miss + 35x → ~$186; 30% miss + 28x → ~$130.

Margin of safety: the Semilab incremental revenue ($100-120M, high-confidence due to contracted pipeline) and the SK Hynix $240M agreement give downside support even if AI packaging slows modestly.

Decision log

  • 2026-02-28 — Canonical page created; thesis-status set to watching. Quick take framed ONTO as the #3 process-control player with an AI advanced-packaging niche; SK Hynix $240M agreement and Semilab ($495M, Nov 2025) as anchors. (updated 2026-05-15)
  • 2026-04-26 — Intel supply-chain research swarm ran four skills the same day: /profile, /deep-dive, /mgmt-dd, /checklist. Price reference ~$292-308 (Apr 24-26).
    • /deep-dive verdict: Medium-High conviction; thesis clear and well-anchored but valuation demanding and Q3 2025 step-down raises Semilab-integration questions. No target price set (new initiation).
    • /mgmt-dd verdict: Overall management grade B+ — competent, domain-credible, clean governance, no red flags; yellow flags = low insider ownership (~0.86%) and systematic CEO 10b5-1 selling (~$18M, no open-market buys). 'Trust and deploy' team.
    • /checklist verdict (pre-buy): SCALE BUY — initiate a starter position (30-40% of target size), scale in on pullback to $260-275 or on Q1 2026 earnings confirmation above $275M revenue. All 5 FundamentEdge gates PASS (revenue-growth primacy; second derivative recovery; valuation-is-not-the-thesis; quality genuine; estimate revisions upward). Scorecard: thesis clear Yes; understand business Yes; mgmt incentives Partially (non-GAAP vs GAAP tension); financials healthy Yes; valuation Stretched; behavioral traps None clear (recency + confirmation bias noted); technicals Wait/scale (overbought near 52-wk highs, RSI 59-75, +135% in 52 wks); position sized Yes; exit plan Yes.
    • Position plan: target 3-5% of portfolio within the AI-packaging/Intel supply-chain basket (alongside 8027, 6809, AMKR, AWX, S71). Tranche 1: 30-40% now at ~$290-300; Tranche 2: 30-40% on pullback to $260-275; Tranche 3: 20-40% post-Q1 2026 confirmation. Add if Q2 2026 guide >$310M or Semilab tracking >$100-110M. Trim if >$360-380 without earnings confirmation or position >7%. Max acceptable loss ~25-30% (~$200-220); below ~$190, full thesis re-evaluation.
  • 2026-04-26 — Stop-loss / re-evaluation triggers set: Q1 2026 revenue misses $275M by >5%; Semilab FY2026 revenue tracking below $80M; KLA announces a direct Dragonfly competitor; FY2026 GAAP gross margin fails to recover above 50%; AI capex cycle reversal with broad semicap cuts >20%.
  • 2026-04-26 — Briefing published: Onto Innovation (ONTO) · vault.
  • 2026-05-15 — Rigaku hybrid x-ray/optical metrology co-development flagged (Collyer Bridge channel check); added to catalysts as a watch item pending Q1 2026 earnings disclosure. See 268A.
  • 2026-05-30 — Fragments consolidated into this single thesis-first canonical page (vault reorg W3). No new research; reconciled the ~$640M vs ~$622M net-cash and EV figures (both retained, flagged) and the FY2024 China 11.8%/'12%' vs FY2025 <3% framing.
  • Open / pending: Q1 2026 earnings (est. May 2026) is the explicit confirmation gate. A full /filings ONTO refresh is recommended before adding at scale (the onto-filings.md vault stub is unpopulated; underlying SEC sources — FY2024 10-K filed Feb 25 2025, 2026 DEF 14A, Form 4s, Semilab 8-K Nov 17 2025 — were incorporated indirectly).

Sources

Fragments folded into this page (2026-05-30 consolidation):

  • onto-profile.md — /profile, generated 2026-04-26 (Register D)
  • onto-buy-checklist.md — /checklist pre-buy, generated 2026-04-26
  • onto-deep-dive.md — /deep-dive, generated 2026-04-26
  • onto-mgmt-dd.md — /mgmt-dd, generated 2026-04-26
  • ONTO.md — prior canonical (created 2026-02-28, updated 2026-05-15); reconciled, not blindly trusted
  • onto-filings.md — empty placeholder stub; NOT merged (a full /filings refresh remains recommended)

External / cited sources (deduped across fragments):

Consolidation queue (merged 2026-05-30)

These fragment files were folded into this canonical page and remain live pending Pink's archive confirmation.

  • [ ] onto-profile.md
  • [ ] onto-buy-checklist.md
  • [ ] onto-deep-dive.md
  • [ ] onto-mgmt-dd.md
  • [ ] ONTO.md
Pages that link here (8)
268A — Rigaku Holdings CorporationTSEM — Tower SemiconductorAdvanced Packaging & SubstratesSemiconductor Capital Equipment & TestSemiconductors (cross-cutting / geopolitics)ONTO — Q3 Beat, Q4 Miss (Nov 2023)ONTO — Haitong Initiation (Jul 2024)ONTO — NANO & RTEC Merger Rationale (2019)
Open in Obsidian