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.
| MaxLinear | ASMPT | Equinix | |
|---|---|---|---|
| Moat rating | none MaxLinear’s FY2025 10-K does not show a durable advantage. Its income statement reports revenue of $693,263 thousand in 2023, $360,528 thousand in 2024 and $467,641 thousand in 2025, with operating losses of $38,221 thousand, $223,352 thousand and $126,890 thousand. The risk factors say “Increased competition has resulted in price pressure, decreased demand, reduced revenue and profitability, and loss of market share”; the business section says competitors include “companies with much longer operating histories, greater name recognition, and substantially greater financial, technical and operational resources”; and because its products “often are building block semiconductors” it also faces integrated-circuit makers, “some of which may be existing customers or platform partners”. Two customers were 28% of 2025 net revenue and the ten largest 65%, and “substantially all of our sales to date have been made on a purchase order basis”. Gross profit held up - $385,663 thousand, $194,782 thousand and $265,814 thousand for 2023 to 2025 - and the AI optical ramp has lifted 2026 results, but a revenue base that nearly halved in one year and three straight years of operating losses do not evidence a moat. | narrow ASMPT has a real but contested position in thermo-compression bonding (TCB), and its returns were thin until 2026. On the durable side, TrendForce (2025-12-12, citing The Elec) reports that SK hynix 'is using about fifty sets of TC bonders for HBM4 production, with roughly half supplied by ASMPT'. It adds that Hanmi had been SK hynix's exclusive TC bonder supplier for HBM 'until last year, when ASMPT joined the supply chain toward year-end'. ASMPT's FY2025 results announcement reports TCB revenue growth of about 146% and calls the company the 'Process-Of-Record' in chip-to-substrate bonding, which is its own claim. On the limiting side, the market is shared. TrendForce (2026-06-09, citing ET News) reports a 44.2 billion won SK hynix TC bonder order to Hanmi, and Kulicke & Soffa's FY2025 10-K lists ASM Pacific Technology among eight major equipment competitors. Returns were modest: the FY2025 announcement shows a continuing-operations adjusted gross margin of 38.3%, down 172 bps, and an adjusted SEMI segment profit margin of 8.4%. The 2026 interim results show a recovery, with SEMI's adjusted segment margin at 18.1% for 1H 2026. A contested technology lead with cyclical and recently thin returns supports a narrow moat, not a wide one. | wide The FY2025 10-K (filed 2026-02-11) grounds the advantage in an asset that took 27 years to assemble and that a competitor cannot buy: "Over our 27-year history, we have curated a diverse, industry-leading ecosystem of more than 500,000 interconnections", "over 10,500 customers, including 2,000+ network service providers and a leading market share of cloud-on ramps", across "280 data centers, in 77 markets in 36 countries", with "99.9999%+ operational uptime" delivered in 2025 and no single customer at 10% of revenue. Because the value of each IBX rises with who else is already inside it, incumbency compounds rather than decays. The counterweight is real and disclosed: Item 1A says "The global multi-tenant data center market is highly fragmented. It is estimated that we are one of more than 2,400 companies that provide these offerings around the world", and warns that competitors "may adopt aggressive pricing policies". That caps the pricing that the moat converts into, not its durability — the fragmentation sits in commodity space-and-power, while the interconnection density the filing describes has no comparable substitute. FY2025 revenue of $9.217B with operating margin recovering to 20.0% from 15.2% in FY2024 is consistent with the incumbency holding. |
| Moat type | none No single source of advantage in the 10-K is strong enough to name. MaxLinear has “over one thousand issued patents” and says consideration under intellectual property sale agreements “has previously been and is currently expected in the future be material”, but the same filing describes its products as building blocks that larger vendors can integrate, says some optical-interconnect customers are “module makers who are vertically integrated, where we compete with internally supplied components”, and claims only that “We believe that we compete favorably” on factors from product performance to price. Its RF-CMOS integration know-how is real, but nothing in the filing shows customers locked in or rivals unable to match it. | intangibles