{
  "@context": "https://orbyd.app/schemas/dossier.v1.json",
  "ticker": "ATEYY",
  "name": "Advantest Corporation",
  "url": "https://orbyd.app/dossiers/ATEYY/",
  "json_url": "https://orbyd.app/dossiers/ATEYY.json",
  "status": "DORMANT",
  "current_conviction": "LOW",
  "graded_conviction": null,
  "archetype": {
    "code": "a2",
    "n": 2
  },
  "current_thesis": "AI test-intensity picks-and-shovels leader, but the narrative is MATURING: back-to-back upstream capex raises (TSMC $60–64B, ASML €43–45B) failed to lift the tape, and the ADR sits ~20% off its 2026-06-25 record, shedding another 6% on 07-24 into the binary 2026-07-29 Q1 print. Stand aside until the print resolves the H2 'digestion' flag.",
  "invalidation_trigger": "A weekly close below $158 loses the June–July consolidation shelf and the 2026 ADR uptrend; a 2026-07-29 Test System book-to-bill under 1, or 'digestion' language hardening into deferred orders, would confirm the theme flipping toward SATURATED.",
  "catalyst_date": null,
  "outcome": "OPEN",
  "outcome_date": null,
  "invalidation_fired": null,
  "themes": [
    "semi-foundry-equipment",
    "ai-chips-memory",
    "ai-datacenter-infrastructure"
  ],
  "tags": [],
  "sources": [],
  "notes": [
    "FY2026 company guide (ending 2027-03): net sales ¥1,420.0B, operating income ¥627.5B, net income ¥465.5B. Advantest revised FY2025 upward three times; the initial guide has historically been a floor.",
    "Earnings blackout: do not initiate inside 3 trading days before the quarterly print — Advantest earnings gaps have run 6–14% in both directions across 2024–2025.",
    "Next print is Q1 FY2026 on 2026-07-29 15:30 JST; the following H1 FY2026 read lands ~late-October 2026.",
    "Teradyne (TER) prints 2026-07-28 after the US close — its book-to-bill and memory-test commentary is the direct read-through hours ahead of Advantest.",
    "Test System book-to-bill and HBM/memory-test order commentary are the cleanest per-print narrative tells — track separately each quarter.",
    "ADR (ATEYY) liquidity is thin (<$5M/day avg); the 6857.T Tokyo ordinary is the deeper venue for size.",
    "USDJPY is a meaningful daily P&L driver for the ADR: >155 tailwind, <150 headwind — track as a macro overlay.",
    "FY2026 guide to beat: net sales ¥1,420.0B, operating income ¥627.5B (+25.7%), net income ¥465.5B, on FX of ¥150/USD and ¥170/EUR."
  ],
  "body_markdown": "## Current Thesis\n\nAdvantest is the largest semiconductor automated-test-equipment (ATE) maker and the most direct instrument for owning AI-chip *complexity* rather than AI-chip *volume*. Chiplets, 3D packaging and HBM stacks multiply test items and stretch test-time per device, so each accelerator generation burns disproportionately more tester hours. FY2025 (ended 2026-03-31, reported 2026-04-27) put net sales at ¥1,128.6B (+44.7% YoY), operating profit ¥499.1B (+118.8%), operating margin near 44%, with Test System up 51.1% to ¥723.1B.\n\nThe problem is not the franchise — it is the tape. On 2026-07-15 ASML raised full-year guidance for the second time in 2026 to €43–45B (from €36–40B), called order intake \"extremely strong,\" and crossed a $500B market cap. On 2026-07-14/16 TSMC lifted 2026 capex to $60–64B (from $52–56B), the single largest customer-side signal an ATE supplier can get. Both are unambiguously bullish for 2027 tester demand. The equipment complex sold anyway: the ADR fell 6.9% on 2026-07-16 to $174.98, and the Tokyo line (6857.T) shed another 6.0% on 2026-07-24 to ¥28,945, leaving the ADR ~20% under its 2026-06-25 record of $216.24. Good news landing while price falls is a MATURING narrative — the marginal buyer is already positioned. The 2026-07-29 Q1 print (15:30 JST), hours behind Teradyne's 07-28 read-through, is the binary that either restores the beat-and-raise reflex or hardens the H2 \"digestion\" flag management raised in April. Standing aside into that print is the disciplined stance.\n\n## Bull Case\n\n- **Customer capex just stepped up.** TSMC's 2026 capex hike to $60–64B (2026-07-16) and ASML's second 2026 raise to €43–45B (2026-07-15) lead ATE orders by two-to-four quarters and are not in the current ¥1,420.0B FY2026 guide.\n- **The FY2026 guide is built to be beaten.** The forecast for the year ending 2027-03 is net sales ¥1,420.0B, operating income ¥627.5B (+25.7%), net income ¥465.5B, on conservative FX of ¥150/USD and ¥170/EUR. Advantest revised FY2025 up three times off an initial forecast that proved ~30% light; a fourth raise cycle beginning 07-29 resets the story.\n- **HBM runs on its own clock.** Management (advantest.com IR, 2026-04) frames HBM3e/HBM4 investment as an independent growth curve; HBM4 mass production carries materially higher test difficulty per stack, and HBM4 qualification volume ramps through 2026–2027.