Skyworks Solutions (SWKS) — 2026Q2 FY2026 Earnings Call Analysis

60% of Revenue From One Customer, Content Flat, Book Still Above 1

They beat by $20M, guided through seasonal margin pressure, and admitted content at their only real customer is flat next year in the same call.

Thesis: Execution is real: beat above whispers, flat guide through input cost inflation, book-to-bill >1. But the equity is a leveraged bet on one customer with flat content per unit and a pending merger that carries China SAMR binary risk. The AI book is under $100M on a $3.7B base. Data center growing 50% adds $50M annually, which does not offset any volume miss at the 60% customer. The Qorvo close is the only path to 50-55% GM and it requires SAMR Phase 2 approval on an uncertain timeline.

Verdict: HOLD — Conviction: MEDIUM

Catalyst: Qorvo close unlocking $500M synergies and the 50-55% GM target. Management said 'increasingly hopeful' for late 2026, formal guidance still early 2027. SAMR Phase 2 is the gate.

Key Risk: The 60% customer cuts volume or changes forecast. Brace admitted customers 'change the forecast all along' and they are 'dealing with some of that now.' A single-digit volume miss at that customer overwhelms every growth engine combined.

The Tell: Brace volunteered book-to-bill >1 and lean inventory, then in the same answer said 'customers change the forecast all along. We're kind of dealing with some of that now.' He was asked about seasonality and volunteered the forecast-change admission unprompted. That is concern leaking through confidence.

Detected Patterns

Beat Above Buy-Side Whispers: Revenue $944M vs $924M high end, EPS $1.15 vs $1.10 high end. Beat was driven by mobile sell-through and broad markets outperformance, not one-time items. Beat above guidance midpoint was $0.11.
Backlog Quality: Brace cites book-to-bill >1 and lean inventory, then immediately says 'customers change the forecast all along, we're kind of dealing with some of that now.' The qualification undermines the headline metric in the same breath.
AI Label Without Substance: Data center is under $100M annually on ~$3.7B revenue base (under 3%). 800G, 1.6T, 800V HVDC named in prepared remarks. Growing 50% adds ~$50M/year, not enough to offset low-single-digit mobile decline at the 60% customer.
Pricing Power Signal: Carter: 'some of the long-life products that we're able to increase price and pass those costs on.' Guiding GM flat at 44.5-45.5% against input cost inflation, expedite fees, and gold price increases versus 5-year seasonal -70bps. Real but narrow.
Margin Expansion Engine: Post-Qorvo target of 50-55% GM vs today's 45%, with $500M synergies from fab rationalization and elimination of last real competitor in premium Android RF. Dependent on close completing.
Clean Answers: Carter quantified China at <$200M total, <$20M handset. DC under $100M. Auto ~$250M. Memory zero impact. No evasion on specific numbers when asked directly.

Friction Level: MODERATE_FRICTION — Both sides agree execution is clean and concentration is 60%. Bull reads flat content plus Qorvo synergies as temporary ceiling. Bear reads it as structural ceiling with no pricing power leverage at the top customer. Same numbers, opposite read on whether RF content ever grows again.

Report not found

The report data is no longer available. Please return to the archive.

60% of Revenue From One Customer, Content Flat, Book Still Above 1 | Silicon Signal