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.
| Visa | SK Hynix | Applied Digital | |
|---|---|---|---|
| Moat rating | wide The FY2025 10-K prints a network comparison for calendar year 2024: Visa at $13,433B payments volume, 311B total transactions and 4,805M cards, against Mastercard's $8,014B / 204B / 3,146M and American Express's $1,750B / 12B / 147M. Footnote (1) sources the American Express, Diners Club / Discover, JCB and Mastercard data to The Nilson Report issue 1288 (June 2025); Visa's own line is Visa's own data. On that table Visa's payments volume is roughly two-thirds larger than the next network's, and the same section states 'Based on available data, Visa is one of the largest retail electronic funds transfer networks used throughout the world.' The rating is wide because the lead rests on a two-sided installed base the filing quantifies — nearly 5 billion payment credentials and more than 175 million merchant locations across more than 200 countries and territories — which an entrant would have to reassemble on both sides at once. The table is a single-year snapshot with no prior-year column, so it evidences the size of the lead, not its direction. | 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. | none The FY2026 10-K (filed 2026-07-29) shows contracted revenue, not a demonstrated competitive edge. About 1,410 MW is leased under 15-year take-or-pay, non-cancellable base terms worth about $36.2 billion, but only about 100 MW of the roughly 1.5 GW that is contracted and either operating or under construction was operating and earning revenue at May 31, 2026, and Item 1A says "lessees may have the right to terminate applicable leases if there are significant delays in construction." Item 1A also concedes "We do not have the resources to compete with larger providers of similar products or services at this time," and the Competition section names 13 power-advantaged developers the company competes with. Signed leases give revenue visibility, but the filing does not show a durable advantage. |
| Moat type | network effects Visa itself names the two sides as the source of advantage: 'We believe our fundamental value proposition of security, convenience, speed and reliability as well as the number of payment credentials and our acceptance footprint help us to succeed.' In fiscal 2025 the 10-K counts nearly 5 billion payment credentials, which it defines as issued Visa card accounts, available at more than 175 million merchant locations, with nearly 14,500 financial institutions among the clients that build payment programs on Visa products. The two bases are joined through VisaNet in what the filing calls the 'four-party' model. Neither side is worth joining without the other, so the advantage is the mutual pull of the two installed bases rather than a patent estate or a cost curve. | 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. | none The 10-K claims three advantages: power-advantaged sites (it believes securing power and interconnection ahead of demand is 'the principal constraint on new HPC capacity and a core differentiator for us from many of our competitors'), a standardized 'franchise-style' design, and hyperscaler master service and master telecom service agreements 'that are difficult to obtain.' The filing does not show any of them to be durable. Its Competition section says competition 'centers on securing and developing sites with access to large-scale, reliable, and cost-competitive power and interconnection' and names 13 power-advantaged developers going after the same leases, and Item 1A concedes it lacks the resources to compete with larger providers. Signed leases are take-or-pay and non-cancellable, so a tenant leaving for convenience owes 'the full remaining contractual value,' but that is contractual lock-in on each lease rather than a moat source, so no moat type is assigned. |
| Leadership | clear leader On the 10-K's CY2024 comparison table, Visa carried 311 billion total transactions and $13,433B of payments volume against 204 billion and $8,014B for Mastercard, the largest competitor listed. American Express is next at $1,750B and 147M cards — under a seventh of Visa's payments volume and under a thirtieth of its cards — with Diners Club / Discover ($253B, 72M cards) and JCB ($319B, 167M cards) smaller still on volume. Visa's line is its own data; the competitor lines are sourced to The Nilson Report issue 1288 (June 2025) per footnote (1). | 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. | behind The 10-K makes no leadership claim and gives no ranking or share figure. Item 1A concedes "We do not have the resources to compete with larger providers of similar products or services at this time" and that some rivals have "substantially greater liquidity and financial resources than we do." Its Competition section places APLD against established operators (Digital Realty, Equinix), hyperscalers that build their own capacity, independent developers and 13 named power-advantaged developers (IREN, Cipher Digital, TeraWulf, Hut 8, Riot, CleanSpark, HIVE, Core Scientific, Bitdeer, Galaxy Digital, Fermi, Keel Infrastructure, MARA). |
