Compare moats
Up to three covered companies, band by band. Every call is a curated editorial judgment, never a disclosed figure — and every band carries its cited basis.
| Marvell Technology | SK Hynix | Celestica | |
|---|---|---|---|
| Moat rating | narrow FY2026 10-K (filed 2026-03-11): differentiated platform IP — over 10,000 issued patents and pending applications as of 2026-01-31, plus a proven custom ASIC platform leveraging ultra-high-speed SerDes, silicon photonics, co-packaged optics and custom HBM — but Marvell itself calls its markets 'intensely competitive' with 'pricing pressures', notes customers 'have chosen to develop certain semiconductor products internally', and discloses two >=10% customers with the ten largest at 82% of FY2026 net revenue. Real, defensible IP in a concentrated, contestable customer base = narrow, not wide. https://www.sec.gov/Archives/edgar/data/1835632/000183563226000011/mrvl-20260131.htm | narrow SK hynix held ~62% of the AI-critical HBM market in Q2 2025 (Counterpoint, via TweakTown) and was first to mass-produce HBM4, but memory remains a cyclical three-player business where Samsung and Micron are closing the HBM gap — a real but contestable edge. | narrow The FY2025 10-K shows a real but bounded edge. On the durable side, Celestica is increasingly engaged as an ODM: its HPS business co-designs (JDM) or fully designs the platforms it builds, it states 'We have hardware and software patents that are integral to our HPS business' and that 'our increased R&D activities have resulted in the growth of our dependence on our patent portfolio', and HPS reached 41% of total revenue in 2025 (from 21% in 2023) at a margin profile the filing says is higher than traditional EMS work. On the limiting side, the same filing concedes 'Some of our competitors have greater scale and provide a broader range of services than we provide', the master supply agreements 'do not typically guarantee a particular level of business or fixed pricing', work is won 'on a program-by-program basis', and the top 10 customers were 79% of 2025 revenue with three customers individually at 32%, 14% and 12%. An advantage that must be re-won each program against greater-scale rivals, for a handful of buyers who can in-source, is narrow rather than wide. |
| Moat type | intangibles ip The moat rests on hard-to-replicate mixed-signal IP: the 10-K describes the custom ASIC platform built on ultra-high-speed SerDes, ARM compute, security, storage, silicon photonics and advanced packaging (die-to-die interconnects, chiplets, CPO, custom HBM), with multiple 5nm designs executed, 3nm in progress and a 2nm platform in development; a secondary switching-cost element comes from multi-year custom design wins co-developed to individual customer specifications. https://www.sec.gov/Archives/edgar/data/1835632/000183563226000011/mrvl-20260131.htm | intangibles ip The advantage is process-and-packaging technology leadership: SK hynix achieved industry-first HBM4 mass production in September 2025 and has positioned itself as NVIDIA's key HBM partner, per its record FY2025 results. | switching costs Cost scale is explicitly not the source: the 10-K states competitors have greater scale and a broader service range. What does bind a customer is the design-in. Within HPS, Celestica 'design[s] and manufacture[s] products, either as customized solutions, white box solutions or under Joint Design and Manufacturing (JDM) engagements', holds 'hardware and software patents that are integral to our HPS business', and delivers 'complete platform solutions... integration and orchestration of various technologies into rack-scale designs'. The filing adds that 'a majority of these supply agreements also require the customer to purchase unused inventory that we have purchased to fulfill that customer's forecasted manufacturing demand', a contractual cost of walking away mid-program. Moving a qualified, jointly designed rack-scale platform to another ODM means requalifying a design Celestica partly owns, which is friction the commodity assembly work does not carry. |
| Leadership | co leader Leader in its optics niche, #2 in custom silicon: ~60% of high-end PAM4 DSP share (36kr, 2026-06-27, https://eu.36kr.com/en/p/3870758441178373) but an estimated 20-25% of custom AI ASIC design services versus Broadcom's ~70% (hashrateindex, 2026-05-13, https://hashrateindex.com/blog/design-partners-ai-asic-market-part-2/) — net, a co-leader in AI data-center connectivity/custom silicon behind Broadcom overall. | clear leader Clear leader in HBM (#1 at ~62% share, Q2 2025) and became the world's top overall memory supplier in Q2 2025; caveat: it sits #2 to Samsung in commodity DRAM by volume and Micron/Samsung are narrowing the HBM gap heading into HBM4. | at parity The 10-K claims no share leadership. It states plainly that 'Some of our competitors have greater scale and provide a broader range of services than we provide', and lists Hon Hai, Flex, Jabil, Sanmina, Benchmark and Plexus in EMS plus Quanta, Wiwynn and Accton in ODM. Its stated competitive advantage is execution quality, not position: 'our track record in advanced manufacturing capabilities, design and engineering, quality, delivery, managing complexity and responsiveness'. It is also winning: CCS revenue grew 42% to $9.19 billion and total revenue reached $12.39 billion in FY2025. Competing on comparable terms with a field it neither leads nor trails is parity. |
