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.
| Trane Technologies | GitLab | Cerebras Systems | |
|---|---|---|---|
| Moat rating | narrow Trane's FY2025 Form 10-K (filed 5 February 2026) makes exactly one rank claim about itself and it is hedged: 'We are one of the leading manufacturers in the world of HVAC systems and services and transport temperature control products and services'. It sits inside a Competitive Conditions section that opens 'Our products and services are sold in highly competitive markets throughout the world', describes 'a wide variety of competitors that vary by product line and services' including 'well-established regional or specialized competitors, as well as larger U.S. and non-U.S. corporations or divisions of larger companies', and names price first among 'the principal methods of competition in these markets', ahead of quality, delivery, service and support, technology and innovation. The risk factors push the same way: 'The markets that we serve are highly competitive', 'consolidation and new entrants (including non-traditional competitors) within our industries' 'could result in increased competition and pricing pressures', Trane competes 'with large companies and with smaller, local operators who may have customer, regulatory or economic advantages in the geographies in which they are located', and 'some of our competitors may employ pricing and other strategies that are not traditional'. What the filing does evidence is an advantage that gets paid for: 2025 net revenues of $21,321.9 million, up 7.5% from $19,838.2 million, of which the company's own bridge assigns 3.0 points to pricing; gross margin up 50 basis points to 36.2%; operating income of $3,967.4 million, 18.6% of revenues; and firm backlog up to $7,769.4 million from $6,747.7 million. But that advantage is not uniform in the same document: the EMEA pricing component was (0.3)% and its segment adjusted EBITDA margin fell 150 basis points to 18.3%, and Asia Pacific organic revenue fell 2.5% on 'lower volumes in China'. A position the company itself will only call 'one of the leading', defended in markets it calls highly competitive and priced unevenly across its three segments, is protection that is real but bounded. | narrow The FY2026 10-K (year ended 2026-01-31) shows real expansion inside the installed base: 118% Dollar-Based Net Retention, Base Customers up to 10,682 from 9,893, customers above $100,000 of ARR up to 1,456 from 1,229, and $1.0 million ARR customers "to 155 from 123, an increase of 26%". But the filing concedes "limited barriers to entry" and names "Microsoft Corporation, which owns GitHub" as principal competitor. Retention decelerates (130% FY2024, 123% FY2025, 118% FY2026), and since DBNR is reported "on a threshold basis of 130%," the FY2024 figure is a cap and the slide is steeper than it looks. Narrow, not wide: expansion must be re-won against a bundled hyperscaler. | narrow One-of-a-kind silicon sold to very few buyers: the IPO prospectus discloses G42 at 24.0% of 2025 revenue (85.0% in 2024), MBZUAI at 62.0% of 2025 revenue, and a December 2025 MRA with OpenAI representing 'a substantial portion of our projected revenues over the next several years' — against a competitive field the same document names as NVIDIA, AMD, Intel, hyperscaler accelerators and the major clouds. |
| Moat type | switching costs The filing rules out the obvious alternative in its own words. Item 1's Patents and Licenses paragraph says the Company 'does not consider any single patent, trademark, copyright, trade secret, proprietary technology, technical data, business process or any other confidential information (or any related group of any such items) to be of material importance to any segment or to the business as a whole', and the matching risk factor is blunter: 'we do not believe that our business is materially dependent on a single intellectual property right or any group of them. In our opinion, engineering, production skills and experience are more responsible for our market position than our patents and/or licenses.' That is a company disclaiming an IP moat. Where the 10-K does locate durability is the installed equipment and the channel wrapped around it. The overview frames the entire growth plan that way - 'As an industry leader with an extensive global install base, our growth strategy includes expanding recurring revenue through services and rental options' - the principal products and services table carries 'Service agreements', 'Repair and maintenance services', 'Parts and supplies (aftermarket and OEM)' and 'Rental services' alongside the equipment itself, the Distribution paragraph puts U.S. sales through Trane's own 'branch sales offices, distributors and dealers across the country' and non-U.S. sales through 'numerous subsidiary sales and service companies with a supporting chain of distributors throughout the world', and the accounting policy confirms the contracts are multi-period: 'For extended warranties and long-term service agreements, revenue for these distinct performance obligations are recognized over time on a straight-line basis over the respective contract term.' Long-lived equipment serviced under term agreements by the manufacturer's own branches is the asset this filing keeps pointing at, which is why the moat is read as an installed-base tie rather than a patent estate. | switching costs Stickiness comes from platform embedding, not