The New York Times Company (NYSE: NYT) — A Subscription Flywheel Finally Earning Its Keep, Re-Rated to Its Richest-Ever Price

Independent fundamental research. 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. Price reference: $76.71 (close 2026-07-16). All figures reconcile to SEC filings and the Q1 2026 earnings call unless noted. This is general information, not investment advice.

⚡ Claude’s Take

This block is the author’s own independent, subjective opinion — general information, not investment advice. The analytical body that follows carries no position and no price target.

Verdict: HOLD — a genuinely high-quality, net-cash, self-funding compounder that has finally converted its subscription scale into real operating leverage (17% ROIC, ~16% and rising operating margin, ~$580M FCF), but whose stock has been re-rated to the top of its own decade-long price-to-sales and price-to-book range. This is “great business, full price.” Not a short — the fundamental momentum is real and the balance sheet is pristine — but the margin of safety at ~$77 (≈22x EV/EBITDA, ≈4.5x EV/sales, ~1.4% FCF-yield-plus-buyback) is thin. Accumulate on weakness toward the low-$60s / high-$50s (≈17–18x EV/EBITDA, ~3.5x sales), where you’d be paying a fair price for a mid-teens grower rather than a rich one.

The market is pricing NYT correctly as one of the very few news organizations to win the digital transition — a rare, brand-anchored subscription franchise with pricing power, a genuinely differentiated non-news bundle (Games/Wordle, Cooking, Wirecutter, The Athletic), and now a demonstrated ability to expand margins ~200bps a year while still growing revenue ~10–12%. What it is not pricing is any of the tail risk: that the same generative-AI wave now inflating NYT’s licensing optionality (the Amazon deal; the OpenAI suit) also erodes the top-of-funnel referral traffic management openly acknowledged is under pressure from “a small number of tech companies,” or that digital-ad’s recent 30%+ growth (a real 2025–26 tailwind) normalizes. The framing is a quality-compounder-at-a-price / mild-momentum name, not a falling knife and not a value setup — the factor model confirms it: low beta (~0.50), positive alpha, +44% trailing-year return, but a highly idiosyncratic R²~10% tape whose factor “twins” are low-vol income vehicles, i.e., the stock now trades like a defensive quality bond-proxy, which is exactly how a fully-priced compounder behaves. Catchy version: “They won the news war — and the stock now charges you full price for the trophy.”

• Conviction: medium. Clean numbers, clean balance sheet, but the entry price does the work here and it isn’t there.
• Flips bullish: a drawdown to the high-$50s/low-$60s, OR evidence the 15M-subscriber target is conservative (sustained 1.2M+ annual digital net adds with ARPU still rising) that would re-underwrite the multiple on faster earnings.
• Flips bearish: a demonstrable step-down in engagement/subscriber adds from AI-driven traffic loss (the Google/LLM “zero-click” risk), or an adverse, precedent-setting outcome that reframes the AI story from “licensing windfall” to “content commoditized.”

📈 Stock Price Action — Five-Year Event Map

Factual price history — not a recommendation and not a price target. The price move is a Fact; the attributed driver is Interpretation. Source: public split/dividend-adjusted daily price series; events cross-referenced to earnings prints and news.

The arc. Over five years NYT round-tripped and then broke out. From a COVID-era low near $24.6 (Mar 2020) it ran to ~$55 (late 2021) on the pandemic news/subscription surge and the growth-stock melt-up, gave almost all of it back to ~$26 (mid/late 2022) as rates rose, the ad market cracked, and investors questioned the ~$550M Athletic acquisition — then re-rated steadily and powerfully to an all-time high of $86.83 (7 Apr 2026) as margins inflected. It now trades ~$77, roughly 12% off that high, with a 52-week range of $50.44–$86.83. The last three years have been a near-uninterrupted one-way street up (3-yr annualized total return ~+24%, 1-yr ~+44%), driven by the business, not the multiple alone — though the multiple did the last leg.

Cycle narrative. (1) The pandemic made The Times a habit for millions and the market extrapolated it into a growth stock. (2) 2022 was the reckoning — rates repriced all long-duration equities, the ad market cratered, and the Athletic deal looked like empire-building into losses; NYT briefly traded near its 2020 lows. (3–4) The recovery was earned, not multiple-driven: the digital bundle worked, Games/Wordle became a genuine acquisition and engagement engine, and the subscriber base marched from ~8M toward 11M+. (5) The 2025–26 leg is the important one here — it is where operating leverage finally showed up (incremental operating margins ~40%, AOP margin +200bps/yr), advertising re-accelerated to 30%+ digital growth, and the AI narrative flipped from threat to optionality (the Amazon licensing deal; a credible legal claim against OpenAI). (6) The recent ~12% pullback is digestion after a near-triple, not a thesis break.

1. Executive Summary

The New York Times Company is the clearest single example of a legacy news organization that has successfully crossed to a digital, subscription-first, direct-to-consumer model — and, as of 2025–26, has begun to prove that the model produces expanding margins and abundant free cash flow, not merely growth. Revenue rose from $1.78B (2020) to $2.82B (2025), a ~9.6% CAGR; operating margin expanded from 9.9% to 16.0%; diluted EPS more than tripled from $0.60 to $2.09; ROIC improved from ~12% to ~17%; and the company generated $584M of free cash flow in 2025 on a net-cash balance sheet (cash + short-term investments $642M, essentially no financial debt). Q1 2026 extended the trend: consolidated revenue +12%, adjusted operating profit +27%, AOP margin 16.6% (+200bps), digital subscription revenue +16%, digital advertising +32%, and total subscribers above 13 million (12.78M at year-end 2025) against a stated target of 15 million by year-end 2027.

The moat is real but narrower than the bulls imply. It is a genuine intangible-asset and habit moat — the Times brand as the default American general-interest news product, reinforced by a bundle (News + Games/Wordle + Cooking + Wirecutter + The Athletic + Audio) that raises engagement, retention, and pricing power. That combination shows up in the financials the way a moat should: 50%+ gross margins, ~90% subscription revenue that is recurring, rising ARPU, and ~40% incremental operating margins. The vulnerability is at the top of the funnel: NYT depends on third-party platforms (Google Search, social, and increasingly LLM answer-engines) for discovery, and management itself flags that “a small number of tech companies whose moves continue to impact traffic to publishers.” Generative AI is simultaneously an optionality (content-licensing revenue — the Amazon deal; the OpenAI litigation as a value-establishing enforcement action) and a structural risk (zero-click answers eroding referral traffic and the free-to-paid conversion funnel).

Capital allocation is conservative and shareholder-friendly without being aggressive: a dividend compounding ~24%/yr (to $0.68/share, ~32% payout), steady buybacks that have shrunk the share count from 167M to 162M, and one sizeable acquisition (The Athletic, ~$550M, Jan 2022) that is strategically coherent (sports as a subscription and advertising vertical) even if its standalone economics remain undisclosed. Governance is family-controlled via a dual-class structure: the Ochs-Sulzberger family, through the 1997 Trust, controls the Class B shares that elect roughly 70% of the board — a structure that has protected editorial independence and long-term orientation for over a century but that also insulates management from the market.

