They beat on every line, raised WFE to mid-teens, then guided operating margins down 160 basis points and admitted the mix is shifting toward the most cyclical segment.
Thesis: Nova owns real metrology intensity at the 2nm GAA and hybrid bonding inflection points, and 400bps annual share gains prove customers are locking in. But the guided operating margin compression from opex ramp, a memory mix that management says will get more cyclical this year, and an admitted supply constraint with no pricing commentary mean the quarter is priced for perfection at a mix they themselves flagged as temporary. The structural thesis is intact. The near-term setup is not.
Verdict: HOLD — Conviction: MEDIUM
Catalyst: Hybrid bonding pull-ins from memory customers converting to revenue, plus the new Asia facility coming online end of 2026 removing the capacity ceiling into a mid-teens WFE year.
Key Risk: DRAM capex pause flips the memory mix back to logic foundry where process control intensity is lower per Beckett's analysis, breaking both the margin and growth thesis simultaneously.
The Tell: CEO told Beckett 'it's too early to call how process control specifically will look like this year' in the same answer where he committed to outperforming mid-teens WFE growth. He cannot size his own core growth driver but still guaranteed outperformance. That is a claim without an engine.
Friction Level: MODERATE_FRICTION — Both sides agree share gains and beat-and-raise execution are real. Disagreement is whether guided 59% GM with accelerating memory mix and a 12-month capacity gap creates enough downside risk to offset structural metrology growth.
Report not found
The report data is no longer available. Please return to the archive.