Gartner, Inc. (NYSE: IT) — A Forty-Year Research Monopoly Priced Like a Melting Ice Cube

Independent Equity Research · July 3, 2026

Except for the clearly-labeled “Claude’s Take” block below, this article carries no investment recommendation and no price target; it discusses valuation only as embedded expectations and scenarios.

⚡ Claude’s Take

This is the author’s own independent opinion and general information only — not investment advice. The analysis that follows is deliberately position-free and carries no price target.

Verdict: HOLD / accumulate-on-weakness — a genuinely elite franchise at a genuinely washed-out price, but the knife is still falling and the one number that matters has not yet turned. Start scaling in sub-$130; back up the truck toward the ~$95–115 bear zone; do not chase a bounce. Not a short. Conviction: medium.

Gartner is one of the highest-quality business models in the public market — a ~68% gross-margin, ~22% ROIC, near-100%-recurring syndicated-research annuity that writes its research once and sells it to 13,000+ enterprises at close to zero marginal cost, throws off $1.1–1.2B of nearly-uncapexed free cash flow, and has retired a quarter of its shares in five years. It has been cut −75%, from $551.80 (Nov 2024) to ~$136, into its cheapest-ever multiple on both earnings and sales (P/E ~13.5x GAAP / ~11.5–12x on normalized EPS of ~$11.5–12; ~8x EV/EBITDA; ~12% free-cash-flow yield). The market has reclassified it out of the S&P-Global/Moody’s/Verisk data-compounder bucket and into the low-growth IT-services bucket. On the numbers, the pessimism is extreme: a reverse-DCF says the price embeds roughly −2% perpetual FCF decline for a business that still grew revenue +3.7% and whose logo retention (85–86%) never cracked. That is the setup value investors dream about.

So why only HOLD? Because the break is real, not imagined, and it has not stopped getting worse. Contract Value — the leading indicator that becomes next year’s revenue — decelerated every quarter of 2025 to +1% FX-neutral and was still +1% in Q1-2026; wallet (dollar) retention fell below 100% in both channels (GTS 102%→96%); and management is shrinking its own sales force (−1% to −3% quota-bearing heads) — a demand read you cannot spin. The bull case (federal/DOGE is a one-time reset now in the base; AI substitution shows up in essentially zero deal logs; downselling not defection) is plausible and evidence-backed — but it is a hypothesis about 2026–27, not yet a number. The factor tape agrees: this is a positive-Quality, deeply-negative-Momentum name — washed-out quality, not trash — but a −0.65 momentum stock keeps getting sold until CV inflects. And the insiders, tellingly, bought zero shares personally through the entire collapse; the only buyer was the buyback. Framing: quality-at-a-washed-out-price / falling-knife hybrid — you are early, and you should be paid for being early. I want a bigger margin of safety than $136 offers against the genuine risk that FY2025 was step one down a structural staircase rather than a cyclical trough. Accumulate into weakness, size for a second leg down, and let the CV print — not the narrative — tell you when the trade is de-risked.

• What flips me decisively bullish: two consecutive quarters of re-accelerating CV (ex-federal back toward high-single-digits) with wallet retention back above 100% — proof the trough held.
• What flips me bearish: a second full year of sub-2% CV with wallet retention still <100% and AI substitution finally appearing in the deal-loss data — confirmation of structural erosion, at which point 13.5x is a fair multiple on a shrinking annuity, not a cheap one.

📈 Stock Price Action — Five-Year Event Map

Text-only by design. Every price move is a Fact; every attributed cause is Interpretation. No price target, no recommendation, no chart-pattern or support/resistance levels.

The arc (Fact). Over five years Gartner ran from the low-$150s (early 2021) to an all-time high of $551.80 on 2024-11-13, then round-tripped violently to a 2026 low of $125.73 (2026-06-22), closing $136.32 (2026-07-02) — −75.3% off the high, essentially at its −77% lifetime maximum drawdown. The 52-week range is roughly $126–$400; the stock sits below its 21-, 50- and 200-day moving averages (~$140 / ~$149 / ~$203). The entire post-2024 move has been one direction.

Cycle narrative. Events 1–4 are the up-leg: Gartner compounded from ~$152 to a $551.80 ATH on post-COVID CV re-acceleration, expanding margins, and a multiple that peaked near 30–40x earnings in late 2024. Event 5 is the fracture — a single −27.6% session on 2025-08-05 when the Q2-2025 report crystallized fears of decelerating Contract Value, the DOGE-driven federal cuts, and AI substitution (FactorsToday logs a ~−25% stock-specific move that day — this was idiosyncratic, not a market event). Events 6–7 are the grind lower: from ~$244 to a $125.73 trough by June 2026 as the overhang persisted and no CV inflection arrived to arrest it, compressing the multiple to the cheapest in the company’s history. The mispricing/opportunity judgment on this arc belongs to Claude’s Take above; this block only records what happened and why.

1. Executive Summary

Gartner is the dominant global franchise in IT-and-business research and advisory — ~$6.5B of revenue, of which the Insights subscription segment ($5.1B, 78% of revenue) earns a 77% gross contribution margin and is essentially the entire economic engine; Conferences ($645M) and Consulting ($553M) are lower-margin adjacencies monetizing the same analyst bench. The model is genuinely excellent: research written once and sold to 13,000+ client enterprises, ~68% company gross margin, ~22% ROIC, minimal capex, and $1.1–1.2B of annual free cash flow. Its moat is a rare three-layer stack — brand/“cover value” (the 40-year default reference; “nobody got fired for citing Gartner”), scale economies (a 2,400-analyst bench amortized across a client base ~13x its nearest independent peer, Forrester), and demand-side habit/switching costs (85–86% logo retention).

The stock has nevertheless been cut −75% because, for the first time in a decade, the leading indicators broke. Contract Value grew only +1% FX-neutral in 2025 (GTS flat, GBS +3%) and remained +1% in Q1-2026; wallet retention fell below 100% in both channels (GTS 102%→96%, GBS 106%→99%); and the company cut its own quota-bearing sales force. The causes are a genuine mix: a one-time U.S. federal/DOGE reset (less than half of end-2024 federal CV was retained; federal fell from ~$275M toward ~$114M), a tariff/macro budget freeze, tech-vendor retrenchment, and an AI-substitution narrative that has crushed the multiple even though management reports it appears in “a very, very, very small number of transactions.”

The reported earnings line is nearly useless in both directions and must be normalized: GAAP diluted EPS of $9.65 (FY25) is depressed by a $150M Digital Markets goodwill impairment, while FY24’s $16.00 was flattered ~$5/share by a $300M event-cancellation insurance gain and a ~$162M discrete tax benefit (9.6% tax rate). Normalized EPS is ~$11.5–12, growing mid-single digits. Capital allocation is a single-minded, dividend-free buyback machine that has retired ~24% of shares in five years ($6.0B, funded conservatively at ~1x net leverage) — intelligent in form, but with much of the $6B deployed at $250–500 (now underwater) and, tellingly, zero personal insider buying through the entire crash.