The tension is entirely valuation. On its own history, NYT trades at the top of its range on the metrics that matter for a now-profitable franchise: EV/sales ~4.5x (vs. a ~3.1x five-year average), EV/EBITDA ~22.7x (vs. mid-to-high teens historically), price/sales in the ~94th percentile of its own decade, and price/book in the ~96th. The P/E percentile looks deceptively “cheap” (~30th) only because 2020–21 earnings were so depressed that the historical P/E range is inflated; in absolute terms NYT trades at ~33x trailing and ~30x forward earnings. This is a company being priced as a secular compounder that has definitively won — a defensible view, but one that leaves little room for the AI-traffic risk to bite or for advertising’s recent 30%+ growth to normalize. The body below argues the quality case and the price case in full; the analysis itself takes no position (see Claude’s Take for the one labeled exception).

2. Business Overview

The New York Times Company is a subscription-first digital media company built around a single flagship news product and a widening set of adjacent lifestyle and interest verticals. Importantly, NYT reports as one reportable segment / one reporting unit — the President & CEO manages the business as a single integrated bundle (FY2025 10-K, Note 15). The New York Times Group and The Athletic (acquired January 2022) are managed as parts of that whole, not as separate reportable segments, which means there is no audited segment-level P&L for The Athletic — its economics surface only through episodic management commentary (a genuine disclosure gap).

Revenue architecture. FY2025 revenue of $2,824.9M (+9.2%) breaks into three streams (Q4 2025 press release, Statements of Operations):

• Subscription — $1,950.8M (69.1% of revenue) — the core and the growth engine, of which digital-only subscription was $1,434.3M (50.8% of total, +14.3%) and print subscription $516.4M (18.3%, −3.2%). Digital revenue crossed $2B for the first time in FY2025. Total subscription revenue grew 11.3% in Q1 2026 to ~$517M; subscription revenue is ~90%+ recurring and is the highest-quality revenue in the business.
• Advertising — $566.0M (20.0%, +11.8%) — split between digital and print. Digital advertising has re-accelerated dramatically (+31.6% in Q1 2026 to $93.3M; +24.9% in Q4 2025), driven by growth in ad supply (Games, sports, and new formats), strong marketer demand, and a large first-party data asset. Print advertising is in secular decline.
• Affiliate, licensing & other — $308.1M (10.9%, +5.7%) — Wirecutter affiliate-commerce commissions, content licensing (including the new AI/LLM licensing category, e.g., the Amazon deal), commercial printing/distribution, HQ floor leasing, and live events.

The bundle is the strategic core. Rather than sell a single newspaper subscription, NYT sells the “All Access” portfolio: hard News, plus Games (the Crossword, Spelling Bee, and especially Wordle, acquired 2022, a low-cost, high-frequency engagement and acquisition machine), Cooking (recipes), Wirecutter (product reviews), The Athletic (sports), and Audio (podcasts including The Daily, and Serial) — each also sold standalone. An All Access Family tier ($30/month, up to four users) launched September 2025. The strategic logic is that multiple daily “habits” across a household raise engagement, lower churn, and justify recurring price increases as subscribers migrate from single-product to bundle to higher tiers. Bundle & multiproduct subscribers — now 6.48M, or ~53% of the digital-only base — carry meaningfully higher ARPU (~$12.9 bundle vs. ~$3.4 for a single non-news product), so the mix shift toward the bundle is itself an ARPU tailwind. Management’s disclosed playbook is explicitly ARPU-plus-volume: grow the subscriber base and raise ARPU through promotional-to-full-price step-ups and periodic price increases.

How it makes money. The unit economics are attractive and asset-light: acquire a subscriber (often through a low-priced product like Games or a promotional offer), engage them across multiple products to build habit and reduce churn, migrate them to the full-price all-access bundle over time, and monetize the resulting large, engaged, affluent, first-party-identified audience again through advertising and affiliate commerce. The marginal cost of serving an additional digital subscriber is near zero, which is why incremental operating margins run ~40% and why the model throws off free cash flow well in excess of net income.

At year-end 2025 the business served 12.78 million total subscribers (12.21M digital-only), ~26% of them international.

Verdict: A coherent, multi-revenue-stream, subscription-anchored digital media model — one of the very few in the industry that demonstrably works. The business is well-understood and the revenue base is high-quality and predominantly recurring.

3. Industry Dynamics

The industry NYT operates in is, structurally, one of the worst in media — and NYT’s achievement is to have escaped the average. Newspaper publishing has been in secular decline for two decades: print advertising, once the profit engine, collapsed as classifieds went to Craigslist/Google and display went to Meta/Google; local newspapers have been decimated; and the open web’s economics have been captured by a handful of platforms. The average participant in “newspaper publishing” (SIC 2711, NYT’s own classification) is a melting ice cube.

What separates the winners. Digital news is a winner-take-most structure at the national/general-interest tier. A small number of destinations — NYT, The Wall Street Journal, The Washington Post, a few others — command the scale, brand trust, and breadth to sustain a direct paid-subscription relationship; everyone else fights for scraps of programmatic ad revenue on platform-intermediated traffic. NYT is the clear scale leader in general-interest digital news subscriptions in the U.S., with a subscriber base (13M+) that dwarfs domestic peers. Scale in original journalism is self-reinforcing: the fixed cost of a global newsroom (~$300M+ of the cost base) is spread over a growing subscriber base, so unit economics improve with scale — a genuine economies-of-scale-plus-captivity advantage in Greenwald’s taxonomy, where the “captivity” is habit/subscription stickiness and the scale economy is newsroom fixed-cost leverage.

Profit pools and the value chain. The profit pool has migrated decisively from advertising (intermediated, cyclical, platform-captured) to consumer subscription (direct, recurring, pricing-power-bearing). NYT’s strategic masterstroke was to pivot early and hard to reader revenue: subscription is now ~two-thirds of revenue vs. advertising’s declining ~one-fifth, inverting the historical newspaper model. This is the single most important structural fact about the business — it has diversified away from the bad part of the industry (cyclical, secularly-declining advertising) toward the good part (recurring consumer revenue).

Regulation and structural factors. Light-touch regulation (First Amendment protections; no rate/licensing regime). The dominant structural forces are: (a) platform dependency — Google, Meta, Apple, and now LLM answer-engines control discovery and referral traffic; (b) the generative-AI disruption — simultaneously a licensing revenue opportunity and a “zero-click” traffic threat; and © content-cost inflation — quality journalism is expensive and labor-intensive (unionized newsroom), and AI does not meaningfully lower the cost of original reporting.

