The stat that surprised me
Two weeks ago, the New York Times confirmed that Wordle is going to television, an NBC game show hosted by Savannah Guthrie and produced with Jimmy Fallon’s company. The interesting part was buried in the announcement. Wordle was played 4.2 billion times in 2025, down from 5.3 billion the year before. A decline. And rather than quietly move on, the Times is spending money to pull lapsed players back onto the daily grid.
I noticed because I have spent the last year building a small app that helps designers rehearse high-stakes workplace conversations, and there is exactly one question that has ever mattered for it. It is the same question sitting under that Wordle number: will you come back tomorrow?
That question turns out to be the entire product strategy for two very different companies.
Two companies, one playbook
The New York Times bought Wordle in 2022 for a low seven-figure sum and folded it into what had been a crossword-only games section. Four years later, the internal joke, as reported by several outlets, is that the Times is now a games company that happens to publish news. Its puzzle portfolio logged 11.2 billion plays in 2025. Roughly half of new subscribers now choose the bundle rather than news alone, and games are one of the stickiest reasons they stay.
LinkedIn watched that happen and quietly copied the pattern. In March 2024, app researchers found unreleased game code in the app before any announcement. By May, Queens, Pinpoint, and Crossclimb were live. Tango followed in October, Zip in early 2025, then Patches, a small Sudoku, and in June 2026 a word game called Wend, the eighth title in the lineup. LinkedIn reports that 86 percent of players return the next day and 82 percent are still there a week later.
Neither company launched with one big swing. That restraint is the real story.
How they actually test the idea
Here is the part product people should study.
When the Times built Pips, its first original logic puzzle, it did not ship it worldwide and hope. Pips was proposed in 2024, developed through 2025, and soft-launched as a beta in Canada first. The team watched retention in a contained market, and when the beta ended, players on Reddit openly asked for it back. That organic demand, not a slide deck, was the signal. Only then did Pips go global, in August 2025, with drag-and-drop play and what the head of games called a human-crafted twist.
LinkedIn ran a different version of the same discipline. Instead of one geographic beta, it treated each new game as a live experiment. Ship one puzzle, measure next-day and seven-day return, prove the daily habit holds, then add the next title on a roughly quarterly cadence. The validation metric was never downloads or reach. It was the retention curve.
Both companies also validated something subtler. When the Times shipped Connections and then Pips from scratch, it proved it could design a hit, not just acquire one. That is a more durable capability, and it is what justifies the whole bet.
What this means if you build products
Three takeaways I keep coming back to.
First, validate on the behavior you actually need, not the one that is easy to measure. For a habit product, a million installs mean nothing if day-two retention is flat. Pick your true north metric before you launch, and let it gate the next investment.
Second, soft-launch in a place where you can be wrong cheaply. A single-country beta or a single new feature is a way to learn with real users while limiting the blast radius. The clamor from Canadian Reddit told the Times more than any focus group could.
Third, and this is the thread running through my own work, taste is the moat. At the same conference where LinkedIn and the Times talked strategy, a Disney animation executive said the industry will never automate taste and imagination. The Times leans on human-crafted puzzles and named editors precisely because AI can generate a grid but cannot decide which grid is delightful. As AI makes execution cheap, judgment about what is worth making becomes the differentiator. That is as true for a puzzle as it is for the practice app on my laptop.
The open question
The product arc here is worth naming: acquire a habit, prove you can build one yourself, grow it into a portfolio, then extend it into new formats like a prime-time show. Each step is a bigger bet resting on the same daily return.
So here is what I am sitting with. When a beloved daily ritual starts to soften, is a TV show a smart way to re-energize it, or a sign the core loop has peaked and the growth has to come from somewhere else? I genuinely do not know. But I will be watching the retention curve, because that is where the answer always lives.
References
“The New York Times Wants the Wordle TV Show to ‘Re-Energize’ Lapsed Players,” A Media Operator: https://www.amediaoperator.com/analysis/wordle-tv-show-new-york-times-games-jonathan-knight/
“NBC Greenlights ‘Wordle’ Game Show,” NBCUniversal: https://www.nbcuniversal.com/article/nbc-greenlights-wordle-game-show-executive-producer-jimmy-fallon-and-new-york-times
“About Pips, the Newest Game From the New York Times,” TODAY: https://www.today.com/life/pips-new-york-times-game-rcna225538
“LinkedIn Launches Wend as First New Word Game in Two Years,” PocketGamer.biz: https://www.pocketgamer.biz/linkedin-launches-wend-as-first-new-word-game-in-two-years/
“LinkedIn Wants to Add Gaming to Its Platform,” TechCrunch: https://techcrunch.com/2024/03/16/linkedin-wants-to-add-gaming-to-its-platform/
“LinkedIn Has a New Way to Keep You Hooked,” TechRadar via Yahoo: https://tech.yahoo.com/social-media/articles/linkedin-way-keep-hooked-more-100214799.html
“NYT Earnings and the Wordle Acquisition,” Axios: https://www.axios.com/new-york-times-earnings-wordle-acquisition-users-0cb11458-5b27-493b-a14b-d0d96eb82f7a.html