| Pricing power | strong Visa's own take is insulated from the fees regulators target: it sets default interchange rates but does not collect them — the 10-K says 'Generally, IRFs are paid by acquirers to issuers' and that 'the fees we receive from issuers and acquirers are not derived from IRFs or MDRs.' The reported margin is high but moved down in fiscal 2025: operating income of $23,994M on $40,000M of net revenue is a 60.0% operating margin, against 65.7% ($23,595M on $35,926M) in fiscal 2024. The cause is printed two lines above operating income in the same statement — the litigation provision rose from $462M to $2,562M — and the filing says litigation provisions 'do not correlate to the underlying performance of our business' and that it excludes them 'to facilitate a comparison to our past operating performance.' Adding that provision back to operating income in both years leaves 66.4% against 67.0%, so the pricing base held and the decline is a legal charge, not a fee concession. The real limits are indirect: Dodd-Frank and the EU IFR cap interchange Visa never earns, client incentives are paid back to clients and rise with payments volume, and the UK Payment Systems Regulator holds 'wide-ranging powers and authority to review our business practices, systems, rules and fees with respect to promoting competition and innovation in the UK, and ensuring payment systems take care of, and promote, the interests of service users.' | 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 Item 1A says "Due to the limited number of hyperscalers, we expect that a limited number of customers will continue to account for a high percentage of our revenue for the foreseeable future," and that if customers' equipment usage declines or they discontinue use of its facilities, APLD "may be compelled to lower our lease prices in some instances or risk losing a significant customer." One customer was 59% of FY2026 revenue from continuing operations. Take-or-pay, non-cancellable terms protect contracted revenue over the base term, and Note 19 reports a $39.1M HPC Hosting segment profit on $385.3M of segment revenue in FY2026, but those terms are agreed with a small group of concentrated buyers. |
| Summary | Visa runs the switchboard, not the bank. The FY2025 10-K is explicit that 'Visa is not a financial institution. We do not issue cards, extend credit or set rates and fees for account holders of Visa products nor do we earn revenue from or bear credit risk with respect to any of these activities.' That disclaimer is scoped to issuing and credit extension, not to risk at large — the same filing says indemnifying issuers and acquirers for one another's settlement failures 'creates settlement risk for us' because of the timing gap between a payment transaction and its settlement. What Visa earns is service, data-processing and international-transaction revenue, reduced by client incentives, for moving other people's money: 329 billion payments and cash transactions carried the Visa brand in fiscal 2025, 258 billion of them processed by Visa, an average of 901 million a day across more than 200 countries and territories on $17 trillion of total payments and cash volume. The pull between nearly 5 billion credentials and more than 175 million merchant locations is what a rival would have to buy on both sides at once. The filing is candid about what is arriving anyway: real-time payment networks have launched in at least 80 countries behind 'strong government sponsorship and regulatory initiatives' the filing names as FedNow, PIX and UPI; B2B blockchain payments including stablecoins 'can operate globally 24/7' for cross-border transactions; and Dodd-Frank and the EU IFR both cap interchange and limit network exclusivity and routing restrictions. | 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. | Applied Digital designs, builds and operates purpose-built, liquid-cooled HPC data centers, which it calls 'AI factories', and leases the capacity to CoreWeave and investment-grade hyperscalers. At May 31, 2026 its 10-K lists five campuses (Polaris Forge 1-3 and Delta Forge 1-2) with about 1,410 MW contracted under roughly 15-year take-or-pay, non-cancellable leases worth about $36.2 billion over the base terms. The filing claims three sources of advantage: it controls power-advantaged sites, it uses a standardized 'franchise-style' design built to deliver about 150 MW in about 14 to 18 months, and it holds hyperscaler master agreements that are 'difficult to obtain.' The same document shows how early the company is. About 100 MW was operating and earning revenue. One customer was 59% of FY2026 revenue from continuing operations. It competes with Digital Realty, Equinix, hyperscalers that build their own capacity and 13 named power-advantaged developers, and it concedes that it lacks the resources to compete with larger providers. Signed leases give long-dated revenue visibility, but the filing does not show a durable competitive advantage. |
| Chain position | Both a buyer and a seller of AI inside payments, not an AI infrastructure supplier: the 10-K claims 'early adoption and integration of artificial intelligence (AI) models in payment systems', sells 'risk detection and prevention solutions underpinned by real-time AI-driven scores' to issuers and acquirers, and opens its rails to third-party AI systems 'via on-demand APIs, our MCP server that enables AI systems to interface with our Visa Intelligent Commerce APIs, and fully managed solutions.' | Primary supplier of HBM (high-bandwidth memory) stacked onto NVIDIA's AI accelerators; HBM represented roughly 77% of Q2 2025 revenue. | Developer and landlord of power-advantaged, liquid-cooled AI data-center capacity, leased long-term to CoreWeave and investment-grade hyperscalers. |
| Products (share / barrier) |
|
|
|
| Long-horizon vote | +0.42 at weight 0.20 · swarm neutral Editorial prior, not backtested. | +0.24 at weight 0.20 · swarm bullish Editorial prior, not backtested. | -0.20 at weight 0.20 · swarm neutral Editorial prior, not backtested. |