{
  "@context": "https://orbyd.app/schemas/dossier.v1.json",
  "ticker": "FORM",
  "name": "FormFactor, Inc.",
  "url": "https://orbyd.app/dossiers/FORM/",
  "json_url": "https://orbyd.app/dossiers/FORM.json",
  "status": "DORMANT",
  "current_conviction": "MEDIUM",
  "graded_conviction": "MEDIUM",
  "archetype": {
    "code": "a2",
    "n": 2
  },
  "current_thesis": "Probe-card pick-and-shovel to the HBM4 test ramp; after a parabolic run to a fresh $160.27 ATH (2026-06-30) on Russell 1000 inclusion, the 2026-07-02 chip selloff cut FORM ~24% to ~$121. HBM inflection intact (record DRAM quarter, second SmartMatrix customer), but the 2026-07-29 Q2 print is the binary that re-rates it.",
  "invalidation_trigger": "A weekly close below $110 forfeits the post-Investor-Day recovery base and turns the July correction into a trend break; a Q2 revenue miss versus the $240M ±$5M guide, or SK hynix confirming a slower HBM4 ramp on its late-July call, would confirm it.",
  "catalyst_date": "2026-08-26",
  "outcome": "INVALIDATED",
  "outcome_date": "2026-07-07",
  "invalidation_fired": true,
  "themes": [
    "semi-foundry-equipment",
    "ai-chips-memory"
  ],
  "tags": [],
  "sources": [],
  "notes": [
    "Customer concentration: probe cards into the top-3 HBM makers, so the growth vector is SK hynix / Micron / Samsung capex and HBM4 qualification timing.",
    "Sell-side dispersion is unusually wide: TD Cowen Hold $100 and Evercore Outperform $150, both set on 2026-07-30 after the same print.",
    "GAAP and non-GAAP gross margin differ materially for this name; check the basis before comparing any reported figure to the 49.5% non-GAAP guide.",
    "Russell 1000 inclusion on 2026-06-29 added a passive holder base; that mechanical bid is elapsed and is not a forward driver.",
    "The 2026-05-11 Investor Day 2030 model ($1.6B revenue, 55% GM, $5.00 non-GAAP EPS) anchors a rich multiple that derates on any quarter of imperfection.",
    "Next company-specific binary is the Q3 2026 print, expected late October 2026; nothing company-dated falls inside the next 30 days."
  ],
  "body_markdown": "## Current Thesis\nThe binary resolved in favour of the fundamental leg, and the stock is still 26.6% below its high. FormFactor's 2026-07-29 Q2 print put revenue at $258.242M against a $239.999M consensus and adjusted EPS at $0.82 against $0.61, then guided Q3 to $260–280M revenue and $0.77–$0.95 adjusted EPS versus street numbers of $247.354M and $0.63. Shares rose 26.6% on 2026-07-30. Seven sessions later the adjusted close was $117.39 (2026-08-07), against a 52-week high of $159.93 and a three-month return of -20.6%. The probe-card content story — HBM4 layer count and pin count driving test intensity, and therefore probe-card wear and replacement frequency — is compounding in the reported numbers. The multiple is not following it, and both a bull and a bear cut their targets the morning after the beat.\n\n## Bull Case\n- Q2 revenue of $258.242M (2026-07-29) came in above the top end of the company's own $240M ±$5M guide issued on the 2026-04-29 Q1 call — the beat was against management's range, not only against consensus.\n- Q2 revenue grew 14% QoQ and 32% YoY to a company record, per the Q2 results materials (2026-07-29), with management attributing demand to DRAM, Foundry & Logic and systems, driven by HBM and co-packaged optics.\n- The Q3 guide raises rather than digests: $260–280M revenue versus $247.354M consensus and $0.77–$0.95 adjusted EPS versus $0.63; GAAP EPS guided $0.66–$0.84 versus $0.51 estimated (2026-07-29). Guiding a second consecutive record after a 32% YoY quarter is the specific thing a \"peak HBM test\" bear case does not survive.\n- 2026-07-28: expanded strategic partnership with Keystone Microtech covering probe-card assembly, test and distribution in Taiwan — a supply/capacity node next to the foundry and OSAT customer base, announced one day before the print.\n- Evercore ISI maintained Outperform on 2026-07-30 with a $150 target — roughly 28% above the 2026-08-07 close of $117.39.\n- The 2026-05-11 Investor Day 2030 model ($1.6B revenue, 55% gross margin, 23% opex, $5.00 non-GAAP EPS) now has one quarter of evidence behind its revenue slope rather than none.\n\n## Bear Case\n- Targets were cut into the beat. On 2026-07-30, TD Cowen kept a Hold and lowered its target to $100 — below the 2026-08-07 close of $117.39 — while Evercore, still Outperform, took its own number down to $150. A record quarter that produces target reductions on both sides of the rating spread says the sell side is marking down the multiple faster than it is marking up the estimates.\n- The best print in the company's history has not reclaimed the 2026-06-30 high.\n- Margin basis is an open question. The Motley Fool's 2026-07-30 write-up cited Q2 gross margin of 43.97% against a non-GAAP guide of 49.5% ±150bps given on 2026-04-29. Those are not the same measure and the spread between FormFactor's GAAP and non-GAAP margin is normally wide, but no reconciled non-GAAP gross margin appeared in the headline coverage — worth checking in the Q2 filing before treating the beat as clean.