| Pricing power | moderate The 10-K characterizes Marvell's markets as having 'pricing pressures' and intensifying competition (https://www.sec.gov/Archives/edgar/data/1835632/000183563226000011/mrvl-20260131.htm), yet the Q1 FY2027 release reports 52.1% GAAP / 58.9% non-GAAP gross margin on record revenue (https://www.sec.gov/Archives/edgar/data/1835632/000183563226000014/q127_8kx522026ex-991.htm) — differentiated-IP margins, tempered by hyperscaler buyer power. | strong FY2025 operating margin hit a record 49% (operating profit +101% YoY) as HBM revenue more than doubled — unusually strong pricing power for a memory maker, per SK hynix's FY25 release. | weak The filing describes price as contested by contract, not set by Celestica: master supply agreements 'do not typically guarantee a particular level of business or fixed pricing', 'Some of these agreements require us to provide specific price reductions to our customers over the term of the contracts', and the Item 1A risk factor 'We operate in an industry comprised of numerous competitors and aggressive pricing dynamics' says competitors may be 'willing to, or able to make sales or provide services at lower margins than we do'. The site's fundamentals show gross margin at 12.06% in FY2025, up from 8.96% in FY2022, 9.47% in FY2023 and 10.72% in FY2024 — improving, but still low double digits, and the filing attributes the higher margin to mix (HPS is 'higher margin profile than our traditional EMS businesses' and went from 21% to 41% of revenue over the same span) rather than to raising prices. |
| Summary | Marvell is a fabless data-infrastructure silicon supplier whose center of gravity has shifted decisively to the AI data center: the data center end market was $6,100.3M, 74% of FY2026 revenue, up from 40% two fiscal years earlier (FY2026 10-K). Its strongest position is electro-optics — in high-end PAM4 optical DSPs for 400G+ transceivers it holds roughly 60% share on Inphi-inherited SerDes/FEC IP, with Broadcom above 30%, the two together over 90% (36kr, 2026-06-27). In custom AI silicon it is the structural #2 design partner at an estimated 20-25% of the custom AI ASIC design-services market versus Broadcom's ~70%, anchored by AWS Trainium and Microsoft Maia wins (hashrateindex, 2026-05-13). The Q1 FY2027 release (2026-05-27) shows the flywheel turning — record $2.418B revenue (+28% YoY), Q2 guided to $2.7B mid-point (+35% YoY), management citing 'exceptional AI-related bookings' across 800G/1.6T optics, 51.2T switches, CPO/NPO and custom XPU — and the Celestial AI (Photonic Fabric) and XConn (PCIe/CXL switching) acquisitions closed in February 2026 extend the interconnect moat toward scale-up fabrics. The offsets that keep the moat narrow are in Marvell's own filing: intense competition (AMD, Alchip, Astera, Ayar, Broadcom, Credo, GUC, Lightmatter and others), hyperscaler in-housing risk, and heavy customer concentration. | SK hynix's durable advantage rests on an HBM process-and-packaging lead: it held roughly 62% of the HBM market in Q2 2025 (Counterpoint) as NVIDIA's key HBM partner, and delivered industry-first HBM4 mass production. That lead drove record FY2025 results with a 49% operating margin, though rivals are qualifying competing HBM parts. | Celestica is a contract design and manufacturing company that has been pulled up the value chain by the AI data-center build-out. Its CCS segment, which builds networking switches, optical systems, data center racks, servers and storage for hyperscalers and other cloud and AI service providers, grew 42% to $9.19 billion in 2025, and Communications alone went from 33% of revenue in 2023 to 57% in 2025. The part of that business with a defensible position is HPS, where Celestica is the designer rather than the assembler: HPS revenue rose 81% in 2025 to 41% of the total, carries a higher margin than traditional EMS work, and rests on a patent portfolio the filing calls integral to the business. That design-in position is what separates it from pure build-to-print capacity. What caps the moat is the customer side of the ledger. The 10-K describes an industry where 'aggressive pricing is a common business dynamic', where master supply agreements guarantee neither volume nor price and some of them 'require us to provide specific price reductions to our customers over the term of the contracts', and where the company bids program by program against Hon Hai, Flex, Jabil, Sanmina, Benchmark and Plexus on the EMS side and Quanta, Wiwynn and Accton on the ODM side, plus Arista and Cisco where a customer might buy an off-the-shelf switch instead. Revenue concentration has tightened as the AI mix grew, from 64% of revenue in the top 10 customers in 2023 to 79% in 2025, with a single customer at 32%. The company also notes its HPS offerings can compete with a customer's own hardware, which may 'negatively impact our relationship with, or result in a loss of business from, such customers'. So the profile is a genuine but program-scoped advantage, held by a company whose fortunes turn on a few buyers' capital plans. |
| Chain position | Fabless supplier spanning 'data center core to network edge': it sits between hyperscaler AI compute (custom XPU/XPU-attach ASICs) and the optical layer (PAM4/coherent DSPs, CPO/LPO, DCI, AEC, PCIe retimers), outsourcing fabrication to independent CMOS foundries; the Feb-2026 Celestial AI and XConn acquisitions push it further into scale-up photonic fabric and PCIe/CXL/UALink switching (FY2026 10-K, https://www.sec.gov/Archives/edgar/data/1835632/000183563226000011/mrvl-20260131.htm). | Primary supplier of HBM (high-bandwidth memory) stacked onto NVIDIA's AI accelerators; HBM represented roughly 77% of Q2 2025 revenue. | Celestica sits between component suppliers and the hyperscale buyer, and owns the design only in part of that span. It 'procure[s] substantially all of our materials and components on behalf of our customers pursuant to individual purchase orders that are generally short-term in nature', then designs, assembles, integrates at rack scale and tests the switches, optical systems, servers, storage and data-center racks that cloud, AI and enterprise customers deploy. In ATS it is a pure EMS contractor that does 'not generally collaborate with ATS customers on the design of the solutions we manufacture'; in HPS it is the designer of record or a joint designer. The economics follow that split: the segment where it holds design IP is the one with the higher margin profile. |
| Products (share / barrier) |
|
|
|
| Long-horizon vote | +0.13 at weight 0.20 · swarm bullish Editorial prior, not backtested. | +0.24 at weight 0.20 · swarm bullish Editorial prior, not backtested. | -0.01 at weight 0.20 · swarm neutral Editorial prior, not backtested. |