IP. The 10-K rests differentiation on "our single platform with a unified data model," on "Consolidation of multiple tools into a single platform," and on deployment inside the customer's own perimeter for regulated buyers. Replacement means re-integrating a toolchain and re-certifying compliance. IP is the wrong label: only "16 issued patents and 25 pending patent applications," plus open-source licenses granting "broad permissions to use, copy, modify, and redistribute." The no-lock-in claim is the filing's own characterisation, footnoted to a Forrester study "commissioned by GitLab." | intangibles ip Wafer-scale integration is the moat's substance: the prospectus contrasts the WSE-3's 4 trillion transistors on 46,225 mm² of silicon with NVIDIA's B200 at 208 billion on roughly 1,600 mm² — a commercial chip category with one occupant. |
| Leadership | co leader The company's own claim is comparative but never exclusive: 'We are one of the leading manufacturers in the world of HVAC systems and services and transport temperature control products and services' - one of, with no rank, no share figure and no competitor named anywhere in the filing. Competitive Conditions sets that against 'well-established regional or specialized competitors, as well as larger U.S. and non-U.S. corporations or divisions of larger companies', and the risk factors add that Trane competes 'with large companies and with smaller, local operators who may have customer, regulatory or economic advantages in the geographies in which they are located'. The scale behind the claim is substantial and disclosed - $21,321.9 million of 2025 net revenues, products sold in approximately 100 countries, backlog of $7,769.4 million - but the 10-K never asserts the top position in any market it serves. | fast follower The 10-K positions GitLab against a leader rather than as one: "Our principal competitor is Microsoft Corporation, which owns GitHub." Competition reads from the challenger side, differentiating "from GitHub through flexible deployment options that work within enterprise security and compliance requirements, LLM neutrality with self-hosted gateway support" and an open core model. On AI it follows: Duo Agent Platform went "Generally available in January 2026," the last month of the fiscal year. What keeps it a follower rather than behind is cadence, "a new version of our software every month for 172 months in a row," plus a regulated-deployment franchise a SaaS-first rival does not cover. | clear leader Scoped to the category the record supports: wafer-scale integration has no other commercial occupant per the prospectus's own comparison, and the Q2 release claims — attributed, via AMD and AWS partnerships — leadership in disaggregated inference, a technique the prospectus defines. |
| Pricing power | moderate Price is a separately disclosed component of growth in this filing and it is positive: of the 7.5% increase in 2025 net revenues, the company's bridge assigns 3.0% to pricing against 3.2% volume, 0.8% acquisitions and 0.5% currency, and gross margin rose 50 basis points to 36.2% 'primarily due to gross productivity and price realization, partially offset by inflation'. In the Americas the price component was 3.8% of an 8.0% revenue increase, and that segment's adjusted EBITDA margin rose 70 basis points to 21.6% 'primarily due to price realization and gross productivity, partially offset by inflation and continued business reinvestment'. What holds the band below strong is in the same bridge and the same risk factors: EMEA pricing was (0.3)% and its segment margin fell 150 basis points to 18.3%; Asia Pacific pricing was 0.4% on organic revenue down 2.5%; consolidation and new entrants 'could result in increased competition and pricing pressures'; and 'in the event there is deflation, we may experience pressure from our customers to reduce prices.' Price that sticks in the Americas, disappears in EMEA and is disclaimed as durable in the risk factors is moderate rather than strong. | moderate FY2026 gross margin was 87% against 89%, with cost of revenue up $35.6 million "primarily due to an increase of $18.4 million in third party hosting costs for SaaS and cloud usage," and management expects SaaS and Duo mix to bring costs that "may adversely impact our gross margins." Strain is explicit: GitLab "may be required to reduce our prices," competitors "may offer their products and services at a lower price or for free," and it "implemented user limits on our free SaaS product." Power retained: Premium lists at $29 per user/month billed annually, and seats bundle $12 and $24 of GitLab Credits. But "A majority of our subscriptions are on a one-year period." | moderate The CFO states Q2 'significantly improved core gross and operating margins compared to a year ago' — improving, from a base where three named customers hold most of the revenue and the leverage that comes with it. |