At ~$136 the stock trades at its cheapest-ever P/E and P/S, ~8x EV/EBITDA, ~12% FCF yield — a reverse-DCF implying roughly −2% perpetual decline. The entire investable question is binary: is FY2025 a cyclical air-pocket (federal one-off + macro freeze, with a durable moat intact) or the first step of a structural, AI-driven impairment of the syndicated-research model? The FY2025 data cannot adjudicate it; FY2026 Contract Value will. This memo lays out the evidence on both sides and takes no position (see Claude’s Take for the labeled exception).

2. Business Overview

What Gartner does. Gartner sells syndicated research, one-to-one advisory (“inquiry”), conferences, and consulting to corporate and government decision-makers. Founded in 1979 and now ~$6.5B in revenue, it is the largest of the independent research houses. It employs 2,400+ analysts who produce proprietary research, benchmarks, and tools, and who take client calls to de-risk major decisions (which software to buy, how to structure an IT organization, where a market is heading). Gartner is “in steady contact with over 13,000 distinct client enterprises” and logged 510,000+ direct client interactions in 2025 (Fact, FY2025 10-K, Business).

Why the model is beautiful. The economics rest on one structural feature: research is written once and sold to thousands of clients. An analyst note costs the same to produce whether one subscriber or ten thousand read it — the ten-thousandth subscription is nearly pure margin. This is a fixed-cost-leverage / scale-economy model wrapped in a subscription. It shows up directly in the segment margins: Insights earned a 77% gross contribution margin in FY25 and FY24, and company gross margin was 68.4% (Fact, 10-K; ROIC.ai). Marginal cost of an incremental subscription is near zero, so incremental revenue drops through at extraordinary rates.

Segment breakdown (FY2025).

(Source: 10-K MD&A segment tables.) Insights’ ~$3,890M of gross contribution is ~85%+ of all segment gross contribution — Conferences (~$322M) and Consulting (~$188M) are small by comparison. Interpretation: any valuation of Gartner is, at bottom, a valuation of the Insights subscription annuity. Digital Markets (Capterra/GetApp/Software Advice), a structurally weaker transactional business, took a $150M goodwill write-down in 2025 and was sold to G2 for ~$110M, closing 2026-02-05 — a clean refocus on the core.

GTS vs. GBS — the two channels inside Insights. Global Technology Sales (GTS) sells to technology users and providers (CIOs, IT leaders, tech vendors) — the legacy franchise, ~three-quarters of Insights CV. Global Business Sales (GBS) sells to all other functional leaders (HR, finance, supply chain, marketing, legal, sales) — the newer, faster-growing land-grab into non-IT functions, largely built on the 2017 CEB acquisition.

Contract Value (CV) — the KPI that matters. CV is the annualized dollar value of all subscription contracts in force at a point in time. Because Insights revenue is recognized ratably, CV leads revenue — today’s CV becomes next year’s revenue. That is why +1% CV growth (with GTS flat) is so consequential even as reported Insights revenue still rose +5% (harvesting stronger 2024 CV). Recurring vs. non-recurring: ~84% of revenue is recognized over time; ~78% is the Insights subscription itself. Conferences (recognized at event delivery) and Consulting ($173.7M backlog) are the cyclical layers. Headcount was 20,244, but the growth engine went into reverse — GTS quota-bearing associates fell 3% to 3,704, GBS fell 1% to 1,280, with a $56.6M workforce-reduction charge.

Verdict: A genuinely high-quality, high-margin, recurring subscription model with a small tail of lower-margin adjacencies — but one visibly decelerating at the KPI (CV +1%) and retention (wallet <100%) level for the first time in a decade. The business is beautiful; the momentum has broken.

3. Industry Dynamics

The industry. Gartner sits at the apex of the IT-and-business research & advisory industry — producing independent, syndicated analysis and one-to-one advisory that helps enterprises make large, uncertain technology and operating decisions. The top of the market is shorthanded as “the FIGs” — Forrester, IDC, and Gartner — the three at-scale, vendor-neutral independents. Adjacent sit specialist advisory/benchmarking firms (ISG, Everest, HFS, Constellation), the strategy consultancies on the high end (McKinsey, Bain, BCG), integrator/reseller advisory arms (Accenture, Deloitte), and — the newest and most-debated — free and AI-generated substitutes (search, vendor content, LLM assistants). The 10-K names all of these, and now explicitly flags competition from “large language models” (Fact, 10-K, Competition + Risk Factors).

Market structure — concentrated at the top, fragmented at the fringe. The scale gap between Gartner and its nearest independent peer is enormous and widening:

(Sources: Gartner 10-K; Forrester FY2025 release; IDC estimate.) Gartner’s Insights CV alone ($5.16B) is ~13x Forrester’s total revenue. More telling: while Gartner’s CV grew +1% in a hard year, Forrester’s revenue shrank 8% and its CV shrank 6% — the scale leader gained relative share even in a downturn, exactly what a scale-economy framework predicts.

TAM and growth. The addressable market spans enterprise IT/technology research plus the GBS non-IT adjacencies plus conferences and consulting; management frames a “$200B+” total opportunity of which syndicated research is a low-single-digit-billions penetrated slice. Interpretation: the structural growth rate of enterprise willingness-to-pay for third-party advisory is probably GDP-plus in normal times, with Gartner’s historical mid-teens growth coming from share gains + price + seat expansion + the GBS land-grab, not a fast-growing pie. FY2025 exposed the vulnerability: when budgets froze (tariff uncertainty, macro caution) and the federal buyer was actively cut (DOGE), the algorithm collapsed toward flat. This is a discretionary, budget-sensitive spend — first cut in a freeze, rarely cancelled outright (hence logo retention held while wallet retention fell).

Greenwald tests. Barriers to entry are high for scale, low for niche — replicating Gartner’s coverage, brand, and analyst bench from scratch is prohibitively slow (no one has in 40 years), but niche entry is cheap (why the fringe is crowded and LLM substitution is a low-end threat). Capital intensity is negligible (people + IP, ~$118M depreciation on $6.5B revenue), so ROIC is structurally high (~22–25%). Pricing power is real but conditional — low-to-mid-single-digit annual increases stuck for years, until FY2025’s sub-100% wallet retention proved pricing power depends on the customer’s budget environment.

Marathon capital-cycle read. High returns normally attract capital, competition, and mean-reversion — but in this industry the binding input is credible senior-analyst talent and brand trust, which cannot be manufactured with capital quickly. That intangible barrier is why decades of Gartner’s high returns never summoned a credible new at-scale competitor. The genuine risk to the capital cycle is not a well-funded entrant; it is a technology shift (generative AI) that lowers the replacement cost of the underlying “answer,” letting compute substitute for the analyst bench. If that holds, the cycle bites from the technology side, not the balance-sheet side.

Verdict: structurally a good industry — for the scale leader — but at its first genuine inflection in a decade. A high-return, low-capital, high-barrier oligopoly where Gartner is ~13x its nearest independent peer and gaining share. But attractiveness is now contingent on two open questions: is the demand freeze cyclical, and does generative AI relax the talent/IP supply constraint that has historically blocked the capital cycle?