ip The advantage sits in bonding process technology rather than in customer lock-in. The FY2025 results announcement credits chip-to-wafer TCB orders to an 'ultrafine pitch TCB solution equipped with proprietary plasma AOR technology', and says its HBM4 12H TCB solutions 'were the first to secure orders from multiple players'. TrendForce (2026-01-28, citing EE Times China) says ASMPT holds TCB and hybrid bonding capability, 'which the industry widely regards as the second most important technologies after lithography'. Kulicke & Soffa's FY2025 10-K says assembly equipment competes on 'price, speed/throughput, production yield, process control, delivery time, innovation, quality and customer support', which rewards technology. Switching costs are the weaker candidate: TrendForce (2025-12-12) shows SK hynix sourcing HBM4 TC bonders from both ASMPT and Hanmi. | network effects The 10-K names the mechanism itself, twice and unprompted: "As more customers choose Equinix for high connectivity and performance reliability at the metro edge, it benefits their suppliers and business partners to colocate in the same data centers and connect directly with each other. This adjacency creates a network effect that attracts new customers while continuously enhancing our value proposition to existing customers", and in the Competitive Landscape section, "This ecosystem creates a network effect that improves performance and lowers the cost for our customers". The evidence is the count of participants rather than any patent or unit-cost claim — 500,000+ interconnections, 2,000+ network service providers, a leading share of cloud on-ramps, an Internet Exchange the filing calls "the largest global peering solution". Switching costs are a genuine second layer (fixed-duration contracts billed on space and power, physical cross connects into resident counterparties), but they are what holds a customer already inside the ecosystem; the reason to enter in the first place is who is already there. |
| Leadership | fast follower In its fastest-growing line MaxLinear is a second source. Deep Fundamental’s September 27, 2024 deep dive (https://deepfundamental.substack.com/p/deep-dive-optical-module-market) says “Marvell ($MRVL) and Broadcom ($AVGO) are two major suppliers of DSPs, with Marvell holding the top position in the market”, that “Coherent also sources heavily from Marvell, with Broadcom/ Maxlinear potentially serving as second supplier with 20-30% share”, and that MaxLinear offers “DSPs at about half the price of Marvell's if it can achieve a meaningful mass production volume of at least 100K units per month”. Keystone has since reached volume - management said on the Q2 2026 call that it “continues to ramp into high volume production at major hyperscale customers across U.S. and Asia” - but the 10-K names Broadcom, Qualcomm, Realtek, Skyworks, Credo, MediaTek, Marvell, MACOM, Texas Instruments, Analog Devices, Renesas, Microchip and Semtech as primary merchant competitors, and no third-party source found ranks MaxLinear first in any of its markets. | co leader In HBM TCB, TrendForce (2025-12-12, citing The Elec) reports that roughly half of the about fifty TC bonders SK hynix uses for HBM4 came from ASMPT, and that SK hynix ordered seven more ASMPT systems at about 4 billion won each. Hanmi remains a major supplier: TrendForce (2026-06-09, citing ET News and Hankyung) reports a new SK hynix order to Hanmi worth 44.2 billion won, believed to cover around 15 units. In logic, ASMPT calls itself the 'Process-Of-Record' for chip-to-substrate TCB, which is its own claim. In wire and die bonders, Kulicke & Soffa's FY2025 10-K names ASM Pacific Technology among its major equipment competitors. The only third-party supplier split found covers one customer, SK hynix, and no third-party share figure for TCB, wire bonders or SMT was found, so the band is co_leader rather than clear_leader. | clear leader Leadership is claimed on interconnection, not on square footage, and the filing's own evidence is about density: "our position is unmatched in the industry" is supported by 2,000+ resident network service providers, "a leading market share of cloud-on ramps", 500,000+ interconnections and an Internet Exchange described as "the largest global peering solution", across 36 countries. The band is read against neutral, ecosystem-dense colocation, where that footprint has no direct analogue. It is deliberately not read against total data centre capacity: the same 10-K puts Equinix among "more than 2,400 companies" in a "highly fragmented" MTDC market, and Item 1A concedes the company must compete for land and power against "new market entrants" drawn in by AI. |