| Summary | Trane Technologies is the Trane and Thermo King equipment maker - the FY2025 10-K opens by calling it 'a global climate innovator' selling HVAC, transport refrigeration and custom refrigeration through 'our strategic brands, Trane® and Thermo King®' - organised in three regional segments that produced 2025 net revenues of $17,168.8 million (Americas), $2,802.1 million (EMEA) and $1,351.0 million (Asia Pacific), with 'Approximately 25% of our net revenues in 2025 ... derived outside the U.S.' across 'approximately 100 countries' and 'no single external customer that accounted for more than 10% of our consolidated net revenues in 2025, 2024 or 2023'. The filing is candid that the markets are contested and equally candid that its position does not rest on intellectual property: 'engineering, production skills and experience are more responsible for our market position than our patents and/or licenses.' The durable asset it does describe is the installed base and the aftermarket attached to it - an 'extensive global install base' whose monetisation through 'services and rental options' is the stated growth strategy, reached through branch sales offices, dealers and distributors, with service agreements, repair and maintenance, aftermarket and OEM parts and rental all listed among the principal products. The 2025 financials carry that signature: revenues up 7.5% to $21,321.9 million split 3.2 points volume, 3.0 points price, 0.8 points acquisitions and 0.5 points currency; gross margin up 50 basis points to 36.2% 'primarily due to gross productivity and price realization, partially offset by inflation'; and backlog of $7,769.4 million on equipment where 'orders for specialized equipment or specific customer applications are submitted with extended lead times'. The limits are in the same filing. Price contributed 3.8% in the Americas but (0.3)% in EMEA and 0.4% in Asia Pacific; EMEA's segment adjusted EBITDA margin fell 150 basis points to 18.3%; Asia Pacific organic revenue fell 2.5% on 'lower volumes in China'; the Americas' otherwise strong year was 'partially offset by weaker volume in our Residential business'; and the risk factors warn that refrigerant regulation 'could make some of our existing HVAC and refrigeration products non-compliant or obsolete', with the company 'developing and selling our next generation products that utilize lower global warming potential solutions' against $347.6 million of 2025 research and development spending. For the AI build-out specifically, the link exists in the filing but is unsized: 'Data center HVAC systems', 'Data center liquid cooling solutions', 'Data center facility controls' and 'Data center services' appear in the principal products and services table, and those four entries are the only places the phrase occurs anywhere in the 10-K - no revenue, no ranking, no share, and no mention at all in the MD&A discussion of 2025 growth. | GitLab sells a single-platform DevSecOps toolchain defended by workflow lock-in rather than intellectual property: a unified data model plus deployment flexibility, including single-tenant "Dedicated for Government with FedRAMP compliance," makes it hard to remove where data residency is contractual. The installed base expands: 118% Dollar-Based Net Retention, 10,682 Base Customers, 1,456 above $100,000 of ARR, 155 above $1.0 million (up 26%), over 70% of ARR from public sector and enterprise. The filing supplies the counterweight: "limited barriers to entry," Microsoft/GitHub as principal competitor, 16 issued patents. Retention (130 to 123 to 118, the 130 a cap), gross margin (89% to 87%) and a January 2026 AI launch point to compression at the commodity end. No source measures market share, so every share band is "unknown". | Cerebras bet that AI compute should be one enormous chip rather than many small ones — 'quantities of compute and memory never before assembled on a single commercial chip,' avoiding 'the latency and the power-draw induced by the traditional approach,' in the prospectus's words — and is converting that bet into an inference cloud: Q2 2026 cloud revenue grew 281% with 600 MW of data center capacity under contract and $25.4 billion of remaining performance obligations. The concentration that funded the bet is the risk that remains: two related-party Gulf customers were 86% of 2025 revenue between them, with the OpenAI MRA now layered on top. |
| Chain position | Trane sits on the facility side of the AI build-out rather than in the compute stack. Its FY2025 10-K lists 'Data center HVAC systems', 'Data center liquid cooling solutions', 'Data center facility controls' and 'Data center services' among its principal products and services, alongside 'Thermal energy storage' and 'Smart and AI-enabled services'. Those four entries are the only occurrences of the phrase in the entire filing: there is no data center revenue figure, no ranking, no named competitor and no separate reporting - segment disclosure stops at Americas, EMEA and Asia Pacific, and the MD&A attributes the Americas' 7.4% organic growth to 'realization of price increases and higher volumes led by strong demand within our Commercial HVAC business' without naming an end market. | GitLab sits at the software-tooling layer, not compute or models. It consumes third-party LLM capacity — marketing "LLM neutrality and support for self-hosted AI gateways, including air-gapped environments" — so it captures no model-layer economics and bears inference as COGS, visible in the $18.4 million rise in hosting costs. Its distinctive position is the regulated perimeter. | Layer-4 wafer-scale accelerators plus an inference cloud — hardware vendor and specialized cloud in one node. |
| Products (share / barrier) |
|
|
|
| Long-horizon vote | +0.13 at weight 0.20 · swarm neutral Editorial prior, not backtested. | +0.06 at weight 0.20 · swarm neutral Editorial prior, not backtested. | +0.20 at weight 0.20 · swarm bearish Editorial prior, not backtested. |