Capital cycle (Marathon lens). Unusually for media, capital is not flooding into premium general-interest news — if anything, capital has fled the sector (local news collapse, layoffs, closures), leaving the scaled survivors with less competition for attention and talent. That supply-side withdrawal is a tailwind for NYT: the barriers to recreating a global newsroom and a trusted brand are enormous, and no one is trying. The one place capital is flooding — generative AI — is the wild card that could either pay NYT (licensing) or route around it (answer engines).

Verdict: A structurally bad industry in which NYT occupies a structurally good niche. The generic newspaper business is in secular decline; the scaled, brand-anchored, direct-subscription national-news-plus-lifestyle niche NYT has built is defensible and growing. The verdict is favorable for NYT specifically, with the large caveat that the AI-driven reshaping of discovery is a genuine, unresolved structural risk to the entire open-web content model.

4. Competitive Position

The moat is an intangible-asset (brand/trust) moat reinforced by habit and bundle-driven switching costs — and it is corroborated by the financials. Name the mechanism precisely: The New York Times is the default trusted general-interest news brand in the United States, a position built over 170+ years and impossible to replicate with capital. That brand converts into (a) the ability to acquire subscribers efficiently, (b) pricing power (repeated price increases with modest churn), and © an advertising audience that is large, affluent, and first-party-identified. The proof that this is a real moat and not a narrative: 50%+ gross margins, ~90%-recurring subscription revenue, rising ARPU (+2.4% in Q1 2026 even as the base grew), ~40% incremental operating margins, and ROIC of ~17% — economics that would deteriorate sharply if the brand/habit advantage evaporated.

The bundle deepens the moat. Games (especially Wordle), Cooking, Wirecutter, The Athletic, and Audio are individually modest businesses but collectively create multiple daily touchpoints per household. A subscriber who does the Wordle every morning, checks scores on The Athletic, and pulls a recipe from Cooking is far stickier than a pure news subscriber — and management has repeatedly cited “performance at our pricing step-up points” and bundle strength as the ARPU driver. This is switching cost of a soft but real kind: not contractual lock-in, but accumulated habit, saved recipes, game streaks, and the friction of unbundling.

Direct competitive comparison. Against national news peers — WSJ (Dow Jones/News Corp), Washington Post, The Atlantic, and free/ad-supported players — NYT has the largest subscriber base, the broadest bundle, and the strongest standalone digital-product portfolio (no peer has a Wordle/Games engine of comparable scale). Against information-services franchises like Thomson Reuters (Westlaw) or Gartner, NYT’s moat is weaker: those franchises have contractual, mission-critical, workflow-embedded switching costs and ~90%+ retention on must-have professional tools, whereas NYT sells a discretionary consumer subscription that a household can cancel with two clicks. NYT’s ~$77 stock trades at ~22.7x EV/EBITDA versus Thomson Reuters at ~14x — i.e., the lower-moat consumer franchise trades at a higher multiple than the higher-moat professional one, a comparison the valuation section returns to.

The vulnerability is discovery, not retention. The moat protects the installed base well; it protects new-subscriber acquisition less well, because the top of the funnel runs through platforms NYT does not control. If Google’s AI Overviews and LLM answer-engines satisfy news queries without a click-through, the funnel that feeds free-to-paid conversion narrows. Management is candid that traffic is under pressure. This is the single most important pressure-test of the moat’s durability and it is not yet resolved.

Network effects: largely absent. Do not overstate them. NYT has scale economies and habit stickiness, but no true two-sided network effect (a new subscriber does not make the product better for existing subscribers, except marginally in Games leaderboards/multiplayer and comments). Be rigorous here: this is a scale-plus-brand-plus-habit moat, not a network-effects moat.

Verdict: A durable but discretionary-consumer moat — genuine intangible-asset and habit advantages that show up cleanly in margins and retention, but weaker than professional-information franchises and exposed at the discovery layer to platform/AI disruption. Durable advantage: yes. Impregnable: no.

5. Growth History and Forward Opportunities

History. The growth record is strong and, more importantly, improving in quality:

Revenue compounded ~9.6% over five years, but the story since 2023 is operating leverage: incremental operating margins of ~38–43% in 2024–25 mean nearly half of each new revenue dollar is dropping to operating profit. That is the flywheel doing what it was designed to do — the fixed newsroom cost is now spread over a much larger, mostly-digital revenue base.

Growth is organic and high-quality, with one large exception. The subscriber and ARPU growth is organic; the one meaningful acquired contributor was The Athletic (2022), which added a sports vertical and audience but also added losses that weighed on margins in 2022 (the −6.9% incremental margin that year partly reflects Athletic integration and the ad recession). The 2024–25 margin inflection is the market’s evidence that Athletic and the broader bundle are now accretive rather than dilutive.

Forward opportunities:

1. Subscriber runway to 15M by year-end 2027. The subscriber base has compounded steadily — digital-only subscribers grew 6.69M (2020) → 8.0M → 8.83M → 9.70M → 10.82M → 12.21M (2025), with total subscribers at 12.78M end-2025 (from 11.43M end-2024). FY2025 added ~1.40M net new digital-only subscribers. Against management’s explicit target of 15 million total subscribers by year-end 2027 (unchanged since it was set in February 2022) and its framing of “many tens of millions more” addressable, the volume runway is real, though the rate of adds will naturally slow as the base matures.

(NYT switched from counting paid “subscriptions” through FY2021 to deduplicated “subscribers” from FY2022, so the 2020–21 totals are not directly comparable to 2022+; the digital-only series is the cleaner read.) 2. ARPU. The larger long-term lever. As promotional subscribers step up to full price and bundle prices rise, ARPU compounds — +2.4% in Q1 2026 and management frames pricing as addressing “the whole demand curve.” A subscription base migrating from ~$10–15/mo promotional to ~$25/mo bundle full price embeds years of ARPU growth. 3. Advertising re-acceleration. Digital advertising +32% in Q1 2026 on new supply (Games, sports) and first-party data — a genuine, if more cyclical, tailwind. 4. Video. The current flagship investment: reporter video, visual investigations, and shows, aimed at capturing linear-TV-news defectors and deepening engagement. Early, unmonetized, and a cost headwind today — optionality, not yet a proven revenue line. 5. AI/LLM content licensing. The Amazon deal established a template; management is “open to deals” that meet its conditions and treats licensing as a potential recurring high-margin revenue stream.

Verdict: High-quality, primarily-organic growth with multiple compounding levers (volume × ARPU × advertising) and genuine optionality (video, AI licensing). The quality of growth — recurring, high-margin, self-funding — is well above the media-sector average. The one caveat is that the rate of subscriber adds will naturally decelerate as the base matures, shifting the growth burden increasingly onto ARPU.

6. Financial Quality

This is a high-quality set of financials. The signatures of a good business are all present: high and rising gross margin (50.8%), expanding operating margin (16.0%), strong and improving returns on capital (ROIC ~17%, ROE 14%), abundant free cash flow, and a fortress balance sheet.

