Autodesk Inc · Moat analysis

Report previewADSK.USReport date Last updated
← Back to reports
Report contents
01

Industry landscape

Where the company competes: three narrow niches, one global standard

Autodesk sells the software that professionals use to design and document physical things. The core market is professional design authoring: 2D drafting and 3D modeling applications that architects, civil engineers, mechanical engineers and animation studios use to turn concepts into drawings, models and construction documents. The company reports across three arenas — architecture, engineering and construction (AEC), manufacturing, and media and entertainment — and each has its own rivals. Autodesk sits in the middle of the value chain: it buys cloud capacity and a small number of software components upstream, then sells subscriptions to design firms, contractors, building owners, manufacturers and studios downstream. The arena is global and Autodesk's file formats act as interchange standards, but this is not one market. It is three adjacent markets, and the barriers differ sharply among them.

Geography is broad and fairly even. On the company's regional revenue disclosure, the Americas supply just under half of revenue, EMEA close to 30%, and Asia-Pacific the remaining fifth or so. That matters because the buyer is local — a 30-person architecture practice, a regional contractor, a mid-size machine shop — while the product standard is global. A Revit model created in Chicago has to open in Manila. Global reach is what lets a subscription business of this kind spread fixed research and development across a large paying base, and it is why Autodesk's scale shows up in gross margin and free cash flow rather than in unit volume.

Competition is fragmented by application rather than by vendor. Few companies meet Autodesk across all three arenas; most meet it in one.

CompanyWhere it meets AutodeskAnnual revenue, latest reported fiscal year
AutodeskAEC, manufacturing and media design software$6.1bn (FY2025, ended January 31, 2025)
Dassault SystèmesManufacturing CAD and PLMabout €6.6bn (2025)
Siemens Digital Industries SoftwareManufacturing CAD, PLM and simulationnot separately disclosed
PTCManufacturing CAD and PLMabout $2.5bn (FY2025)
Bentley SystemsInfrastructure design and digital twinsabout $1.5bn (2025)
NemetschekAEC design (Allplan, Archicad)about €1.1bn (2025)
TrimbleConstruction and geospatial softwareabout $3.6bn (2025)

Revenue scale is a rough proxy for position, not a share measure: none of these companies is a pure play on the same submarket as Autodesk, and no consistent published share series exists for design software. IDC and Gartner cover pieces of the field, such as product lifecycle management and engineering software, but not one comparable CAD or building information modeling series. Autodesk itself does not publish share.

Where the field narrows is worth stating precisely. In BIM authoring — the tools that produce a coordinated 3D building model — Revit is the default in most English-speaking markets, and the credible alternatives in a large commercial project number fewer than five. In infrastructure design, Bentley is a serious rival with its own installed base. In manufacturing CAD, Autodesk is the challenger: Dassault, Siemens and PTC all sell at comparable or larger scale, and cloud-native entrants price aggressively. In media and entertainment, Maya is one of several animation packages competing with Unity, Epic and SideFX.

Applying the practical entry-barrier test to the whole company overstates the problem. Across all three arenas, credible competitors run well past five, which normally signals weak barriers. Within the niches Autodesk leads, the leaders can be counted on one hand, and the barriers look high for four concrete reasons:

  • File formats. DWG and RVT are what subcontractors and fabricators have to accept; converting a project to another format breaks links, drawings and data.
  • Trained labor. Architecture and engineering schools teach Autodesk tools, so new hires arrive productive on them, and retraining a whole office is expensive.
  • Ecosystem. Thousands of third-party plugins and content libraries target Autodesk formats. A challenger has to rebuild that support layer to win large projects.
  • Workflow entanglement. The design file now feeds estimating, fabrication and, increasingly, building operations, which raises the cost of leaving each year.

The soft spots are equally specific. Low-end 2D drafting is contested by inexpensive clones and by domestic vendors in China and India. Manufacturing CAD is the hardest fight, where Autodesk's Fusion partly cannibalizes its own Inventor installed base. And the construction management layer above design is led by Procore, not Autodesk, which leaves the company selling into workflows its software does not own.