4. Competitive Position

Naming the moat. Gartner’s advantage is a rare three-layer stack in Greenwald’s taxonomy:

1. Intangibles / brand (“cover value”). Gartner is the 40-year default reference for enterprise technology decisions; its Magic Quadrant is cited in board decks, RFPs, and vendor marketing. The value is not only information but institutional cover — “we chose the Leader in the Gartner Magic Quadrant” is a career-safe justification. The classic “nobody got fired for citing Gartner.” The 10-K leads its own differentiation with exactly this: “known for its high quality, independence and objectivity… for more than 40 years.”
2. Economies of scale + high fixed cost. A 2,400-analyst bench amortized across 13,000+ clients — Forrester’s entire revenue ($397M) is smaller than the incremental Insights revenue Gartner adds in a good year. Result: the 77% Insights gross contribution margin. No sub-scale rival can match both coverage breadth and price.
3. Demand-side captivity / switching costs & habit. Analyst-inquiry relationships, benchmark data feeds, and seat-based access are woven into how teams make decisions; switching means losing continuity, historical benchmarks, and the “cover” of the standard reference. Habit/search-cost captivity (contracts are annual), which is why logo retention runs a durable 85–86%.

Does the moat show up in the P&L? The moat is falsifiable through the numbers, which is the right test. If it were illusory you would expect: gross contribution margin far below the mid-70s (it is 77% and stable); logo retention collapsing (it holds 85–86%); no pricing power (historically strong); and ROIC compressing toward WACC (it is ~22%, multiples above). Three of four still hold firmly. The one that cracked is pricing power — wallet retention fell to 96% (GTS) / 99% (GBS) in FY2025, the single most important datum in this section: the moat’s pricing-power leg is, for the first time in a decade, not holding.

Pressure-testing the Magic Quadrant vs. analyst-inquiry. The honest answer: the Magic Quadrant brand is real moat; the analyst-inquiry function is the part exposed to AI. The Magic Quadrant is a coordination/standard good — valuable because everyone treats it as the standard, a self-reinforcing network property an LLM cannot confer (an LLM can summarize a market; it cannot grant the industry-agreed “cover” of a Leaders quadrant). But much of what clients historically paid for was analyst inquiry — asking an expert a question — precisely the task generative AI does cheaply and instantly. The moat is not monolithic: its brand/standard leg is AI-resistant; its ask-an-expert leg is AI-exposed.

The central debate — does generative AI structurally impair the moat?

Bear (eroding). The core service — paying five/six figures a year for analysts who answer questions and summarize markets — is the exact LLM use case. If a CIO gets an 80%-good answer instantly and free, willingness-to-pay for the marginal inquiry falls, seats get cut, price increases stop sticking. The FY2025 numbers are consistent with this: wallet retention <100%, CV +1%, and Gartner shrinking its own sales force. The 10-K concedes AI “could affect… how we generate revenue” and that its content is “exposed to… large language models.”

Bull (durable). Three things an LLM cannot easily replicate: (1) proprietary, current, practitioner-sourced data — benchmarks, spend/salary data, and peer insights from 510,000 annual interactions across 13,000 enterprises, primary data not on the public internet; (2) the “cover”/accountability value — a $50M platform decision needs a citable, independent, accountable authority (“ChatGPT told me” is not board-defensible), and LLMs’ confident-but-wrong tendency raises the value of a vetted source; (3) distribution + trust compounding — Gartner is embedding AI into its own delivery (AskGartner at 100% of license users; publishing time −75%), using LLMs to surface its proprietary corpus rather than being disintermediated. On this view, strip out the double-digit federal hit and the tariff freeze and the underlying commercial book still grew mid-single-digits — an AI panic wearing a macro air-pocket’s clothes.

Weighing it. What is Fact: the pricing-power leg cracked (wallet retention <100%) and management is behaving cautiously (sales cuts). What is Interpretation: whether that is cyclical or structural. The most defensible read: the brand/standard/proprietary-data core is durable; the average-inquiry revenue layer is being commoditized at the margin; and the cycle is masking which force dominates. The decisive evidence is FY2026 CV — a re-acceleration off +1% would confirm the cyclical read and the moat; a second sub-2% year with wallet retention still <100% would confirm structural erosion. No thesis claiming certainty either way is honest.

Verdict: a durable, genuine, multi-layer moat that has taken its first real financial hit in a decade. Proven by 77% Insights contribution margin, 85–86% logo retention, and ~22% ROIC a commodity business could not sustain — but the pricing-power leg has broken below 100% and generative AI is aimed squarely at its lowest-differentiation service layer. Durable, but on watch; the burden of proof has shifted to FY2026 CV.

5. Growth History and Forward Opportunities

The long arc: an acquisition doubling, then an organic single-digit compounder, now stalled. Two eras. First, transformational M&A: the April 2017 acquisition of CEB (~$3.3B enterprise value) roughly doubled the company (revenue ~$2.4B FY16 → ~$4.0B FY18) and its non-IT functional research became the core of today’s GBS. Since ~2019 the story has been organic and good: revenue compounded from $4.10B (FY20) to $6.27B (FY24), Insights doing the heavy lifting. That engine decelerated sharply — FY25 revenue +3.7% to $6.50B, and FY26 is guided to only +1–2% FX-neutral. This deceleration is the central fact of the thesis.

Contract Value decelerated every quarter and has not turned. Total CV growth (FX-neutral, YoY): Q1’25 +7% → Q3’25 +3% → Q4’25 +1% → Q1’26 +1%. Year-end CV $5.155B, +1%. The bulls’ ex-federal CV is better but also fading: +6% (Q3’25) → +4% (Q4’25) → +3.5% (Q1’26) — meaning the slowdown is not purely federal. By segment: GTS CV $3.910B, dead flat (“decreased spending from existing clients”); GBS CV $1.245B, +3% (new clients, led by supply-chain/legal). GBS remains healthier; GTS (the historic IT-buyer core) is where the damage concentrated.

Retention shows where it broke — downselling, not defection. FY24→FY25, wallet retention collapsed (GTS 102%→96%, GBS 106%→99%) while logo retention was essentially stable (GTS 84%→85%, GBS 87%→86%). Clients are not leaving; they are buying less — cutting seats, downgrading tiers. Management corroborates: “a client who has tight times saying… I’m going to go down to 9 seats” (Hall, Q4’25). A wallet-retention problem is cyclical and recoverable far more readily than logo churn (which would signal moat erosion). The one genuinely negative sub-trend — declining client count on small-tech-vendor churn (logos going out of business) — has minimal wallet impact.

The growth algorithm is being throttled. Gartner grows by adding quota-bearing headcount (QBH) and letting ramping sellers compound. With CV near zero, that engine stalls: 2026 QBH is guided to only low-single-digit GTS / mid-single GBS, biased toward new-logo hunters over account managers, and new business was down ~4% ex-federal through 2025. Management’s counter is that current seat capacity is “ample” for 2026 and the BD-tilted hires are “really about 2027 and 2028” — i.e., deliberately not spending into the downturn, protecting margin. Defensible, but the sales-capacity flywheel is idling.