Margins and returns. Gross margin of ~51% is high for any media business and reflects the digital-subscription mix. Operating margin has expanded ~600bps in five years (9.9%→16.0%) with EBITDA margin at 19.3% and the Q1 2026 exit rate (AOP margin 16.6%, +200bps) suggesting continued expansion. ROIC of ~17% comfortably exceeds any reasonable cost of capital and has risen every year since 2022 — the capital-cycle “high returns” signal that, per Marathon, would normally attract competition, except that the barriers to entry (brand, newsroom scale) are prohibitive.

Cash flow and its quality. FY2025 operating cash flow was $584.5M and, after minimal capex (~$34M), free cash flow was $550.5M (up sharply from $381.3M in FY2024) — capital intensity is low (the business owns its headquarters and a printing plant but requires little growth capex). FCF/share was ~$3.4, and cash conversion (OCF/net income) was 1.7x, a healthy sign that earnings are backed by cash (D&A and SBC add-backs, favorable working capital). One quality caveat: stock-based compensation has risen sharply, from $14M (2020) to $74M (2025) — now ~13% of operating cash flow and a real, if manageable, expense that partly offsets buybacks. A second, explicitly-flagged caveat: 2025–26 cash flow is flattered by a non-recurring tax benefit — the One Big Beautiful Bill Act lowered cash taxes by ~$65M in FY2025 and is expected to add ~$60M to FY2026 operating cash flow, most of which management says will not recur beyond FY2026. Normalizing for that, sustainable FCF is closer to ~$490–520M than the reported ~$550M.

Balance sheet — a genuine fortress. At year-end 2025: cash and marketable securities of ~$1.2B (cash + short- and long-term investments; +$256M year-over-year), zero financial debt (only $36.6M of capital-lease obligations), an undrawn $400M revolver, and therefore a large net-cash position. Total equity $2.04B; book value/share $15.74, tangible book value/share $8.60. The one legacy liability is a frozen single-employer defined-benefit pension — and it is better than the headline suggests: at year-end 2025 the plan obligation was $1,182.5M against $1,092.3M of assets, with the qualified plans overfunded (a +$75.9M asset) and only the non-qualified plans unfunded (−$166.1M); net across all plans was −$90.2M. The ~$211M balance-sheet “pension and postretirement” line combines that unfunded non-qualified pension with retiree-medical obligations. Contributions are modest (~$13M in 2025, ~$14M expected 2026); this is a well-de-risked legacy artifact, not an overhang. Current ratio 1.5x. This balance sheet gives NYT complete strategic freedom — fund video and product investment, acquire, return capital, and absorb a downturn without financing risk.

Unit economics. The core digital-subscription unit economics are excellent: near-zero marginal cost to serve, rising ARPU, and multi-year retention driven by bundle habit. The ~40% incremental operating margin is the clearest quantitative proof that economics improve with scale.

Verdict: Economics clearly improve with scale — this is a high-return, cash-generative, financially unassailable business. The two blemishes are rising SBC and a temporary tax-driven cash-flow flattering; neither is thesis-altering, but both should be normalized out of any valuation.

7. Capital Allocation

Management’s capital-allocation record is conservative, consistent, and shareholder-aligned — competent rather than brilliant, which for a family-controlled franchise is arguably the right setting.

Returns to shareholders. The dividend has compounded ~24%/year (from $0.23/share in 2020 to $0.68/share paid in 2025) at a conservative ~32% payout, and in February 2026 the Board raised the quarterly dividend +28% to $0.23/share (~$0.92 annualized) — leaving ample room for continued double-digit growth. Buybacks have been steady but not aggressive: the share count has declined from 167.2M (2020) to 161.8M (2025), roughly −3% cumulatively; the Board approved a new $350M Class A repurchase authorization in February 2026 (~$334M remaining as of January 30, 2026, on top of prior programs), and repurchased 883,602 shares for ~$55.4M in Q4 2025 alone. Combined 2025 capital return (~$275M of dividends + buybacks) was ~50% of FCF, so the company is building cash even after returning capital — consistent with a net-cash balance sheet that keeps growing. The critique: with a fortress balance sheet, ~17% ROIC, and a stock the family clearly views as core, one could argue for more aggressive buybacks — but buying back stock at ~4.5x sales / record valuations is not obviously accretive, so management’s restraint is defensible.

M&A. One material deal: The Athletic (~$550M cash, January 2022), funded from the balance sheet (the 2022 cash-flow statement shows ~$516M net cash paid for acquisitions). Strategically coherent — sports is a large, advertiser-friendly, subscription-additive vertical, and The Athletic brought subscribers and a national sports newsroom. Management has disclosed (via commentary, not audited segment data — NYT is one reportable segment) that The Athletic reached its first profitable quarter in Q3 2024 (~+$2.6M operating profit) and has been “solidly profitable” since, supporting the read that the 2024–25 consolidated margin inflection is partly Athletic turning accretive. The remaining critique is the disclosure gap: without a segment P&L, investors must take the profitability claim on management’s word. Smaller tuck-ins (Wordle, 2022; Wirecutter earlier; Audm; Serial Productions, 2020) have been cheap and strategically sound.

Reinvestment intensity. R&D/product and the newsroom are the real “capex” of this business, and management is investing counter-cyclically in journalism and now video while peers cut — a defensible use of the moat’s cash flow to widen it. S&M is disciplined. The company is not empire-building.

Incentive alignment. Compensation is tied to revenue, adjusted operating profit, and subscriber growth (per the proxy) — reasonable metrics that align management with the operating flywheel. The dual-class governance is the defining structural feature: Class A shareholders (the public float) elect only 30% of the Board (4 of 13 directors); Class B elects the other 70% (9 directors), and the Ochs-Sulzberger family’s 1997 Trust holds 94.6% of the Class B shares (738,810 of 780,724). Because Class B is under 0.5% of the ~161.9M total shares, the family exercises ~70% board control on a low-single-digit percentage of the economic equity. The Trust is contractually barred from selling or converting the Class B block or voting for a change of control — a lock that has protected editorial independence and long-term capital allocation for over a century (A.G. Sulzberger is Chairman; Meredith Kopit Levien is President & CEO). The trade-off is stark: public shareholders benefit from patient, non-raidable, franchise-first stewardship, but have no meaningful governance recourse — there is no activist path and no takeover premium optionality, and if allocation ever deteriorated, the float could not force change.

Verdict: Intelligent, conservative capital allocation — strong organic reinvestment into the moat, a fast-growing well-covered dividend, steady buybacks, and one strategically-sound acquisition. The main critiques are the opacity of Athletic’s economics and rising SBC. Nothing here damages the thesis; capital allocation is a modest positive.

