Competitive Moat

Competitive Moat

Gartner's advantages are real and show up in the numbers: a 77% contribution margin on its Insights subscriptions, ~3–4% price increases taken every year for more than fifteen, and retention that beats its only pure-play peer on every measure while the business runs roughly eighteen times that peer's size. But the moat has clear limits — it rests on annual contracts, roughly one-in-seven logo churn, "limited barriers to entry" by the company's own account, and a sharpened, self-disclosed risk that large language models erode the value of the content.

What the moat is made of

The franchise is built on scale and reuse. Gartner is in steady contact with over 13,000 client enterprises, fields more than 2,400 business and technology experts, and logged more than 510,000 direct client interactions in 2025; 77% of Insights contracts are multi-year [1]. The company describes its differentiators as the breadth and quality of its research, a brand carried under the Gartner name for more than 40 years, a global footprint across roughly 90 countries, and the operating leverage of distributing one body of intellectual property across subscriptions, conferences, and consulting [2].

That reuse is what a moat looks like in the margin line. The Insights segment converts 77% of its revenue into gross contribution — the same rate in 2025 and 2024 — because the marginal cost of serving one more subscriber the existing research library is close to zero [3]. Conferences (50%) and Consulting (34%) are respectable businesses, but the economic moat lives almost entirely in the subscription.

Source: FY2025 Annual Report (Form 10-K), segment results [4].

Sources: FY2025 10-K, Item 1 Business and segment results [5] [6]; Q3 FY2024 earnings call [7]; Q1 FY2026 earnings call [8].

The switching cost is behavioral, not contractual. A licensed user builds Gartner into how decisions get made — the Magic Quadrants, the analyst inquiries, the benchmarking, and the 100,000-plus executive peer community [9]. Management's clearest evidence for stickiness is engagement-linked: clients who used AskGartner, its new AI search layer, renewed at "substantially higher" rates than those who did not, even at the same engagement levels [10]. But the contracts are annual, and client retention of about 85% means roughly one in seven logos turns over each year [11]. The moat has to be re-earned every twelve months.

Pricing power, and its ceiling

Gartner has raised prices every year, on a set schedule, for a long time. The CFO put the mechanism plainly: most price adjustments take effect on November 1, the 2025 increase was "just under 4" percent, and the company has "typically raised prices between 3% and 4% for over 15 years" — with the stated purpose of offsetting wage inflation [12]. That is durable pricing power, and few subscription businesses can point to a fifteen-year record of it.

It is also its ceiling. Inflation-matching price increases are a defensive tool, not a growth engine, and they frame what the retention data means. When GTS wallet retention fell to 96% and GBS to 99% in 2025 — the first sub-100% readings in years — it happened despite a ~4% list-price increase, which means the underlying seat and module base contracted by close to a tenth on a like-for-like basis [13]. Pricing held; volume did not. The moat protected the dollar-per-seat, not the number of seats — a distinction the Contract Value Stall traced through the federal cliff and the commercial book.

The competitive set is thin, and Gartner is winning it

The peer screen surfaces six names — Forrester, Information Services Group, Informa (which owns Omdia), S&P Global (451 Research), Clarivate, and Thomson Reuters. Only one, Forrester, runs the same model: syndicated technology-and-business research, analyst advisory, Magic-Quadrant-style vendor evaluations, and an events business [14]. The others are conglomerates in which research is a sliver, competing at the edges rather than head-to-head.

Against that one true comparable, Gartner is not losing share — it is pulling away. Forrester's research contract value fell 6% to $292.4 million in 2025 [15], its client base shrank 7% to 1,797 [16], and it posted a net loss of roughly 30% of revenue after a goodwill impairment [17]. Gartner's Insights contract value, roughly eighteen times larger at $5,155 million, still grew 1% [18]. On both retention measures Gartner sits well ahead.

Sources: Gartner FY2025 10-K, Insights segment [19]; Forrester FY2025 10-K, contract value [20] and key metrics [21].

The read that follows is important: the softness in the numbers is not a rival taking Gartner's customers. Forrester is stalling harder, and the same pattern shows in its metrics — logos holding while wallets thin (its client retention actually rose to 77% as wallet retention fell to 87%) [22]. The pressure on the category is coming from somewhere other than competitors. Gartner's own filing names the direction it expects it to come from: "We anticipate encountering more competition with increased adoption of AI services," alongside the candid admission that "limited barriers to entry exist in the markets in which we do business" [23].

The AI question cuts both ways

For now, AI is a demand driver. It is Gartner's single highest-demand topic; in 2025 the company held more than 200,000 in-depth client conversations on AI, answered more than 500,000 AI questions through AskGartner, and built a library of over 6,000 AI documents [24]. The pitch is that Gartner sits between the buyers and sellers of AI and offers independent guidance neither side can — "No one else does what we do at our scope and scale" [25]. On the current evidence, more AI in the world means more questions Gartner gets paid to answer.

The same technology is also the clearest structural threat the company discloses, and the FY2025 language is sharper than prior years. The 10-K now warns that disruptive technologies including AI "could affect the nature of how we generate revenue," that LLMs surfacing substantive content "could reduce the need to enter our websites," and — most directly — that "although generally contractually prohibited, clients or others may load our proprietary information into large language models, which could reduce the value of our offerings" [26]. Gartner's response is to build with the technology rather than around it: AskGartner launched in August 2025, now updates every two weeks, and answers in 25 languages [27].

Management is watching the substitution question closely enough to instrument it. The CEO described asking every salesperson to log any instance of a client mentioning "using AI as a substitute," routed through a help desk so the company can track whether clients are "considering AI instead of Gartner" [28]. That a company builds a monitoring pipeline for a threat is not proof the threat is materializing — but it is a fair measure of how seriously the risk is taken internally. So far the displacement does not show in the hard data: client retention held or rose, and the retention pattern is category-wide rather than Gartner-specific.

The Durability Read

Two things would move the read toward wide. If wallet retention re-crosses 100% as the federal drag anniversaries out, and if the AskGartner engagement-to-renewal link holds up across a full renewal cycle, the case is that AI deepens the moat by making the library more usable and more essential. Two things would move it toward eroding: client retention breaking below its long 83–86% band, or new-business generation drying up — either would signal the product is being displaced rather than merely trimmed. Neither has happened yet. Whether the commercial book compounds after the cycle turns remains a live question the retention data raises but cannot yet settle — and it feeds directly into whether the ~75% derating priced a cyclical stall or a structural fade.