Forward opportunities — real, but each is a “prove-it.” Five planks: (1) easier federal comps (federal rebased to ~$114M, lapping DOGE from Q2’26 — mechanical); (2) the BTI “transformation” (AskGartner AI at 100% of license users; active-insights library +~50%; publishing time −75%) lifting engagement → retention (Hall: “clients who engage frequently… retain at higher rates”; Q1’26 engagement +~170bps YoY) — real leading indicators, but the payoff lags 12–24 months to renewal; (3) pricing — a normal ~3.5% Nov-2025 increase with little pushback; (4) GBS/non-IT expansion into a large TAM (~140,000 target enterprises, ~14,000 penetrated); (5) a potential future enterprise-license/broader-seat model (floated, not committed). None has yet inflected the +1% headline.

Verdict: a high-quality franchise currently producing low-quality growth; deceleration more cyclical than structural, but the model is more macro-sensitive than the “compounder” narrative implied. Economics remain excellent (77% contribution margin, ~22% ROIC, near-100% recurring, real pricing power) — not a broken business. But growth right now is low-quality: near-zero net CV, sub-100% wallet retention, stalled hiring, declining client count. The weight of evidence says the deceleration is primarily cyclical (logos retained, AI-substitution absent from deal data, federal a one-off), but the structural caveat is real (ex-federal CV halved, 6%→3.5%). Re-acceleration is a credible hypothesis on engagement leading-indicators and easier comps — but as of Q1’26 it is management narrative, not yet a number.

6. Financial Quality

Up front: a high-return, cash-generative, asset-light franchise whose economics do improve with scale — but FY2025 is a clear deceleration, and reported EPS is so distorted by one-time items it is nearly useless. The right lenses are ROIC (~22%), FCF (~$1.1–1.2B), and normalized EPS (~$11.5–12), not GAAP net income.

Revenue. Compounded ~9.6% over five years ($4,099M FY20 → $6,497M FY25) but decayed every year: +15.7% → +7.9% → +6.1% → +3.7%. FY25’s +4%/+3% FX-neutral is the slowest non-COVID year in over a decade. The tell is the leading indicator: CV +1% FX-neutral and short-term deferred revenue up only +1.7% to $2,810M — both flashing low-single-digit forward growth. Wallet retention deteriorated (GTS 102%→96%, GBS 106%→99%): the average surviving account is shrinking spend, masked at the top line by new logos and price.

Margins — structurally high, modestly compressing.

Gross margin is remarkably stable (~68%) — the mark of selling the same content to many subscribers at near-zero marginal cost. EBITDA margin drifted down ~280bp off the FY21 peak; GAAP operating margin fell to 15.8%, but that is entirely the $150M goodwill impairment inside operating costs — ex-impairment, operating income was ~$1,176M, +1.7%. The underlying trend is near-stall growth with SG&A (+6%) outpacing revenue (+4%) — negative operating leverage, the margin story to watch. FY26 EBITDA margin is guided down to 23.5% (from 24.8%), the “new baseline.”

Quality of earnings — the headline EPS is a trap in both directions. Reported diluted EPS reads $11.08 (FY23) → $16.00 (FY24) → $9.65 (FY25) — a swing with almost nothing to do with the operating business:

• FY24 flattered by ~$5/share: a $300M pretax event-cancellation insurance gain (resolving 2020/21 COVID claims; $0 in FY25) plus a ~$161.9M discrete tax benefit from a Dec-2024 intercompany IP transfer that pushed the tax rate to 9.6% (vs. ~24.7% normal).
• FY25 depressed ~$2/share by the $150M non-deductible Digital Markets goodwill impairment (which actually raised the tax rate).
• FY23 carried a +$135.4M divestiture gain in operating income.

Stripping these out, normalized diluted EPS is ~$9.8 (FY23) → ~$11.0 (FY24) → ~$11.6 (FY25) — real earnings power ~$11.5–12, growing mid-single digits. Anyone valuing IT off reported GAAP EPS mis-underwrites it in both directions.

Cash. CFO was $1,290M in FY25 (1.77x net income) — inflated by the non-cash impairment add-back; on normalized earnings, conversion is ~1.3–1.4x. Capex is trivial (~1.8% of revenue), so FCF ≈ CFO ≈ $1.0–1.2B run-rate ($1,175M FY25). SBC of $155.9M (2.4% of revenue) is low for a name the market files under “IT,” and buybacks more than offset dilution, so FCF/share is genuinely distributable. Accounting is conservative — no capitalized-software games, deferred-revenue growth tracks the +1.7% CV signal (no revenue pull-forward).

ROIC vs. ROE — why book equity is a rounding error. Reported ROE fell to 11.5% (from 20–30%) but is meaningless: book equity is only $319.9M and tangible book is negative because $9,040M of cumulative treasury stock (five years of buybacks) has consumed the equity account. The business is funded by negative working capital (deferred-revenue float) plus debt, not retained equity. The correct metric is ROIC — 21.7% in FY25, ~20–25% for five years — evidence of a genuine capital-light moat. Economics improve with scale: incremental subscription revenue drops through at ~68% gross margin against a largely fixed content base.

Balance sheet. $3.0B senior notes ($800M '28, $600M '29, $800M '30, $350M '31, $450M '35) plus $366M finance leases; undrawn revolver. Against $1,722.5M cash, net debt ~$1.26B ex-leases / ~$1.63B incl. — net debt/EBITDA ~0.9–1.2x, no maturity before 2028. Investment-grade, low-risk; balance-sheet risk is negligible.

Verdict: economics do improve with scale and the cash is real — but the operating engine has decelerated to near-stall (CV +1%, wallet retention <100%), and reported earnings are so noisy they obscure the truth. The debate is not quality; it is the durability of growth and what you pay for a business now compounding EPS mostly via buyback rather than organic expansion.

Six-Year Financial Summary

¹ GAAP operating income includes a +$135.4M divestiture gain (FY23) and a −$150.0M goodwill impairment (FY25); ex-items FY23 ≈ $985M, FY25 ≈ $1,176M. ² Normalized EPS removes the FY23 divestiture gain, the FY24 $300M insurance gain and ~$162M discrete tax benefit, and adds back the FY25 $150M non-deductible impairment. (Source: FY2025 10-K; ROIC.ai, reconciled to filing.)

7. Capital Allocation

Up front: a disciplined, single-minded, buyback-only capital-return machine that shrank the share count ~24% in five years while keeping the balance sheet conservative — textbook mechanics. But whether the $6B deployed created value is unproven: much was repurchased at $250–500, and the stock now sits ~$130–155. The direction is right; the prices paid are the open question.

