The New York Times Answered Every Question But One

Q3 beat on subscribers, missed digital ads by a hair, and turned the call into a referendum on AI agents — while the shareholder lawsuit that dented the stock in September went unmentioned.

NYT · Earnings Call · 2026-11-04

A quarter that compounded quietly

The New York Times did what The New York Times does. Net new digital subscribers of roughly 420,000 took the base to 13.8 million, digital-only subscription revenue rose 14.6% to $421 million, total revenue climbed 10% to $769 million, and adjusted operating profit grew 14% to about $150 million. Adjusted diluted EPS went to $0.68 from $0.59. Meredith Kopit Levien framed it as the machine running on its own logic: “the growth came from strong engagement and retention kind of across the portfolio, and that is really the model working as it was meant to as it's designed” — Meredith Kopit Levien, President and Chief Executive Officer · 2026-11-04. The filed numbers lag the call by a quarter — the latest 10-Q covers the period ending 2026-07-31 — but they tell the same story from a longer vantage. Total revenue of $762M in that quarter, up 11% year over year, sits atop a shape that fell 32% over seven years and then rose 83% over eight. More telling is the operating leverage: operating margin of 16.4%, up 13 percentage points across the last six years and flat year over year. And the cash engine: a free-cash-flow margin of 21.5%, nine points better than a year ago.

What nobody asked

Here is the contrast worth sitting with. The stock closed 2026-11-04 at 22.5% below its April 7 peak of $85.86, and 14.4% below its late-July high; it is down 5% over the last ninety trading days. Part of that is valuation gravity on a name that still trades at 20.8x free cash flow, down 12% year over year. But part of it is September: major holders including the Florida State Board of Administration and the National Center for Public Policy Research sued the company over alleged bias in its Israel-Hamas coverage, and the shares fell about 10% in a week. Eight analysts asked questions on this call. Not one asked about the lawsuit. The agenda instead: ad load in games, the width of the Q4 cost range, whether an ASR was coming. There is a real gap opening between the analyst agenda — which is still modeling a subscription business — and the tape, which is now pricing a reputational and legal tail risk that management did not volunteer a single word on. When the questions run out and the stock is down 14% from its high, the silence is the story.

The question that replaced the licensing question

For five straight quarters, the marquee AI question was about monetizing it. In August 2025, an analyst asked what made the Amazon licensing deal palatable, and Kopit Levien answered that “the deal reinforces our principal, that our journalism and everything we do is worth paying for” — Meredith A. Kopit Levien, President and Chief Executive Officer · 2025-08-06. In May 2026 she was still bullish on the structure: “We are open to doing deals that sort of meet our conditions” — Meredith Kopit Levien, President and Chief Executive Officer · 2026-05-06. In August 2026 the framing had already soured into traffic: “the overall direction of travel is less traffic to publishers from the big platforms” — Meredith Kopit Levien · 2026-08-05. This quarter, the licensing question vanished entirely — no Amazon, no LLM partners, no litigation timeline. The company's own keyword set reflects the drop: licensing momentum fell harder than almost anything else in the quarter. In its place came a different AI question, asked first, from JPMorgan: will personal AI agents give subscribers a reason to cancel or haggle? That is a genuinely market-wide anxiety — the agent theme shows up across this quarter's reporters, from Shopify's "Personal Agent Protocol" to Booking's "horizontal personal agents" to NICE and Paymentus. The Times is not alone in being asked; it is unusual in having had nothing to say about the upside.

It's early days with agents and they seem like they could be an important force, but the story of exactly how they're going to do that, I think it's still to be written... we have the ability to permit, to restrict or to block specific tools.
Meredith Kopit Levien, President and Chief Executive Officer · 2026-11-04

That is a defensible answer — the moat is a direct relationship, and blocking is a real lever — but note what it concedes. Twelve months ago AI was a revenue line and a content-rights asset. Tonight it is a churn vector with a firewall around it. The rotation in narrative, not the numbers, is the change at this company.

Password sharing, ad load, and the recurring furniture

The other familiar question returned for at least the fourth time. Password sharing topped this company's keyword set in early 2026 and still ranks inside the top fifteen; analysts asked about it in February and again in August. Kopit Levien has stopped treating it as a crackdown and now sells it as the opposite: “it's in the sunshine. So sort of almost the opposite of cracking down on passwords” — Meredith Kopit Levien, President and Chief Executive Officer · 2026-11-04 — the family plan as a premium-priced penetration tool rather than a stick. It is a smart reframe, but it is a reframe, and the analyst community keeps asking because the answer keeps changing shape. Advertising is the soft spot. Digital ad revenue grew 12% to $110 million, slightly below guidance, and management leaned hard on unpredictability: “I would regard where we came in, in digital advertising in the quarter relative to our guidance as being more about exactly where ads land in quarters being hard to predict” — Meredith Kopit Levien, President and Chief Executive Officer · 2026-11-04. That is the third consecutive quarter of some version of that sentence, after two years of comfortable beats. Ad load remains the analyst lever of choice; management's answer is that new supply will be "more incremental and more across the portfolio." The incentive comp wobble in the 6%–8% Q4 cost guide got two separate airings, with Will Bardeen conceding the quarter is large and performance-linked enough to swing the range.

The cash is doing the talking

Where management was least evasive was capital allocation. Roughly $436 million of free cash flow year-to-date, $1.3 billion of cash and marketable securities, $276 million returned — and, crucially, “we have at this point, returned about 60% of our free cash flow so far this year” — William Bardeen, Executive Vice President and Chief Financial Officer · 2026-11-04, ahead of the stated 50% midterm target. The Q3 buyback was an explicit act of opportunism at lower prices, exactly the attractive return opportunities trigger built into the framework; repurchases of $35M in the latest quarter, up 50% year over year. Asked directly about an ASR for 2027, Bardeen would only say the mechanism could repeat. Strip away the noise and the operating story is intact: a subscription flywheel adding 420,000 net subs a quarter, 16%+ operating margins, and a balance sheet that funds its own buybacks. What has changed is where the risk now lives. It is no longer in the paywall or the price step-up — it is in distribution and attention, where a handful of platforms and a new class of agents decide how people find The Times, and in the non-financial domain that nobody on the call was willing to name. That is a harder problem to model, and the market's 22% drawdown from the April high is already voting on it.