8. Changes and Headwinds — Last Two Years

Strategic and operational changes:

• Margin inflection (2024–2026). The most important change: operating margin expanded from ~12% (2023) to 16% (2025) with the Q1 2026 exit rate higher still, as the bundle and Athletic turned accretive and operating leverage kicked in. This is what re-rated the stock.
• Advertising re-acceleration. Digital advertising went from a drag to a ~30%+ grower in 2025–26 on new ad supply (Games, sports) and first-party data — a genuine positive surprise relative to secular-decline fears.
• AI content licensing became a revenue category. In May 2025 NYT signed its first generative-AI licensing deal, with Amazon — a multi-year agreement covering NYT core, Cooking, and The Athletic content for training Amazon’s foundation models and surfacing attributed summaries in Alexa. Terms were not disclosed by NYT (press estimates ~$20–25M/year, unconfirmed); FY2025 “affiliate, licensing & other” growth was attributed partly to higher licensing revenue. Management frames further deals as optionality, subject to three conditions (strategic fit, sustainable fair-value exchange, control over use).
• Video investment. A deliberate, cost-bearing pivot into reporter video, visual investigations, and shows — production more than doubled in Q1 2026 — aimed at the next engagement frontier. A near-term margin headwind, a long-term optionality bet.
• Disclosure changes. NYT stopped breaking out The Athletic separately and changed bundle-subscriber disclosure — reducing transparency into two areas investors care about.

Headwinds and adverse developments:

• Platform/AI traffic risk. Management explicitly acknowledges that “a small number of tech companies” are impacting publisher traffic. Google AI Overviews and LLM answer-engines threaten the referral/discovery funnel that feeds free-to-paid conversion — the single most-cited bear point.
• The OpenAI/Microsoft litigation. NYT’s copyright suit (filed December 2023, SDNY) is live and proceeding: the court largely denied the motions to dismiss in March 2025, core copyright claims are advancing through discovery, and in November 2025 the court ordered OpenAI to produce a de-identified sample of 20 million ChatGPT conversation logs (affirmed January 2026); NYT amended its complaint in mid-2026. It is a double-edged binary of uncertain timing: a win or favorable settlement establishes licensing value and validates the AI-optionality thesis; an adverse or narrowing outcome would remove that optionality and signal that content is legally fair-game for training.
• Cost inflation. A unionized newsroom and video investment push operating costs up ~8–9% (guided), and AI does not lower the cost of original reporting.
• The non-recurring tax tailwind (OBBBA) that flatters 2025–26 cash flow rolls off after 2026.

Verdict: On balance these developments strengthen the thesis operationally (margin inflection, ad re-acceleration, licensing optionality) while raising the structural risk profile (AI/traffic, litigation binary). The near-term operating momentum is unambiguously positive; the medium-term structural questions are genuinely open.

9. Risk Analysis

The two risks that matter are (1) the AI/traffic structural question — the only risk with the power to break the growth algorithm — and (2) valuation — the risk that even a business performing exactly as hoped delivers poor forward returns because too much is already in the price. Catastrophic/total-loss risk is very low: net cash, real cash flows, an irreplaceable brand. This is not a balance-sheet or going-concern risk; it is a growth-durability-and-price risk.

10. Valuation Discussion (Embedded Expectations)

Where it trades. At $76.71, NYT carries a market cap of ~$12.4B and, net of ~$600M+ cash, an enterprise value of ~$12–13B. On trailing/LTM figures that is approximately: EV/sales ~4.5x, EV/EBITDA ~22.7x, EV/EBIT ~27x, P/E ~33x, P/FCF ~21x, FCF yield ~4.7% (or ~4.3% normalizing out the OBBBA tax benefit). The dividend yield is ~1.0%.

Against its own history (the key tell). NYT is expensive relative to itself on every metric that isn’t distorted by trough earnings:

• Price/sales ~4.3x sits in the ~94th percentile of its own decade and near the top of its 2.3x–4.9x range; price/book ~6.3x is in the ~96th percentile and at a record; price/tangible book ~8x is a record.
• EV/sales ~4.5x vs. a ~3.1x five-year average; EV/EBITDA ~22.7x vs. a mid-to-high-teens history.
• The P/E percentile (~30th) is misleading: 2020 P/E was ~86x and 2021 ~37x on depressed earnings, inflating the historical range. In absolute terms, ~33x trailing / ~30x forward is a full multiple for a ~10–12% revenue grower, even a high-quality one.

The consistent reading across metrics: the market has re-rated NYT to price it as a proven secular compounder that has won, not as a cyclical publisher. The own-history valuation composite (73rd percentile) understates the richness precisely because the earnings-based metric is distorted; the asset- and sales-based metrics (record highs) are the truer read.

Embedded expectations. To justify ~22.7x EV/EBITDA on ~$573M LTM EBITDA, the market is underwriting something like: sustained low-double-digit revenue growth (volume to 15M+ subscribers × rising ARPU × advertising) for years, continued ~150–200bps/yr margin expansion toward the low-20s%+ operating margin, and — critically — that AI does not impair the traffic funnel. A reverse-DCF at a ~4.5% FCF yield with a ~9% discount rate implies the market expects roughly 8–10% FCF growth sustained for a decade-plus. That is achievable if the subscriber/ARPU/margin flywheel runs uninterrupted, but it prices in essentially none of the AI-traffic downside and assumes advertising’s recent surge is durable.

Scenario analysis (illustrative, ~3-year forward; no price target):

• Bull: Subscribers to ~15–16M, ARPU compounding mid-single-digits, digital ad staying double-digit, operating margin to ~20%+, EPS to ~$3.25–3.50 by 2028; the market keeps a premium ~30x+ multiple → materially higher. Requires AI-traffic risk to prove benign and licensing to scale.
• Base: Subscribers to ~15M, revenue compounding ~8–9%, margin to ~18–19%, EPS to ~$2.90–3.10 by 2028; multiple compresses modestly toward ~25x as growth matures → mid-single-digit annualized returns (earnings growth largely offset by de-rating). This is the “great business, full price” outcome — you earn roughly the FCF yield plus buyback, not more.
• Bear: AI erodes top-of-funnel; net adds fall toward ~500–700K/yr; ad normalizes; margin expansion stalls; the market re-rates a decelerating consumer-subscription name toward ~15–17x EV/EBITDA (still a premium to legacy media) → 25–40% downside even with flat-to-modest earnings.