The buyback is the entire story. Gartner returns capital one way. Cumulative repurchases FY21–25: $6,032M ($1,656M/$1,044M/$606M/$735M/$1,991M), reconciling to the $9,040M treasury-stock balance. No dividend, ever (zero dividend line every year). Diluted shares fell 90.0M (FY20) → 75.6M (FY25); basic shares 88.8M → 70.8M (Dec-25) → 67.5M (Apr-2026) — a ~24% reduction, still running hard into 2026. FY25 was the largest buyback in company history — $1,991M at ~$255–265 (the disclosed Q4-25 window averaged $239.06), funded with cash plus an $800M note issuance; the Board added $500M to the authorization in Jan-2026 (~$1.245B remaining).

The uncomfortable counterpoint. The stock has since fallen to ~$130–155 (Form 4 prices: $154.09 May, $129.62 Jun-30). So the record $1,991M spent in 2025 at ~$260, and ~$3.3B spent in FY21–22 at even higher prices, are currently underwater. Applying Marathon’s capital-cycle test — buybacks create value only below intrinsic value — the verdict splits: the behavior is counter-cyclical and shareholder-oriented, but a material portion of the $6B was deployed at rich prices into a decelerating business, and value-accretion is unproven. “Intelligent capital return” and “financial-engineering a low-growth business” blur here; the honest answer is it looks like some of both.

M&A, spend intensity, debt. Post-CEB (2017, ~$3.3B — the origin of the debt), Gartner has been almost acquisitive-abstinent (FY23–25 acquisition spend ~$0–4M/yr). The one modern blemish is the $150M Digital Markets impairment (a CEB-era asset, since sold to G2). The company does not break out R&D; its “growth spend” is the ~47%-of-revenue SG&A line, overwhelmingly the quota-bearing salesforce. Debt is investment-grade, laddered 2028–2035, revolver undrawn, net leverage ~1x — the buyback is funded from strength.

Incentive alignment — well-designed, tied to the leading indicator (not EPS). The 2026 proxy: short-term bonus = 50% Adjusted EBITDA + 50% Revenue (FX-neutral, excluding the volatile federal book); long-term 100% performance-based (7-year SARs → TSR-aligned, plus PSUs), with Contract-Value growth = 70% of 2025 LTI and the Pay-vs-Performance “Company-Selected Measure.” CEO pay is 88% long-term. Crucially, comp is not tied to EPS or FCF, so the buyback does not mechanically inflate payouts — removing the classic “engineer EPS to hit the bonus” conflict.

Insider read — no conviction buying into the crash. Insider ownership is modest and non-controlling: directors/NEOs/officers (24 people) hold 1,760,321 shares = 2.6%, CEO Hall ~1.19M; institutions dominate (BlackRock 10.8%, Baron 6.4%). Across ~23 sampled Form 4 transactions spanning Oct-2025 → Jul-2026 — the entire ~$240→~$130 drawdown — there were ZERO code-P open-market purchases. Every transaction was routine comp mechanics (grants, vesting, tax-withholding), plus one small sale and one gift. In a ~75% decline, not one insider stepped in personally. The only buyer of IT stock during the crash was the company’s own program — a genuine dissonance worth flagging.

Verdict: disciplined, focused, shareholder-friendly in form — no dividend, no empire-building, relentless share-count reduction, conservative leverage, incentives on the right metric — but more ambiguous in substance: much of the $6B was executed well above today’s price into a decelerating business, and no insider has backed the stock personally through the collapse. Intelligent in process; not yet vindicated in outcome.

8. Changes and Headwinds — Last Two Years

The last two years took Gartner from a ~$550 stock (Nov-2024 ATH) to ~$136 (−75%). The causal chain:

1. The CV deceleration (2025) — the master driver. Total CV growth fell from +7% (Q1’25) to +1% (Q4’25/Q1’26). Management frames the cause as external “market forces” (DOGE, tariffs, state/local funding cuts, non-AI tech softness) creating “increased scrutiny, elevated deal approval authority and extended buying cycles.” The numbers largely support the external attribution (logos retained, pricing held, pipelines up) — demand deferred, not lost — though the ex-federal fade means it is not only external.

2. The U.S. federal / DOGE reset — the largest quantifiable hit. DOGE cuts began March 2025. The 10-K: “Less than half of our December 31, 2024 [federal] Insights contract value was retained in 2025”; federal CV fell to ~$126M (Dec-2025) from ~$165M (Sep-2025) toward ~$114M (Mar-2026); a 250bps headwind to total CV in Q1’26 alone. Mitigant: federal contracts are one-year, “any client that wanted to cancel… will have already had the chance,” so federal is modeled flat in 2026 then growing — largely in the base now, a cleansing rather than an ongoing negative.

3. AI-disruption narrative — the overhang the numbers don’t yet support. The reason the multiple compressed to cheapest-ever. Management pushes back with data: sellers must log every AI-substitution instance, and it appears in “a very, very, very small number of transactions” — the real drags being budgets, federal, tariffs. Gartner positions AI as its #1 demand topic (6,000+ AI documents, 200,000+ AI client conversations in 2025). The AI-substitution thesis is a valuation driver, not yet a fundamental one — but it is the key unfalsified bear risk.

4. Guidance rebased down. No dramatic cut/withdrawal, but FY26 embeds a step-down: EBITDA margin 24.8%→23.5%+ (“the new baseline”), and the headline commitment shifted from the medium-term 12–16% CV objective to a new “12%+ adjusted-EPS CAGR over 3 years” — which, with FX-neutral revenue +1–2%, is driven by buybacks and margin, not organic growth. An honest but telling repackaging: the growth promise migrated from the top line to financial engineering.

5. Portfolio & capital-structure changes — rational. Digital Markets sold to G2 (~$110M, closed 2026-02-05), refocusing on core Insights (a Q3’25 goodwill impairment preceded it); an inaugural investment-grade bond (+~$500M) raised specifically to fund buybacks; ~$2B repurchased in 2025 (−8% shares, ~−10% YoY by Q1’26) leaning into the de-rated stock.

6. Leadership — continuity, not disruption. CEO Gene Hall’s contract extended to end-2031; he assumed Chairman (July 2024). CFO Craig Safian remains. Two new directors added, committee chairs rotated (2025). The only downturn-linked “leadership” change was a BTI staff restructuring in H2’25 that management insists was skills-repositioning, “nothing to do with cost” (treat as Interpretation — a standard euphemism).

7. Conferences — the one segment recovering. FY25 revenue +11% to $644.7M (+8–9% same-conference FX-neutral); 56 destination conferences planned for 2026 (from 53), plus a new C-level communities format. Rising NPS/attendance is management’s cited leading indicator for future CV — a genuine bright spot.

Verdict: the changes weaken the near-term thesis but are, on balance, cleansing rather than value-destructive. Negatives are real and concentrated in 2025 (CV collapse, wallet retention −6–7pts, federal gutted, margin/growth rebased down). But the character matters: the federal hit is a one-time reset in the base; AI-substitution is not corroborated by deal-level or retention data; the portfolio and capital-structure moves are rational and shareholder-friendly; and leadership is locked in through 2031. What the changes genuinely expose is that the subscription base is more cyclically sensitive than its double-digit-compounder reputation suggested — a downgrade to the quality-of-growth narrative, not (yet) to the moat. Net: near-term negative, medium-term neutral-to-constructive, contingent entirely on whether engagement-led re-acceleration lifts CV above low-single-digits in 2026–27.