\n- **Sell-side is re-engaging below the tape.** 2026 target raises (Bernstein Outperform ¥34,300; a separate lift to ¥39,200 from ¥32,300) now sit 15–35% above the 07-24 ¥28,945 close — upgrades clustering after a 20% drawdown carry different information than upgrades at the high.\n- **Capacity is a multi-year commitment.** The 70%+ production-capacity expansion targeted for end-2026 vs 2024 (announced 2025) is a throughput build sized for durable demand, spanning several accelerator generations.\n\n## Bear Case\n\n- **The tape is rejecting its own bull case.** Two upstream guidance raises inside 48 hours produced a -6.9% ADR session (07-16), then another -6.0% Tokyo session on 07-24. When a name stops responding to good news, the story is priced.\n- **The H2 \"digestion\" flag is unresolved.** April commentary flagged a second-half digestion phase; a Test System book-to-bill under 1 on 07-29, or \"digestion\" language hardening into deferred orders, converts the pullback into a de-rating.\n- **Binary print, three trading days out.** Advantest earnings gaps have run 6–14% in both directions across 2024–2025. Entering ahead of 07-29 is a coin-flip, not an edge.\n- **Concentration and FX.** Revenue leans on a short list of AI-accelerator and HBM customers, so any single-customer capex-timing wobble is amplified. The ADR also carries USDJPY on top of business risk — sub-150 yen is a direct headwind to the dollar line.\n- **Valuation leaves no cushion.** After a roughly 3x two-year run, a beat that merely satisfies the beat-and-raise reflex may not move price; only a guide raise well past ¥1,420.0B re-rates it.\n\n## Setup & Price Structure\n\nThe ADR set a record $216.24 on 2026-06-25, closed $188 on 2026-07-15, then gapped -6.9% to $174.98 on 2026-07-16 as the sector sold the ASML/TSMC news. Tokyo (6857.T) extended the fade to ¥28,945 on 2026-07-24 (-6.0%, from a ¥30,800 prior close), so the ADR sits roughly 20% off its high and drifts into the print on weak breadth. This is a MATURING theme: the 2026 uptrend is intact but the reflexive bid is gone. It is not a stretched-above-MA blow-off — the risk here is the binary catalyst, not extension. A weekly close that loses the June–July consolidation shelf around $158 breaks the structure; holding that shelf into and out of 07-29 keeps the base alive. No US-listed common stock exists — the ADR is the only US vehicle and it is thin (<$5M/day), so clean fills for size favor the 6857.T Tokyo ordinary.\n\n## Catalyst Calendar (next 30 days)\n\n- **~2026-08-14 (est.):** Applied Materials / broader WFE peer prints — confirms or breaks the capex-to-orders read.\n- **Ongoing:** USDJPY as a daily P&L overlay on the ADR; HBM4 qualification volume ramping through 2026–2027.\n\n## Elapsed catalysts\n\n- **2026-07-28 (after US close):** Teradyne (TER) Q2 2026 results — guide was $1.15–1.25B revenue, EPS $1.86–2.15, AI ~70% of Q1 mix. Direct ATE read-through hours before Advantest prints. *(passed 12d ago)*\n- **2026-07-29 (08:30 ET):** Teradyne outlook call — memory-test and book-to-bill commentary sets sector sentiment. *(passed 11d ago)*\n- **2026-07-29 (15:30 JST):** Advantest Q1 FY2026 print — the binary. Watch Test System book-to-bill (>1?), HBM/memory-test order commentary, the H2 \"digestion\" language, and any revision to the ¥1,420.0B sales / ¥627.5B OP guide. *(passed 11d ago)*\n\n## What Would Change Our Mind\n\nA 07-29 print with Test System book-to-bill above 1, HBM/memory-test at fresh highs, and a guide raise beyond ¥1,420.0B would re-accelerate the theme and justify a fresh setup on the first clean higher-low. Conversely, a weekly close below $158 on the ADR loses the June–July shelf and the 2026 uptrend; a book-to-bill under 1, or \"digestion\" language hardening into deferred orders, would confirm the theme sliding toward SATURATED. A clean beat that fails to move price — the maturing-narrative tell — argues for patience over a chase into the print.\n\n## Correlation Notes\n\n- **Teradyne (TER):** the closest ATE comp; its 07-28 print is the best same-week read-through into Advantest's demand.\n- **ASML / TSMC:** upstream capex signals lead ATE orders by 2–4 quarters; the 07-14/15 raises are bullish for 2027 but already public.\n- **Nikkei semi block / SOX:** the ADR trades as high-beta to the global semi-equipment complex; Tokyo drawdowns (07-24 -6%) transmit directly.\n- **HBM demand (SK Hynix, Micron):** memory-maker HBM4 ramps pull Advantest memory-test volume.\n- **USDJPY:** >155 is a tailwind to the ADR's dollar translation, <150 a headwind; treat as a macro overlay.",
  "first_seen": "2026-05-01",
  "last_analyzed": "2026-07-26T11:20:28+00:00",
  "last_synthesized": "2026-07-26",
  "last_update_source": "watchlist_research",
  "license": "Content © orbyd. Cite the canonical URL."
}