| Pricing power | moderate Mixed. Gross profit has stayed in proportion through the cycle - the 10-K’s income statement shows $265,814 thousand on revenue of $467,641 thousand in 2025 - and the Q2 2026 release (https://www.sec.gov/Archives/edgar/data/0001288469/000128846926000050/a06302026exhibit991.htm) reports GAAP gross margin of 57.8% against 56.5% a year earlier, guiding Q3 to 57.0%-60.0%. But the 10-K says “From time to time, we have reduced the average unit price of our products due to competitive pricing pressures, new product introductions by us or our competitors, and for other reasons, and we expect that we will have to do so again in the future”, that under some distributor agreements “we provide protection for reductions in selling prices of the distributors' inventory”, and in optical DSPs it entered as the lower-priced challenger. | moderate Margins are healthy for equipment but not exceptional, and they move with volume. The FY2025 results announcement reports SEMI's adjusted gross margin at 43.3% (down 240 bps) and SMT's gross margin at 32.4% (down 218 bps), with a group adjusted gross margin of 38.3%. The 2026 interim results report Q2 2026 adjusted gross margins of 46.5% for SEMI and 36.8% for SMT, and attribute both year-on-year increases largely to higher volume. TrendForce (2026-01-28, citing EE Times China) says SMT 'mainly serves traditional PCB assembly', while semiconductor packaging equipment, 'especially advanced packaging tools for AI chips', offers 'significantly higher pricing power and margins'. | moderate Contract structure supports price: fixed-duration agreements billed on space and power plus per-connection interconnection fees, an installed base too physically entangled to move cheaply, and 99.9999%+ uptime in 2025 as the thing being paid for. FY2025 revenue of $9.217B grew about 5% on FY2024's $8.748B while operating margin recovered to 20.0% from 15.2%, so pricing and cost were at least held. But the filing itself refuses the strong band: competitors "may adopt aggressive pricing policies, especially if they are not highly leveraged or have lower return thresholds than we do. As a result, we may suffer from pricing pressure that would adversely affect our ability to generate revenues", and some rivals bundle communications or cloud services against bare colocation. Power procurement is a further pass-through risk the filing flags. Price is defended, not dictated. |
| Summary | MaxLinear is a fabless designer of RF, analog and mixed-signal communications SoCs whose core skill, per its FY2025 10-K, is combining broadband RF and analog front ends with digital signal processing in standard CMOS. It sells into broadband access (cable, fiber PON and DSL gateways - approximately 44% of 2025 net revenue), home connectivity (Wi-Fi, MoCA, G.hn and Ethernet), wired and wireless infrastructure including optical data-center DSPs, and industrial and multi-market interface and power products. After a downturn that took revenue from $693,263 thousand in 2023 to $360,528 thousand in 2024, the AI optical ramp is turning it around: the Q2 2026 release reports revenue of $168,847 thousand, up 55% year over year, with the infrastructure business up 145% on the Keystone PAM4 DSP ramp for 800G, and management raised its 2026 optical data-center revenue outlook to $210 million-$230 million on the call. The moat question is whether that growth rests on anything durable. In optical DSPs MaxLinear entered as a lower-priced second source to Marvell; elsewhere it competes with Broadcom, Qualcomm, Realtek and MediaTek, which can integrate the functions it sells; customers are concentrated; and the 10-K still carries the Silicon Motion arbitration over its terminated merger, whose outcome it says it cannot predict. On this record MaxLinear is a technically capable challenger without a moat. | ASMPT makes back-end manufacturing equipment in two segments. Semiconductor Solutions (SEMI) sells die and wire bonders, thermo-compression and hybrid bonders, and photonics assembly tools. SMT Solutions sells surface-mount systems for circuit-board assembly. In FY2025, continuing-operations revenue was HK$13,736.2 million, up 10.0%. SEMI contributed HK$7,380.5 million and SMT HK$6,355.8 million. Advanced packaging revenue was US$532.1 million, up 30.2% and 30% of group revenue, with a significant contribution from TCB. The company estimates the TCB market at about US$760.0 million in 2025, rising to US$1.6 billion in 2028, and targets a 35 to 40 percent share. Business accelerated in 2026. Continuing revenue