9. Risk Analysis

The dominant risks are #1/#2/#4 — all facets of the same cyclical-vs-structural question — and #9, the possibility that even fundamental stabilization does not restore a premium multiple. Balance-sheet and total-loss risk are minimal: this is not a solvency story.

10. Valuation Discussion (Embedded Expectations)

Gartner trades at the cheapest valuation in its history on earnings and sales — the single most important input. The question is whether that cheapness is a correct verdict on a structurally impaired model, or a violent overshoot on a franchise having an air-pocket. The multiples don’t resolve that; they frame how much pessimism is priced.

Live multiples (~$136, 2026-07-02). At ~68M shares (falling via buyback), market cap is ~$9.3B; with ~$1.6B net debt (incl. leases), EV is ~$10.9B. Against TTM revenue $6.5B, EBITDA $1.38B, EBIT $1.18B, and firm FCF ~$1.5B:

Note: ROIC’s snapshot EV (~$20.6B) is stale — it embeds the FY25 year-end $252 price; at the live $136 the EV halves and every multiple roughly halves with it. P/B is not meaningful (a decade of buybacks drove book equity to near-zero / negative tangible book), so the own-history read rests on P/E and P/S — the AZI valuation_index places both at the ~0.5th percentile of the stock’s own ~10-year range: the cheapest Gartner has ever traded.

Own-history context is the headline. For a decade Gartner was a serial 26–40x P/E, 20–28x EV/EBITDA compounder — filed alongside S&P Global, Moody’s, MSCI, FactSet, Verisk. It now trades at ~13.5x earnings (~11.7x normalized) and ~8x EBITDA — a ~55–70% de-rating off its own norms and the 2024 peak. Not a pullback; a wholesale reclassification.

Comp set. The information-services quality group — SPGI (~28–32x P/E), MCO (~33–38x), MSCI (~30x), VRSK (~28–32x), FDS (~22x) — carries large premiums despite Gartner’s superior gross margin and comparable ROIC. Accenture (~18–20x P/E) sits well above IT. The only cohort IT is now cheaper-than-or-in-line-with is the offshore/low-multiple IT-services cluster — Genpact and ExlService (~13–16x) — exactly the factor-peer set the positioning model flags. Pure-play Forrester trades at a distressed low-teens multiple and is shrinking. The market has repriced Gartner out of the data-vendor bucket and into the low-growth IT-services bucket — the crux of the debate.

Embedded expectations / reverse-DCF. Solving a Gordon perpetuity backward from the ~$9.3B equity value at a 9% cost of equity, the price implies roughly −1% to −4% perpetual FCF growth ($1.0B anchor → −0.6%; $1.2B → −2.4%; $1.35B → −4.8%). In plain terms, the market is underwriting a business in gentle secular decline — a shrinking annuity — for a company that grew revenue +15.5%/+15.7%/+7.9%/+6.1%/+3.7% across 2021–2025. The trend is unmistakably decelerating (the bear’s best fact), but even the most decelerated print is positive, and the reverse-DCF has already crossed into pricing contraction. The gap between “growth decelerating to low-single-digit” and “terminal decline” is the entire investable variance.

Scenario analysis (embedded-expectations framing — NOT a price target, NO buy/sell). Illustrative value zones from explicit assumptions on ~68–76M shares:

The width of that range (~$80 to $290) is itself the finding: this is a binary re-rating debate, not a valuation quibble. At $136 the market sits nearer the bear anchor, pricing a fair probability of terminal decline. Where at $550 the multiple demanded flawless double-digit CV, at $136 it only needs Gartner not to shrink to look mispriced. (No price target; no recommendation — the body reports embedded expectations only.)

11. Variant Perception

Consensus belief. The tape and sell-side have converged on structural impairment: generative AI erodes syndicated human research, federal/DOGE and tech-vendor retrenchment slow CV, and the flywheel that drove a decade of 26–40x multiples is broken. Published price targets collapsed into a ~$120–164 band — “de-rated and dead money,” not opportunity. The −75% round-trip and negative alpha (−0.44) confirm consensus has fully embraced the bear.

Strongest bull case. The de-rating is a sentiment overshoot on a still-elite franchise: 68% gross margin, ~22% ROIC, $1.1–1.2B FCF, high retention, pricing power, ~1x leverage. AI is at least as plausibly a tailwind — rising technology complexity historically increases demand for independent advisory to cut through vendor noise, and Gartner is the default arbiter. At ~13.5x / ~12% FCF yield with an aggressive buyback shrinking the count, the equity only needs CV to stabilize at low-single-digit growth to compound attractively from a washed-out multiple. The market is paying nothing for re-acceleration and pricing terminal decline in a business that just grew +3.7%.

Strongest bear case. This is a genuine, not cyclical, decline. CV decelerated four straight years and the rate of change is the tell; AI substitutes directly for the “quick-answer” query that anchors seat-based renewals; federal and tech-vendor exposure is structural; a subscription base that stops adding seats de-rates and de-grows simultaneously. In that world ~13.5x is not cheap — it is a fair multiple on an eroding annuity headed lower, and the reverse-DCF’s −2% terminal is optimistic.

The assumptions that matter most. (1) Does Insights CV growth stabilize/re-accelerate or keep decelerating toward zero/negative? (2) Is AI net substitute or complement? (3) Is the federal/tech-vendor weakness cyclical or permanent? (4) Do gross margin and retention hold as the base matures? (5) Does management keep shrinking the count at these prices, converting the FCF yield into per-share value?

Falsification. Bull breaks if: two-to-three more quarters of decelerating CV with sequential wallet/seat-retention declines, or a gross-margin roll below the high-60s. Bear breaks if: CV inflects up for two consecutive quarters, federal churn stabilizes, and net-new-business turns positive.

Factor-positioning read. FactorsToday (2026-07-02, All-Factors) prices IT with a Momentum beta of −0.649 alongside a positive Quality beta (+0.167) and a large “Financial Data Titans” custom loading (+0.568) — the empirical signature of a quality business the momentum crowd has abandoned/is shorting, not a low-quality name riding hype into a fall (which would show negative Quality). Its factor-peers (ACN, Genpact, EXLS) confirm the market now trades it like IT-services. This is the classic abandoned-value / washed-out-quality setup: positioning is already maximally bearish (near the −77% lifetime max drawdown, deeply negative alpha), which caps further positioning-driven downside and sets up asymmetric snap-back if the fundamental floor holds — but offers no cushion if the bear thesis is right. Idiosyncratic vol is elevated (44.9%; R² 0.42), meaning the crash was overwhelmingly stock-specific (Gartner’s own CV story), dominated by the single −27.6% August-2025 session — not a factor or market event.

