The New York Times Didn't Buy a Word Game. It Bought a Front Door.

On November 1, 2021, a word game built by a Brooklyn software engineer for his partner had ninety players. Sixty-two days later it had crossed 300,000, and a week after that, more than two million - all of it spread by people posting little grids of green and yellow squares to Twitter, never spoiling the answer.⁷ On January 31, 2022, The New York Times bought it for what the company described only as 'an undisclosed price in the low-seven figures.'¹ The internet immediately decided the Times had become a games company. The internet read the headline and missed the play.

The official story is that a 170-year-old newspaper paid a small fortune for a free puzzle and pivoted into entertainment. The real story is quieter and stranger: the Times didn't buy a game. It bought a front door - the cheapest, lowest-friction way yet found to start a relationship with a stranger who would never have paid for the news first.

A free game is a strange thing to pay seven figures for

Look at the deal on its own terms and it makes no sense. Wordle was free. It carried no ads, ran no subscription, and Josh Wardle had deliberately refused to monetize the viral monster he'd created. You don't pay low-seven figures for a revenue line that doesn't exist. You pay it for the people. The Times wasn't buying a product; it was buying a daily habit shared by millions, attached to nothing and pointed nowhere - and it already had somewhere to point it. By December 2021, weeks before the deal, NYT Games had crossed one million Games subscriptions, and its puzzles had been played more than 500 million times that year.² Wordle wasn't a new business. It was rocket fuel for one that already worked.

The funnel runs downhill from a puzzle to a paywall

Here is the mechanism the games-company story skips. Almost nobody wakes up and decides to pay for a newspaper cold. But millions will play a free word game, and a fraction of those will pay a few dollars a month to play all the games, and a fraction of those will, over time, accept an offer to add news, Cooking, Wirecutter, and The Athletic for not much more. Each step is a smaller ask than the one above it. The genius of Wordle is that it lives at the very bottom of the staircase, where the friction is lowest and the audience is largest. The Times spent that same period building the rest of the staircase too: in early 2022 it closed a $550 million all-cash deal for The Athletic, a sports product to bolt onto the bundle.³ Games is the lure; the bundle is the catch. And it's working as designed - by mid-2024, the Times had more single-product subscribers for its non-news products than for news alone, and news-only subscribers had fallen by nearly 40% since September 2022.⁵

The payoff shows up in the only number that ultimately matters: cash from people. The Times ended 2024 with roughly 11.43 million total subscribers, about 10.82 million of them digital-only, and 2024 subscription revenue of $1.788 billion - up from $1.656 billion the year before.⁴ That is not the financial signature of a company that wandered into puzzles. It is the signature of a company that found the cheapest possible top-of-funnel and engineered a path from it to the most expensive product it sells.

But isn't the Times really just a games company now?

The sharpest objection is the one that takes the data seriously: if non-news single-product subscribers now outnumber news-only ones, hasn't the tail become the dog? It's a fair read, and it's the wrong conclusion - because the single-product cohort is the segment that pays the least. By Q4 2025, the 'other single-product' bucket where Games naturally lives carried an average revenue per user of just $3.36, and that figure had actually fallen 6% year over year. The bundle subscriber, by contrast, was worth $12.92 and still rising, up 3%.⁸ That gap is the whole strategy in two numbers. The single-product subscriber isn't the destination; they're the deposit. The Games cohort is the fastest-growing and the lowest-yielding at once, which is exactly what an on-ramp looks like - cheap to fill, valuable only to the degree it feeds something richer downstream. Call the Times a games company and you've mistaken the doormat for the house.

The Times paid low-seven figures for a game it would never charge a cent to play, and that was the point. It wasn't buying flavor or fun or a new identity. It was buying the one thing harder to manufacture than journalism: a daily reason for millions of people who don't read the news to open a Times product anyway. The bet was never that Wordle would make money. It was that Wordle would make members - and that, climbing the staircase one small yes at a time, enough of them would eventually pay for the house.