Comparative check. The subscription/information-services cohort frames NYT’s premium sharply. The starkest comp is Thomson Reuters — a higher-moat, ~80%-recurring, family-controlled information franchise — trading at ~14x EV/EBITDA after an AI-fear de-rating, versus NYT’s ~22.7x:

The market is paying a large premium for NYT’s consumer subscription (softer switching costs, discretionary, two-click cancel) over TRI’s professional subscription (contractual, mission-critical, workflow-embedded). Two readings are possible. The charitable one: NYT deserves the premium because it grows faster (~10–12% vs. TRI’s ~8%), has cleaner AI optionality (a licensing tailwind rather than a Westlaw-style substitution threat), a pristine net-cash balance sheet, and a longer runway (12.8M subscribers of “many tens of millions” addressable). The cautionary one: the lower-moat, more-discretionary franchise trading at a ~60% EV/EBITDA premium to the higher-moat one is a warning that NYT sits on the expensive edge of the cohort — and the AI question the market treats as a positive for NYT and a negative for TRI could just as easily invert, since NYT’s consumer traffic funnel is arguably more AI-exposed than TRI’s embedded professional workflows. The comp does not settle the debate, but it makes explicit that owning NYT here is a bet that its growth/optionality premium over the cohort is durable.

Verdict (no recommendation): The valuation embeds a “won the war, keep compounding, AI is benign or a positive” outcome. That is a defensible base case, but it leaves little margin of safety and prices in almost none of the genuine AI-traffic tail risk. The stock is priced for continued excellence, not for surprises.

11. Variant Perception

Consensus view. NYT is the great survivor and winner of digital news — a proven subscription compounder with a widening bundle, expanding margins, a fortress balance sheet, and AI optionality (licensing) rather than AI risk. Sell-side is broadly constructive (price targets clustered ~$80–82), and the stock’s low-beta, positive-alpha, +44%-trailing-year behavior has made it a “quality-compounder” favorite. Consensus treats the AI-traffic risk as real but manageable and the licensing opportunity as an emerging positive.

The strongest bull case. The flywheel is genuinely self-reinforcing and early: 13M subscribers is a fraction of the “many tens of millions” addressable; ARPU has years of step-up runway; advertising has re-accelerated structurally on new supply and first-party data; incremental operating margins of ~40% mean margins keep climbing toward information-services levels; and AI is a net positive — high-quality, trusted, original content becomes more valuable and scarcer as the web fills with AI slop, and licensing plus enforcement (the OpenAI suit) converts that scarcity into a new high-margin revenue stream. On this view ~33x earnings is cheap for a business that could compound EPS mid-teens for a decade.

The strongest bear case. NYT is a discretionary consumer subscription whose entire acquisition funnel depends on third-party discovery that generative AI is actively disintermediating. As Google AI Overviews and ChatGPT-style answer engines satisfy news queries without a click, the top of the funnel narrows, net adds decelerate, and the growth algorithm shifts entirely onto ARPU — which has a ceiling. Advertising’s 30%+ growth is a supply-driven, cyclical high that will normalize. And all of this is happening at a record price/sales and price/book, ~22.7x EV/EBITDA, versus a higher-moat peer (TRI) at ~14x. The bear does not need NYT to be a bad business — only for it to be a decelerating good business bought at a great-business price.

The 3–5 assumptions that decide it:

1. Does AI erode or leave intact the discovery/conversion funnel? (The single most important variable — determines whether subscriber growth continues.)
2. Is advertising’s recent 30%+ digital growth durable or a cyclical/supply-driven peak?
3. How much ARPU runway remains before price increases drive churn?
4. Does AI licensing become a meaningful, recurring, high-margin revenue line — or a one-off?
5. Does the market sustain a ~30x+/22x-EV-EBITDA multiple, or mean-revert toward the high-teens as growth matures?

Falsification tests. Bull is falsified by two-to-three quarters of decelerating digital net adds (toward <200K/quarter) with softening engagement metrics, signaling AI-traffic bite. Bear is falsified by sustained 300K+ quarterly net adds with rising ARPU and continued margin expansion through 2026–27, proving the funnel is intact and the flywheel self-sustaining.

Factor-positioning read (from the tape). The factor model reinforces the “fully-priced quality” framing rather than either extreme. NYT is low beta (~0.50), positive alpha, low-vol, with a mild value tilt and, notably, no momentum loading despite a +44% trailing year — the return is ~90% idiosyncratic (R²~10%, specific vol ~29%), i.e., driven by company-specific execution, not a factor wave. Its factor “twins” are low-vol dividend/income vehicles — the market is treating NYT as a defensive quality bond-proxy, precisely how a fully-valued compounder trades. There is no crowded-momentum blowoff signature here and no washed-out value signature either; it is a well-owned, well-behaved quality name whose forward return depends on the fundamentals continuing to justify a rich multiple. That is consistent with the valuation section’s conclusion: the risk is not a crash catalyst on the tape, it is mediocre forward returns if the business merely performs as expected into a full price.

12. Fact vs. Interpretation

13. Open Questions

1. What is The Athletic’s standalone P&L? Is it profitable yet, and what did it contribute to the 2024–25 margin inflection? (Disclosure was withdrawn.)
2. What is actually happening to referral/organic traffic and free-to-paid conversion as AI Overviews scale? Management flags pressure but does not quantify it.
3. How large and how recurring is AI-licensing revenue (Amazon and any successors), and what are the economics?
4. What is the realistic ARPU ceiling before price increases meaningfully raise churn?
5. What is the status and likely resolution of the OpenAI/Microsoft litigation, and what precedent would a ruling set for content-licensing value?
6. How much of the 30%+ digital-ad growth is durable structural share gain vs. a supply/cyclical peak?
7. What is the normalized FCF once the OBBBA tax benefit rolls off after 2026?

14. What Must Be True

For the bull case (owning at ~$77 works):

• Subscriber growth continues toward and past 15M — i.e., the AI/traffic risk proves benign and the acquisition funnel stays intact.
• ARPU compounds mid-single-digits for years as promotional subs step up and bundle prices rise, without triggering churn.
• Operating margin continues to expand ~150–200bps/yr toward the low-20s%, sustaining ~40% incremental margins.
• The market sustains a premium (~28–33x earnings / ~20x+ EV/EBITDA) multiple.
• Falsification test: two-to-three consecutive quarters of digital net adds falling below ~200K with softening engagement — evidence the funnel is breaking — falsifies the bull.

For the bear case (this is a decelerating good business at a great-business price):

• AI-driven disintermediation narrows the discovery funnel; net adds decelerate toward <700K/yr; growth leans entirely on a maturing ARPU lever.
• Advertising normalizes from its 30%+ pace back toward low-single-digits.
• Margin expansion stalls as video and newsroom costs outrun revenue.
• The market re-rates toward the high-teens EV/EBITDA as growth matures.
• Falsification test: sustained 300K+ quarterly net adds with rising ARPU and continued margin expansion through 2027 — proving the flywheel self-sustains against AI — falsifies the bear.

The two cases share a single decisive fault line: whether generative AI erodes NYT’s top-of-funnel faster than the brand/bundle/ARPU flywheel can offset it. Everything else is second-order.

15. Source Appendix

See Appendix B below for the full, dated source list. Primary sources: NYT FY2025 Form 10-K and FY2020–FY2024 10-Ks (SEC EDGAR, CIK 0000071691); NYT Q1 2026 Form 10-Q and Q1 2026 earnings-call transcript (2026-05-06); NYT DEF 14A (2026-03-13); and NYT quarterly earnings press releases. Quantitative data cross-checked against public market-data and factor-model providers and reconciled to the filings. All multiples and percentiles as of 2026-07-16 close ($76.71).