What would change this assessment: evidence that new construction starts migrate to a non-Autodesk authoring format, or that generative design shifts work to a cloud platform where file compatibility matters less. The proper benchmark here is Autodesk's own share trend, and the honest gap is that no reliable public series tracks it. Until one exists, the defensible claim is narrower than the product breadth suggests: strong positions in BIM authoring and drafting, a contested position in manufacturing, and a niche position in media.

02

Competitive advantage

Does the company have a moat? Yes, and it holds where Autodesk sets the file standard

Autodesk's economics look structurally protected, not just well run. Two tests settle the question: whether its submarkets stay concentrated enough to keep entrants out, and whether above-normal profits persist instead of getting competed away. Autodesk clears the profitability test with room to spare and the concentration test in two of the three arenas it sells into, which is why the moat is real but uneven.

The record covers five fiscal years, each ending January 31. Revenue rose every year, gross margin stayed inside a 1.5-point band, and operating margin nearly doubled.

Autodesk reported results, USD bnFY2022FY2023FY2024FY2025FY2026
Revenue4.395.005.506.137.21
Operating income0.620.991.131.371.79
Operating margin14.1%19.8%20.5%22.3%24.8%
Net income0.500.820.911.111.12
Net margin11.4%16.4%16.5%18.1%15.5%
Gross margin89.6%89.5%89.8%90.6%91.0%
Free cash flow1.472.021.281.512.41

The one soft spot is FY2026, when net margin fell to 15.5% while operating margin rose to 24.8%. Tax explains it: the provision jumped to $0.48 billion from $0.27 billion. That is a tax line, not a pricing problem, and cash generation improved anyway, with free cash flow of $2.41 billion on $7.21 billion of revenue.

Market position: Fairly Strong, in the niches that set the format

Autodesk does not publish market share, and no third-party series covers design authoring as a whole. IDC and Gartner track product lifecycle management and adjacent engineering software, not a comparable CAD or building information modeling index. Position has to be argued from buyer behavior, and the behavior is informative.

In BIM authoring, the tools that produce a coordinated 3D building model, Revit is the default on large commercial projects in most English-speaking markets, and the credible alternatives on a single project number fewer than five. DWG, the company's 2D format, is what subcontractors and fabricators must accept. These are interchange standards, which is a stronger position than a high share in a market with easy substitutes. Manufacturing CAD is the opposite case: Dassault Systèmes, Siemens and PTC all sell at comparable or larger scale, and Autodesk's Fusion partly competes with its own Inventor base. Media and entertainment is a niche among several animation packages.

For stability, the usual test would be five-year share movement: a swing of more than five percentage points would suggest weak barriers, and less than two would suggest strong ones. No such series exists here, so the proxy is revenue and margin. Revenue compounded at roughly 13% a year from FY2022 to FY2026 with no down year, gross margin moved less than 1.5 points, and no rival has displaced Revit as the default on major projects. Fairly Strong. The leading niches look close to dominant; the portfolio as a whole does not.

Profitability: Strong, at 21% to 25% returns on capital

Reported ROE flatters the business. It was 49.7% in FY2025 and 36.8% in FY2026, but equity is thin by design: $3.04 billion of equity against $4.29 billion of goodwill at January 31, 2026, after years of buybacks. Return on invested capital is the fairer measure. Using reported operating income, the reported tax provision and invested capital of equity plus debt, after-tax ROIC ran about 21% to 25% in FY2023 through FY2025, with FY2026 in the same range. That is an estimate built from reported figures, but the inputs are audited and the level sits at the top of the 15%-25% band that signals a durable advantage. Returns earned through a $4.29 billion goodwill balance are returns after paying for acquisitions, not before.

Scale supports that and does not by itself create it. FY2026 revenue of $7.21 billion is several times Bentley Systems' and Nemetschek's and roughly three times PTC's, though below Dassault's. Strong.

Overall rating: Strong, with manufacturing as the thin flank

Both tests pass where Revit and AutoCAD set the format: fewer than five credible BIM authoring alternatives on a large project, and after-tax returns on capital well above the level that marks a protected franchise. What keeps the call from being unambiguous is that no published share series confirms stability, and manufacturing CAD is a three-way fight against larger vendors with their own entrenched formats. On balance, Strong.