12. Fact vs. Interpretation

13. Open Questions

1. What was Q2-2026 CV growth? The single most important unknown — did the +1% trough hold, re-accelerate, or break lower? (Reports ~August 2026.)
2. How much of the wallet-retention decline is federal vs. commercial? If commercial wallet retention is near/above 100% ex-federal, the cyclical read strengthens materially.
3. Is AI substitution appearing in renewal/downsell reason-codes as it scales, or staying near zero? Management’s deal-logging data is the tell but is not externally verifiable.
4. Does the BTI engagement lift actually convert to higher retention on renewal over the 12–24-month lag, or is it a leading indicator that never reaches the CV line?
5. Will management keep buying at ~$130 even after $6B underwater at higher prices — and does the authorization get materially expanded?
6. What is the succession plan after Hall (2031)? A 22-year CEO; the bench is not publicly demonstrated.
7. Does the new “12%+ EPS CAGR” target hold if revenue stays +1–2% — i.e., how much buyback/margin is required, and is it sustainable?

14. What Must Be True

For the bull case (the de-rating is a cyclical overshoot):

• Insights CV growth re-accelerates off the +1% trough — ex-federal back toward high-single-digits within 2–3 quarters — as federal annualizes out and budgets thaw.
• Wallet retention returns above 100%, proving the installed base resumes net expansion (downselling was cyclical, not structural).
• AI stays a complement — no visible substitution in renewal/downsell data, and AskGartner-style embedding lifts engagement and retention.
• Falsification test: a second full year (through 2026) of sub-2% CV with wallet retention still below 100% — or AI substitution finally appearing materially in deal-loss codes — kills the bull case. At that point ~13.5x is a fair multiple on a shrinking annuity, not a cheap one.

For the bear case (structural, AI-driven impairment):

• CV keeps decelerating toward zero/negative and wallet retention stays sub-100% across the cycle, independent of the federal reset.
• Generative AI measurably substitutes for the analyst-inquiry layer, showing up in seat cuts and lost renewals.
• Gross margin/pricing power erode below the high-60s as the base matures and price increases stop sticking.
• Falsification test: CV inflects up for two consecutive quarters with federal stabilized and net-new-business positive — the +3.7% floor holds and re-firms — which would confirm the cyclical read and invalidate the terminal-decline thesis embedded at ~$136.

15. Source Appendix

See the Source Appendix below for the full citation list — primary sources (Gartner FY2025 10-K filed 2026-02-12, prior 10-Ks, 2026 DEF 14A, 8-Ks, SEC Form 4 corpus, CIK 749251), earnings-call transcripts (Q3’25–Q1’26), and secondary sources (Forrester FY2025 results, trade press on the DOGE/federal and G2/Digital Markets transactions). Every non-obvious fact in this article traces to a dated primary source.

APPENDIX A — Standard Diligence Questionnaire

Gartner, Inc. (NYSE: IT) · Report date: July 3, 2026

Supplemental to the article. Fact / Interpretation / Assumption labels where they matter.

General

What thoughtful questions have other investors asked? The central one dominates every sell-side note: is the Contract-Value deceleration cyclical (federal/DOGE + tariff-driven budget freeze) or the leading edge of structural, AI-driven disintermediation of syndicated research? Secondary questions: (a) how much of the sub-100% wallet retention is federal vs. commercial; (b) whether the record 2025 buyback (at ~$260) was smart counter-cyclical capital return or value destruction into a decelerating business; © whether the “12%+ EPS CAGR” target is a healthy commitment or an admission that organic growth is gone and EPS now comes from buybacks; and (d) why no insider bought a single share personally through a 75% collapse.

Cyclicality & Earnings Nature

Cyclical high or low? Below mid-cycle on the operating engine (CV +1% vs. a normal +8–14%), but not on margins — EBITDA margin (21.2%) is only modestly off peak. Interpretation: earnings power is depressed relative to the franchise’s potential, but not washed-out; normalized EPS ~$11.5–12 is near trend, growing mid-single-digits. External or internal drivers? Predominantly external in 2025 (DOGE federal cuts, tariff-driven budget freezes, tech-vendor retrenchment) layered on an internal choice to protect margin over growth (sales-force cuts). Revenue stability: high — ~84% recognized ratably, ~78% recurring subscription, contracts prepaid, deferred revenue $2.81B. This is among the most visible revenue bases in the market; the risk is growth, not volatility. Market size/direction: a large ($200B+ framed) but low-single-digit structural-growth advisory pie in which Gartner grows by share, price, and seat expansion — currently stalled. Predominantly international-capable but US-concentrated; the US federal slice (~2% of CV, rebased to ~$114M) was the acute 2025 hit.

Business Quality & Competitive Moat

Industry more or less competitive? More, at the low-differentiation end — free/AI substitutes and vendor “research” crowd the fringe — but the top-of-market oligopoly (Gartner ~13x Forrester’s revenue) is, if anything, consolidating as sub-scale peers shrink (Forrester −8%). Profitability: ROIC ~21.7% (~20–25% five years); ROE meaningless (near-zero/negative book equity from buybacks). Industry profitability for the scale leader is excellent (capital-light, 77% Insights contribution margin); for sub-scale players, poor. Barriers to entry: high for scale (40 years of brand + a 2,400-analyst bench + 13,000-client amortization base — never replicated), low for niche. Understandable? Yes — a subscription-research annuity with two adjacencies. Undermined by low-cost foreign labor? Not directly; the product is brand/authority/proprietary-data, not labor arbitrage — though the market now trades it like the offshore IT-services names (Genpact/EXL). Do brands matter? Decisively — the Magic Quadrant “cover value” is the core moat. Nature of competition: brand/standard + scale + habit vs. free/AI substitution at the margin. Switching costs: real but habit/search-cost-based (annual contracts), evidenced by durable 85–86% logo retention.

Financial Condition & Balance Sheet

Assets not on the balance sheet? Yes, materially — the analyst bench, brand, Magic Quadrant franchise, and 13,000-client relationships are internally-generated and largely uncapitalized; economic value vastly exceeds the $8.1B balance sheet. Off-balance-sheet liabilities? None material beyond operating/finance leases already on the sheet ($366M finance leases; office leases). Accounting conservatism: conservative — revenue recognized ratably (no pull-forward; deferred revenue tracks CV), trivial capitalized software, modest SBC (2.4% of revenue). The one aggressive-looking optic — CFO 1.77x net income — is a non-cash impairment add-back, not earnings management. CapEx-hungry? No — capex ~1.8% of revenue; among the most asset-light models in the market.

Capital Allocation & Management

FCF and its use: ~$1.1–1.2B/yr, deployed 100% into buybacks (no dividend, no M&A of scale since CEB 2017). Philosophy: relentless share-count reduction (−24% in five years, $6.0B). Recent acquisitions? None material; a divestiture (Digital Markets → G2, ~$110M, Feb-2026). Buying back shares? Aggressively — FY25 was the largest buyback ever ($1,991M), leaning into the crash; ~$1.245B authorization remains. Issuing shares to insiders? Modest — SBC 2.4% of revenue, more than offset by buybacks. Compensation: 88% CEO pay long-term; LTI 100% performance-based with CV growth = 70% of awards and not tied to EPS/FCF (so buybacks don’t game comp) — genuinely well-designed. Management motivations: shareholder-oriented in structure; but insiders own only 2.6% and bought zero shares personally through the drawdown — a conviction gap worth noting. CEO Hall in role since 2004, extended to 2031.