APPENDIX A — Standard Diligence Questionnaire

The New York Times Company (NYSE: NYT) · Report date: 2026-07-17 · Price reference: $76.71

Fact / Interpretation / Assumption labels applied where material. All figures reconcile to SEC filings and company disclosures.

General

What thoughtful questions have other investors asked about this company? The debate clusters on four axes: (1) AI vs. the funnel — does generative AI (Google AI Overviews, LLM answer-engines) erode NYT’s discovery/referral traffic and free-to-paid conversion faster than the brand/bundle can offset? (2) ARPU ceiling — how much pricing power remains before increases drive churn, given digital-only ARPU is only ~flat-to-up (~$9.7)? (3) Advertising durability — is the +30% digital-ad growth structural or a cyclical/supply-driven peak? (4) Valuation — is ~22.7x EV/EBITDA / record price-to-sales justified for a ~10–12% grower, especially versus higher-moat info peers (Thomson Reuters ~14x)? A fifth, quieter question: what is The Athletic actually earning, given NYT no longer discloses a segment P&L?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: Above mid-cycle but not obviously a peak. Two cyclical tailwinds are elevated — digital advertising (+30%+, likely closer to a cyclical/supply high) and a non-recurring OBBBA tax benefit (~$65M cash-tax saving in 2025, ~$60M to 2026 OCF, largely non-recurring after 2026). The subscription engine (~69% of revenue) is structurally growing, not cyclical. Normalized earnings are somewhat below the reported run-rate once the tax benefit and peak ad growth are adjusted.

Driven by the external environment or internal actions? Predominantly internal — the margin expansion (op margin 9.9%→16.0%) is a deliberate operating-leverage outcome of the subscription/bundle strategy. External factors (ad cycle, big news cycles, rates driving the multiple) amplify but do not create the trend.

How stable are revenues? High stability: ~69% subscription, ~90%+ recurring, low churn on the bundle. Advertising (~20%) and affiliate/licensing (~11%) are more variable.

Outlook for products/services? Growing — subscriber base toward 15M (2027 target), ARPU step-ups, advertising re-acceleration, video optionality, AI licensing.

How big will this market be? Large and global. Management frames “many tens of millions” of addressable subscribers internationally and domestically; the general-interest news + lifestyle-bundle market is large, growing in paid-digital penetration, and NYT is the scale leader.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? For premium general-interest news: less — local news has collapsed and capital has fled, leaving scaled survivors with less competition for attention and talent. At the discovery layer: more — platforms and LLMs increasingly intermediate and disintermediate access.

How profitable is the business (ROIC, ROE)? ROIC ~17.1% (FY2025, up from ~12% in 2020); ROE 14.0%; operating margin 16.0%; gross margin 50.8%. High and rising — a genuinely good business by return-on-capital.

How profitable is the industry — competitors, barriers to entry? The industry average (newspapers) is poor; the scaled-national-subscription niche is highly profitable. Barriers to entry are prohibitive: a 170-year brand, a global newsroom costing $300M+/year, and 12M+ subscriber habit are not replicable with capital. (Greenwald: economies-of-scale + customer captivity.)

Can the business be easily understood? Yes — subscription media with advertising and licensing overlays.

Can it be undermined by foreign low-cost labor? No — original, trusted, U.S.-focused journalism is not offshoreable; AI is the relevant substitution threat, not labor arbitrage.

Do brands matter? Decisively. The Times brand is the moat — it drives efficient acquisition, pricing power, and a premium advertising audience.

Nature of competition? Competition for attention/discovery (platforms, other publishers, all digital media) and for subscription wallet (WSJ, WaPo, streaming, other subscriptions). NYT competes on brand, breadth of bundle, and product quality.

Customers’ switching costs? Soft but real — habit (daily Wordle/Games/Cooking), saved content, game streaks, bundle convenience. Not contractual; a household can cancel in two clicks. Weaker than professional-information switching costs.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes — the Times brand and the subscriber relationships are the core value and are largely unrecognized (only acquired intangibles/goodwill from Athletic/Wordle/Wirecutter, ~$638M, appear). Economic value far exceeds book equity ($2.04B).

Off-balance-sheet liabilities? Multiemployer pension withdrawal-liability risk (noted in the 10-K), operating-lease commitments, and multiyear content/talent commitments. None appears material to solvency.

How conservative is the accounting? Reasonably conservative. Watch items: rising SBC ($74M) as a real expense; adjusted-operating-profit (AOP) add-backs that exclude SBC, D&A, and severance (standard but flattering); and the withdrawn Athletic segment disclosure, which reduces transparency.

How CapEx-hungry is the business? Very low — capex ~$34M on $2.8B revenue (~1.2% of sales). Asset-light digital model; the real “investment” is the newsroom and product/engineering expense (in the P&L, not capex).

Capital Allocation & Management

How much FCF, and how is it used? FY2025 FCF ~$550M. Uses: a fast-growing dividend (raised +28% to ~$0.92 annualized in Feb 2026, ~32% payout), steady buybacks (new $350M authorization, ~$334M remaining), and reinvestment into journalism/video/product. Excess accumulates as cash (net-cash balance sheet keeps growing).

Significant acquisitions recently? The Athletic (~$550M, Jan 2022) and Wordle (2022) are the notable ones; both strategically sound. No large deals since.

Buying back shares? Yes, steadily — share count 167.2M→161.8M since 2020; new $350M program (Feb 2026).

Issuing large amounts of stock to insiders? SBC has risen to $74M/year (~2.6% of revenue) — meaningful and worth watching, but partly offset by buybacks; net share count is declining.

Compensation policy / motivations of management? Comp tied to revenue, AOP, and subscriber growth (aligned with the flywheel). The controlling family (Ochs-Sulzberger, via the 1997 Trust electing 70% of the board) is motivated by long-term franchise durability and editorial independence over a century-plus horizon, not quarterly optimization — generally a positive for capital discipline, but it removes public-shareholder recourse and takeover optionality.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No — U.S. C-corp, NYSE-listed Class A common (1099 dividends). Dual-class (Class B family-controlled, not publicly traded).

Dividend policy? Progressive: ~24%/yr growth to $0.68 paid in 2025; raised +28% to ~$0.92 annualized in Feb 2026; ~32% payout; ~1.2% yield. Ample coverage and room to grow.

How profitable is the business? See ROIC ~17% / margins above — high and rising.

Is net income diverging from cash from operations? OCF ($584M) exceeds net income ($344M) by ~1.7x — a positive divergence (cash-backed earnings, aided by D&A/SBC add-backs and working capital), though the SBC add-back and the temporary tax benefit should be normalized.