\n- Customer concentration is unchanged: probe cards into the top-three HBM makers means the growth vector is SK hynix, Micron and Samsung capex timing. Reporting through mid-2026 has pointed to SK hynix weighing HBM4 ramp pace against conventional DRAM margins; a qualification push-right at the lead customer hits FormFactor's DRAM line directly.\n- Sell-side dispersion of $100 to $150 on the same 2026-07-30 date is a 50% spread on a name trading between them. That is a valuation argument, not an earnings argument, and valuation arguments resolve slowly.\n\n## Setup & Price Structure\n- Adjusted close $117.39 (2026-08-07); 52-week high $159.93; distance from high -26.6%; three-month return -20.6%; RSI(14) 57.2 — mid-range, neither washed out nor extended.\n- Structure since the print is a narrow shelf: the 2026-07-30 gap-up session, a 2026-08-05 close of $114.44 as the low mark, and $117.39 on 2026-08-07. Roughly $114–120 is the working consolidation. Beneath it sits the $108–112 base the stock built after the 2026-05-11 Investor Day.\n- Overhead is the June structure: the 2026-06-29 Russell 1000 inclusion, the 2026-06-30 all-time high, and the trapped supply from the subsequent drawdown into late July. Reclaiming the high requires absorbing every holder from that window.\n- **Narrative life-cycle: MATURING.** Attention peaked around the 2026-06-29 index inclusion and 2026-06-30 high; the story is now well known and still working — the 2026-07-29 numbers are the proof — but flow is moderating rather than expanding. The marker is 2026-07-30: the fundamental news was as good as it gets and the sell-side response was two target reductions, with the stock closing the following week below its gap-day excitement.\n- Positioning observables, stated as observables: shares sit 26.6% below the 52-week high, so there is no extension above a rising trend to unwind; RSI 57.2 is unremarkable; the mechanical passive bid from Russell 1000 inclusion is elapsed (2026-06-29); no company-specific earnings date falls inside the next 30 days; the recent filing record shows no Form 4 insider sales or equity issuance into the June strength, though absence from a feed is not proof of absence.\n\n## Catalyst Calendar (next 30 days)\n- **2026-08-26** — NVIDIA Q2 FY2027 results and call (confirmed by NVIDIA, call 2:00 p.m. PT; quarter ended 2026-07-26). Not a FormFactor event, but the dominant read-through for AI-accelerator and HBM demand pacing that the memory-test complex trades on.\n- **No dated FormFactor-specific event falls inside the window.** The next company binary is the Q3 2026 print, expected late October 2026 (~2026-10-28, est.), which is where the $260–280M guide is graded.\n- **~2026-09-24 (est.)** — Micron fiscal Q4 results, just outside the 30-day window; the nearest direct HBM capex and pricing read from a named FormFactor customer.\n\n## What Would Change Our Mind\nThe structural break is the loss of the post-print shelf. If the $114–120 consolidation that formed after 2026-07-30 gives way and the stock re-enters its pre-earnings range, the market will have rejected a record beat-and-raise — the strongest fundamental input available for another eleven weeks. Gradeable form: **a weekly close below $110** forfeits that shelf and the $108–112 Investor Day base beneath it. A secondary confirmation would be Q3 revenue tracking or landing below the $260M low end of the 2026-07-29 guide, or a top-three memory customer guiding HBM capex or HBM4 qualification timing lower.\nOn the other side, the read strengthens if the stock closes above the $150 Evercore target zone and, more importantly, above the 2026-06-30 high on volume, since that would clear the entire July supply overhang. A more mundane change of mind: a reconciled Q2 non-GAAP gross margin materially below the 49.5% ±150bps April guide would recast the revenue beat as a mix-and-price problem rather than a content-per-stack win.\n\n## Correlation Notes\n- Trades as a high-beta expression of memory capex. The 2026-07-02 chip-sector selloff took FormFactor down with the group regardless of company news, and the 2026-07-30 rebound headlines grouped it with Lam Research and Microsoft — the name follows semicap and AI-capex sentiment before it follows its own bookings.\n- Direct fundamental dependencies: SK hynix, Micron and Samsung capex and HBM4 qualification schedules. Micron's fiscal Q4 (~late September 2026, est.) is the next dated read from a customer.\n- Since 2026-06-29 the shares carry Russell 1000 membership, so index and large-cap passive flows now sit alongside the small-cap holder base that owned the move to the June high.\n- Peer complex for confirmation or divergence: Aehr Test Systems, Camtek, Onto Innovation, Lam Research, Applied Materials. A FormFactor drawdown while the test peers hold would point at a company-specific issue; a synchronized move is sector beta.",
  "first_seen": "2026-04-19",
  "last_analyzed": "2026-08-08T15:34:43+00:00",
  "last_synthesized": "2026-08-08",
  "last_update_source": "watchlist_research",
  "license": "Content © orbyd. Cite the canonical URL."
}