Valuation & Market Data

ADR/MLP/K-1? No — a US C-corp common stock (NYSE: IT). Dividend policy: none; all capital return via buyback. How profitable? Very — 68% gross margin, ~22% ROIC, ~18% GAAP operating margin (ex-impairment). Net income diverging from cash? Only optically (impairment add-back inflates the CFO/NI ratio); on normalized earnings, cash conversion is a healthy ~1.3–1.4x and the deferred-revenue float is real customer cash. At ~$136: ~13.5x GAAP P/E (~11.7x normalized), ~8x EV/EBITDA, ~12–13% FCF yield — the cheapest in the company’s history on P/E and P/S.

Risks & Downside

What would cause the stock to decline (further)? Another quarter or two of decelerating/negative CV; wallet retention falling further; AI substitution finally appearing in deal-loss data; a permanent re-rating into the low-multiple IT-services bucket even if fundamentals stabilize. Catastrophic-loss risk? Low — debt-light (net leverage ~1x, no maturity pre-2028), cash-generative, no single-product/fraud/solvency red flags. Total-loss risk? Negligible — this is a de-rating/growth-quality debate, not a solvency story. The realistic downside is the bear scenario (~$80–95), a further ~30–40%, not zero.

Recent News & Events

Has the environment changed? Yes, materially and recently: DOGE federal cuts (March 2025), “Liberation Day” tariffs (April 2025), a single −27.6% session on the Q2-2025 print (August 2025), the Digital Markets → G2 divestiture (closed Feb-2026), an inaugural investment-grade bond to fund buybacks, and sell-side capitulation (UBS Neutral $164, Wells Fargo Underweight $120, June 2026) with the stock hitting a $125.73 low. Accounting changes? None material — the $150M Digital Markets goodwill impairment (2025) is a one-time write-down, not a policy change. New markets/facilities/management? Continuity — CEO/Chairman Hall extended to 2031, CFO Safian in place; the strategic changes are the divestiture and a margin-over-growth posture, not expansion.

APPENDIX B — Source Appendix

Gartner, Inc. (NYSE: IT) · Report date: July 3, 2026

Primary sources before secondary; every non-obvious fact in the article traces to a dated primary source. Facts distinguished from interpretation throughout.

Primary — SEC filings (US filer, CIK 0000749251)

1. Gartner FY2025 Form 10-K (filed 2026-02-12; period end 2025-12-31; it-20251231) — Business, segments, competition, risk factors, MD&A (segment revenue, Contract Value, client/wallet retention, quota-bearing headcount, workforce-reduction charge), consolidated financials, tax reconciliation, debt schedule, Digital Markets goodwill impairment, event-cancellation insurance gain history. https://www.sec.gov/Archives/edgar/data/749251/000074925126000112/it-20251231.htm
2. Gartner FY2024 / FY2023 / FY2022 / FY2021 Form 10-Ks (filed 2025-02-13, 2024-02-15, 2023-02-16, 2022-02-23) — multi-year revenue, margin, CV, buyback, and share-count trend.
3. Gartner 2026 Definitive Proxy (DEF 14A) (filed 2026-04-15; it-20260415) — executive compensation design (Adjusted EBITDA / Revenue bonus; SARs + PSUs; CV growth = 70% of LTI; Pay-vs-Performance), insider/institutional ownership, basic shares outstanding (67.5M as of 2026-04-02).
4. Form 8-K corpus (2024–2026) — quarterly earnings releases and guidance (Q3’25 2025-11-04, Q4’25/FY26 guide 2026-02-03, Q1’26 2026-05-05), buyback authorizations, Digital Markets/G2 divestiture, bond issuance.
5. SEC Form 4 corpus (2021–2026; ~799 filings, sampled ~23 across Oct-2025–Jul-2026) — insider transaction read: zero code-P open-market purchases through the drawdown; routine grants/vesting/tax-withholding; reference prices $239 → $154 → $130.

Primary — earnings-call transcripts (via ROIC.ai)

6. Gartner earnings calls Q3-2025 (2025-11-04), Q4-2025 (2026-02-03), Q1-2026 (2026-05-05) — management framing of CV deceleration, federal/DOGE reset, AI-substitution (“very small number of transactions”), wallet-retention/downselling, BTI transformation/AskGartner, buyback pace, FY26 guidance and the “12%+ EPS CAGR” target. Treated as hypothesis, validated against filings.

Quantitative data feeds

7. ROIC.ai MCP — income statement, balance sheet, cash flow (FY20–25), profitability ratios (ROIC, margins), enterprise value, valuation multiples (own-history). Third-party aggregated data; reconciled to the 10-K (the filing is authoritative where they differ).
8. AZI (azitrading.com) — 5-year daily price/OHLCV CSV (adjusted, EMAs, beta/alpha) and the valuation_index own-history percentile ranks (P/E ~0.5th, P/S ~0.4th — cheapest ever; P/B not meaningful on near-zero book).
9. FactorsToday factor model (factorstoday.com/api) — stock loadings (Momentum −0.649, Quality +0.167, custom “Financial Data Titans” +0.568), leaderboard (lifetime max drawdown −77.2%, y1 −65.5%, negative Sharpe all horizons), stock-info (beta, alpha −0.44), idiosyncratic vol (44.9%), related/factor-peers (ACN, Genpact, EXLS, Globant). Third-party statistical estimates; overlay subordinate to the thesis.

Secondary — industry & trade press

10. Forrester Research FY2025 results — revenue −8% to $396.9M, CV −6% (peer scale/trend benchmark). https://www.forrester.com/press-newsroom/forrester-research-reports-2025-fourth-quarter-and-full-year-financial-result/
11. G2 to acquire Capterra / Software Advice / GetApp from Gartner — PRNewswire (Digital Markets divestiture, ~$110M). https://www.prnewswire.com/news-releases/g2-to-acquire-capterra-software-advice-and-getapp-from-gartner-302673901.html
12. RBC / channel press on the DOGE-federal CV slowdown — Channel Futures / AInvest (federal reset, tariff/DOGE drag). https://www.channelfutures.com/channel-sales-marketing/gartner-revenue-undeterred-by-heavy-doge-federal-futs
13. Gene Hall named Chairman; CEO contract extended to end-2031 — BusinessWire / Hartford Business Journal (July 2024). https://www.businesswire.com/news/home/20240701879561/en/Gartner-Names-CEO-Eugene-A.-Hall-Chair-of-the-Board
14. Sell-side stance (June 2026) — UBS Neutral (PT lowered to $164); Wells Fargo Underweight (PT lowered to $120) — consensus benchmark for the Variant Perception section.

Every material figure was reconciled to the primary SEC filing; this article rests entirely on public primary sources.