Risks & Downside

What factors would cause the stock to decline? (1) Evidence of AI/traffic-driven subscriber-growth deceleration; (2) multiple compression from record valuations as growth matures; (3) advertising normalization; (4) an adverse OpenAI-litigation/AI-licensing outcome; (5) a broad risk-off/recession de-rating of a full-multiple name.

Risk of a catastrophic loss? Very low — net cash, no debt, real recurring cash flows, an irreplaceable brand. No solvency or going-concern risk.

Chance of a total loss? Negligible over any reasonable horizon. The realistic downside is poor forward returns (multiple compression + growth deceleration), not impairment of the enterprise.

Recent News & Events

Has the business environment changed recently? Yes: (1) digital advertising re-accelerated to +30%+ (2025–26); (2) generative AI became both a licensing opportunity (Amazon deal, May 2025) and a traffic threat; (3) the OpenAI/Microsoft litigation advanced (MTD largely denied Mar 2025; 20M ChatGPT logs ordered Nov 2025); (4) the margin inflection (op margin to 16%) drove a stock re-rating to a record high ($86.83, Apr 2026) before a ~12% pullback to ~$77.

Significant acquisitions? None recent beyond Athletic/Wordle (2022).

Change in accounting policies? NYT reports as one segment and withdrew The Athletic’s separate disclosure and changed bundle-subscriber disclosure — reducing transparency, not a GAAP policy change per se.

Recent changes — new markets, facilities, management? Video is the flagship new investment (production more than doubled in Q1 2026); All Access Family tier launched Sep 2025; management stable (A.G. Sulzberger Chairman, Meredith Kopit Levien CEO since 2020); a $350M buyback and +28% dividend hike announced Feb 2026.

APPENDIX B — Source Appendix

The New York Times Company (NYSE: NYT) · Report date: 2026-07-17

Primary sources first. All quantitative figures were reconciled to SEC filings and the company’s earnings materials; third-party market-data and factor-model providers were used for computed ratios, price series, and factor positioning and are labeled as such. Access date for all online sources: 2026-07-17 (market data as of 2026-07-16 close, $76.71).

Primary — SEC filings (EDGAR, CIK 0000071691)

1. Form 10-K, FY2025 (filed 2026-02-27) — nyt-20251231.htm. Revenue, segment/one-reportable-segment disclosure (Note 15), subscriber counts, 15M-by-2027 target, pension (funded status), buyback authorization, risk factors. https://www.sec.gov/Archives/edgar/data/71691/000007169126000011/nyt-20251231.htm
2. Form 10-K, FY2024 (filed 2025-02-27) — nyt-20241231.htm. Prior-year financials and subscriber base.
3. Form 10-K, FY2023 (filed 2024-02-20) — nyt-20231231.htm. Multi-year trend.
4. Form 10-Q, Q1 2026 (filed 2026-05-06) — nyt-20260331.htm. Q1 2026 financials.
5. DEF 14A proxy (filed 2026-03-13) — nyt-20260313.htm. Dual-class structure, board election (Class A 4 of 13 / Class B 9), Ochs-Sulzberger 1997 Trust holdings, executive compensation, leadership.
6. Q4/FY2025 earnings press release (8-K exhibit, filed Feb 2026) — Statements of Operations (revenue split), subscriber/ARPU footnotes, dividend and buyback announcements. https://www.sec.gov/Archives/edgar/data/71691/000007169126000008/pressrelease12312025.htm
7. Q4/FY2024 earnings press release (filed Feb 2025). https://www.sec.gov/Archives/edgar/data/71691/000007169125000021/pressrelease12312024.htm
8. Historical 10-Ks FY2020–FY2021 — subscriber-methodology change (subscriptions → deduplicated subscribers from FY2022).

Primary — Earnings call

9. NYT Q1 2026 earnings call transcript (2026-05-06). CEO Meredith Kopit Levien, CFO Will Bardeen. Source: ROIC.ai transcript tool. Subscriber count (13M+), digital-sub rev +16%, digital ad +32%, AOP margin 16.6%, LTM FCF $542M, OBBBA tax benefit, Q2 guidance, Amazon AI deal, video strategy.

Secondary — litigation & AI

10. The New York Times Company v. Microsoft Corp. & OpenAI (S.D.N.Y., filed Dec 2023). Case tracker: https://en.wikipedia.org/wiki/The_New_York_Times_Company_v._Microsoft_Corporation ; BakerHostetler AI litigation tracker: https://www.bakerlaw.com/new-york-times-v-microsoft/ . March 2025 MTD largely denied; Nov 2025 order to produce 20M de-identified ChatGPT logs (affirmed Jan 2026); mid-2026 amended complaint.
11. NYT–Amazon AI licensing deal (announced 2025-05-29). TechCrunch: https://techcrunch.com/2025/05/29/the-new-york-times-and-amazon-ink-ai-licensing-deal/ ; Axios: https://www.axios.com/2025/05/30/nyt-amazon-ai-licensing-deal . First generative-AI license; dollar terms not company-confirmed (~$20–25M/yr press estimate).
12. The Athletic profitability — Axios (2025-05-20): https://www.axios.com/2025/05/20/nyt-athletic-profitable . First profitable quarter Q3 2024; management commentary (not audited segment data).

Secondary — recent results & price context

13. Q1 2026 results coverage — Bloomberg (2026-05-06); Investing.com Q1-2026 slides recap. Revenue $712.2M, digital ad +31.6%, adj EPS $0.61 beat, ~310K digital net adds, stock +~5%.
14. Sell-side price-target notes (for consensus context only, not relied upon): JP Morgan Overweight (PT raised to $82, May 2026); Bank of America Neutral (PT lowered to $80, June 2026). Via aggregated financial-news feeds.

Third-party quantitative datasets (computed metrics; reconciled to filings)

15. ROIC.ai — income statement, balance sheet, cash flow, profitability ratios (ROE/ROA/ROIC), enterprise value, valuation multiples, per-share data (FY2020–FY2025 + TTM). Accessed 2026-07-17.
16. Market-data provider (azitrading.com) — valuation_index own-history percentile ranks (P/E 30th, P/B 96th, P/S 94th, composite 73rd, as of 2026-07-16) and the split/dividend-adjusted daily price series (five-year event map). News feed.
17. FactorsToday (factorstoday.com) — factor loadings (Market 0.39, Value 0.21; R²~10%), leaderboard (beta 0.50, alpha +0.14, 1yr +44%/Sharpe 1.42, 3yr +24% ann.), stock-info (RS, 52-week range), specific-vol (~29% idiosyncratic), related-stocks. Accessed 2026-07-16.

Note on data authority: third-party market-data and factor-model providers supply aggregated/estimated data, not primary. Where any figure drives a verdict it was reconciled to the underlying SEC filing or earnings release; the filing governs in case of conflict. Percentiles and factor loadings are statistical estimates reported as-is.