Where the advantage comes from and how long it can last: switching costs first, scale second

Two mechanisms hold Autodesk's position, and they reinforce each other. Customers stay because leaving costs time and rework, and that captivity is what turns the company's scale into an entry barrier instead of just an efficiency edge. Supply-side economics contribute little. Autodesk buys cloud capacity and a handful of software components at market prices, and the 91% gross margin in FY2026 reflects price and product mix, not unusually cheap inputs. A well-funded rival can match the cost base. It cannot quickly replicate the installed base.

Customer captivity: Strong, built on file formats and trained labor

The lock-in works through four steps.

First, the file. A DWG drawing or a Revit model is a live dataset that consultants, fabricators and contractors open, link and revise for years, not a document a customer reads once. Converting a large project to a rival's format breaks links between drawings, schedules and models, and each break has to be repaired by someone with the right skills. The cost scales with project complexity, which is why the biggest projects are the stickiest.

Second, trained labor. Architecture and engineering programs teach these tools, so new hires arrive productive on them. A firm that switched would retrain an entire office and absorb a productivity dip on live work. That is an organizational cost, and it recurs every time the firm hires.

Third, the add-on ecosystem. Thousands of third-party applications and content libraries read and write Autodesk formats. A challenger must persuade those developers to rebuild for a new platform before it can win large customers, and developers follow users. That sequencing problem is harder than writing competitive software.

Fourth, accumulated data. Standards libraries, family components and revision histories sit inside the format. Moving them is a project in itself, and the value of the library grows the longer a firm uses the tool.

The pricing test is what matters. Gross margin stayed near 90% across five years while subscription prices rose and revenue compounded at 13% a year. Captivity only counts as a moat if it produces pricing power, and so far the evidence shows it does. The exceptions are specific: low-end 2D drafting, where inexpensive clones are good enough, and manufacturing CAD, where Dassault's CATIA and Siemens' NX are at least as entrenched. Strong, with those two qualifications.

Economies of scale: Fairly Strong, and sharper because customers stay

Relative share is what counts, not absolute size. Autodesk is the largest vendor in BIM authoring by a wide margin and a challenger in manufacturing, so the scale benefit is concentrated exactly where the captive base is.

The cost structure amplifies the lead. Development, quality assurance and product management are largely fixed, so each additional subscriber carries almost no incremental cost, which is why gross margin sits above 90% and free cash flow reached a third of revenue in FY2026. A smaller rival has to fund the same product breadth from a smaller base, and it has to fund it while cutting price to win customers.

The ecosystem makes that harder still. More users attract more plugin developers; more plugins make the platform more useful; that brings more users. Breaking that loop requires winning users and developers at the same time, and neither side moves first. Education feeds the same machine, because design schools teach these tools and graduates pull them into their employers. That is distribution no start-up can buy quickly, though it also means the company must keep spending to stay in the curriculum.

Two limits are worth stating. Scale advantages are weakest where the customer's format is not the industry's, which describes manufacturing CAD. And scale offers no protection against a change in what customers need to produce, only against rivals producing the same thing more cheaply. Fairly Strong.

Durability: the substitute risk is a different way to design, not a rival seat

Entry barriers protect against new vendors selling the same product. They do not protect against customers solving the problem differently, and that is where the credible threat sits.

Generative and AI-assisted design is the one to watch. If the deliverable shifts from a hand-built coordinated model to a machine-generated design plus a specification, the value of a native file library falls, and a cloud platform that produces the output could matter more than which format it was authored in. That would not require a competitor to beat Revit on features; it would make the contest irrelevant. Low-cost drafting clones in China and India already cover the simple end of the market, and Blender and real-time engines such as Unity cover parts of media production at no license cost.

Above the design tools, the construction management layer is led by others, so Autodesk sells into a workflow it does not own. That limits how much of a project's software spending it can capture, though it does not erode the authoring position.

The format lock should hold for several years because the switching cost is organizational, and organizations change software slowly. The evidence so far only shows that no substitute has displaced Revit on large projects and that price increases have stuck. What would change the assessment: a major market moving new construction starts to a non-Autodesk authoring format, plugin developers porting in volume, or generative tools that make the native file incidental. None of those has happened yet. The brand, the balance sheet and access to capital are not the moat; the switching cost is.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Continue the comparison

Continue reading the full report

  1. Comparison and summaryHow Autodesk Inc compares with its closest rivals, and our moat rating

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on Autodesk Inc