for 1H 2026 was HK$8,902.6 million, up 42.5% year on year, at a 41.1% gross margin. In July 2026 the company received bulk orders for more than 50 chip-to-substrate TCB tools from OSAT customers. In memory, the interim results say the timing of new TCB purchases 'remains dependent on HBM4 product rollout schedules'. The portfolio is being reshaped. The sale of the NEXX deposition business closed on 3 June 2026, and TrendForce (2026-05-05) reported Applied Materials as the buyer at a base value of US$120 million. On 21 January 2026, ASMPT began a strategic options assessment for SMT that includes a possible divestiture, joint venture, spin-off or listing. The verdict is a narrow moat: a credible TCB position in a market shared with Hanmi and others, attached to mainstream bonder and SMT businesses with ordinary margins. | Equinix is a network-neutral, multi-tenant colocation and interconnection REIT: it does not sell compute, it sells the metro-edge real estate where networks, clouds and enterprises physically meet. The FY2025 10-K describes the platform as "280 data centers, in 77 markets in 36 countries" serving "over 10,500 customers, including 2,000+ network service providers and a leading market share of cloud-on ramps", carrying "more than 500,000 interconnections" curated over 27 years, with 61% of 2025 revenue recognised outside the U.S. Revenue is structurally recurring — infrastructure offerings are "billed based on the space and power a customer consumes" under fixed-duration contracts generating MRR, interconnection is "billed based on the outbound connections from a customer" — and no customer reached 10% of 2025 revenue. AI enters the story as demand rather than as a product: the filing positions Equinix as the interconnect point for "model providers, data platforms, neoclouds and gateways", and pushes core hyperscale capacity into xScale, which is "developed and operated through our joint venture partnership arrangements". The bear case is in the company's own Item 1A. The MTDC market is "highly fragmented", Equinix being "one of more than 2,400 companies"; competitors "may adopt aggressive pricing policies"; the AI build-out invites "significant investments in the data center industry by both current competitors and new investors", after which "we could lose market share" and must "compete against certain of these competitors to secure the land and power needed for our expansion plans". Product extension has also failed before — the filing notes past offerings "have been or are being discontinued, including the Equinix Metal product". The honest reading: the interconnection ecosystem is close to unreplicable and the moat sits there; the capacity business around it is a capital race Equinix enters with scale but no immunity. |
| Chain position | MaxLinear sells chips, much of it through Asian distributors, ODMs and module makers - products shipped to Asia were 82% of 2025 net revenue, including 49% to Hong Kong - that build cable modems, PON terminals, Wi-Fi gateways, base-station radios and optical transceivers for operators and hyperscale data centers. In the AI chain it sits beside the optical-module makers as a DSP supplier, competing with Marvell and Broadcom and, at vertically integrated module makers, with internally supplied components. | Back-end equipment supplier to IDMs, OSATs, memory makers and electronics assemblers. China was 41% of 2025 revenue and the top five customers about 16%, per the FY2025 results announcement. | Equinix is the neutral meeting point of the AI and cloud supply chain rather than a link in its manufacturing path: it houses other companies' compute and sells the adjacency between them. The 10-K places it between the network layer (2,000+ service providers), the cloud layer ("a leading market share of cloud-on ramps") and enterprise consumers who "assemble these capabilities into operational stacks", and describes an AI ecosystem "of model providers, data platforms, neoclouds and gateways" curated for enterprise AI demand. Core hyperscale training capacity sits beside that, not inside it, in xScale, built with JV partners so hyperscalers "add to their core hyperscale data center deployments and existing customer access points at Equinix". The revenue is therefore levered to AI's distribution and inference edge more than to training-cluster buildout. |
| Products (share / barrier) |
|
|
|
| Long-horizon vote | +0.01 at weight 0.20 · swarm neutral Editorial prior, not backtested. | +0.13 at weight 0.20 · swarm bearish Editorial prior, not backtested. | +0.38 at weight 0.20 · swarm neutral Editorial prior, not backtested. |