BE Semiconductor Industries (BESI)
BE Semiconductor Industries (BESI)
One-line summary: Dominant advanced-packaging bonder maker; Q2-26 orders kept compounding (€292.9M, LTM record €987.6M) and TSMC AP7 is filling with BESI systems — but CoWoS 4× is not a 1:1 hybrid-bonding identity. HBM hybrid bonding is still evaluation.
What it is
Dutch semiconductor equipment company manufacturing die bonding, flip-chip bonding, thermo-compression (TC Next), and hybrid bonding tools used in advanced semiconductor packaging. Primary listing Euronext Amsterdam (BESI); US OTC BESIY. CEO Richard Blickman (founder, since 1995). Applied Materials owns a 9% stake (2025-04-14) as a hybrid-bonding partner, not a substitute. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
Why it matters to stock-market
BESI is the equipment name on cowos-capacity-expansion-to-besi-hybrid-bonding-backlog. The 2026-08-17 pass found the order cycle is real and still compounding after Q1, but the chain's "each incremental CoWoS line requires a BESI hybrid bonder" step is too tight. CoWoS-like (flip-chip / TCB / oS) is already a significant order slice; hybrid bonding is the SoIC / AP7 / CPO / future-HBM overlay. TSMC is buying BESI tools for AP6/AP7, not making bonders in-house. ASE's 2026 CoW overflow talks name Korean dicing/bonding suppliers, not BESI.
Key facts
- Q1 2026 (official, 2026-04-23): revenue €184.9M (+28.3% YoY); orders €269.7M (+104.5% YoY), "primarily due to a significant increase in bookings for hybrid bonding systems"; GM 63.5%; NI €51.6M; Q2 guide +30–40% sequential / 64–66% GM. From 2026-05-19-autoresearch-cowos-packaging-capacity-may-2026-update · 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- Q2 2026 (official, 2026-07-23): revenue €249.9M (+35.2% QoQ, +68.7% YoY) — beat the +30–40% guide; orders €292.9M (+8.6% QoQ, +128.8% YoY); GM 65.7%; NI €89.0M (net margin 35.6%); H1 revenue €434.7M / orders €562.6M (+116.5% YoY); LTM orders a record €987.6M. Q3 guide +10–15% sequential / GM 63–65%. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- €400M backlog is not a BESI number. Morgan Stanley's Nigel van Putten calculated ~€400M at end-March on the Q1 call; Blickman did not confirm. BESI does not disclose backlog or hybrid-bonder units. Visibility = orders + sequential revenue guides. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- Hybrid bonding customers: 15 at YE2025 → 21 at end-Q2-26. Q2 HB orders: two repeat + one new hyperscaler (CPO, logic, probably wearables — not HBM volume). AI applications ~60% of H1-26 system orders vs ~50% H1-25. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- CoWoS-like is already in the book. Blickman: CoWoS-like is "already a significant part of the orders" (started Q4/Q1/Q2); "increasing penetration of CoWoS, CoPoS" means flip-chip and TCB and hybrid. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- TSMC AP6/AP7. AP6 ~60 bonders. AP7 (TSMC Chiayi) filling with first systems as of Q2; published capacity ~double AP6; Q1 said "anticipated 100 bonders… may be significantly higher" plus CPO. First-round orders pulled into Q1; possible H2 second round; ~6-month lead time. Factory capacity 180 → 250 systems/year. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- HBM hybrid bonding: still evaluation at 1 of 3 memory makers, another starting; all three have tools. Cost/yield hurdles; Taiwan logic took >3 years. Not 2026 volume. 2027–2030 remains the Blickman horizon. Gate: Q3 print 2026-10-22. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- TC Next: qualifications expanding; "single systems, not yet major volume." From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- Applied 9% (2025-04-14): integrated die-to-wafer hybrid-bonding line; no board seat; no further-share plans. Partner, not substitute. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- Cycle caution (first-party). Blickman: typical up-cycle 6–8 quarters then overcapacity; will not call a super-cycle. From 2026-08-17-autoresearch-besi-hybrid-bonding-backlog.
- Unpriced on Twelve Data free plan (recurring). Snapshot structurally incomplete on cowos-packaging-capacity-crunch.
Strengths (from a thesis-input perspective)
- Q2 confirmed the Q1 order surge was not a one-quarter spike; book-to-bill still >1 (computed, not disclosed)
- TSMC captive AP7 is a named-facility (by Blickman + TrendForce) BESI tool fill, not in-house bonder substitution
- Applied partnership endorses the D2W hybrid-bonding stack rather than competing with it
- 64–66% gross margins held through the ramp
Weaknesses (from a thesis-input perspective)
- CoWoS 4× does not map 1:1 onto hybrid-bonder units; units undisclosed
- HBM hybrid-bonding second cycle is still evaluation
- ASE 2026 CoW overflow equipment talks name Korean TCB/dicing suppliers, not BESI
- Blickman's own 6–8 quarter cycle warning; Euronext / BESIY liquidity; Q1 doubling already public in April
Sources
- 2026-05-19-autoresearch-cowos-packaging-capacity-may-2026-update
- 2026-08-17-autoresearch-besi-hybrid-bonding-backlog