On November 1, 2021, ninety people played Wordle. Josh Wardle, a software engineer in Brooklyn, had built the word game for his partner, a puzzle lover, and after it became an obsession in his family's WhatsApp group - "it derailed the family group chat," he told TechCrunch - he released it to the rest of the world in October.
By the first Sunday of 2022, more than 300,000 people were playing. A week later it was 2 million. On January 31, The New York Times announced it had bought the game, for a price Time reported as "in the low seven figures." Ninety players to a seven-figure exit inside three months, with no mobile app, no advertising, and no registration form.
The part I keep coming back to is that none of the growth choices were made to grow anything. TechCrunch ticked off the product's rules: "no mobile app. Only one word every 24 hours. No advertising. No registration required." That is a list a growth team would refuse on sight, and Wardle knew it. "With Wordle, actually, I kind of deliberately did what you're not meant to do if growth is your goal. And bizarrely, I think, those things have led to growth. But obviously, a ton of it is luck, and being in the right place at the right time." The luck part is true and I will not argue with it. Luck explains the size of the wave. It does not explain the shape of the thing the wave landed on. The shape was designed, and the design was rationing.
What the ration actually did
One word a day meant nobody could binge. The game put a hard ceiling on how much of you it would take, and that ceiling turned out to be a floor: the product that takes a sliver of your day gets the same sliver again tomorrow, from the same person, without asking twice.
But one-a-day alone is not the mechanism, and Wardle himself flagged why. "My partner and I play a lot of The New York Times word games, and they follow that one-a-day model," he told TechCrunch, before making the sharper point: a game where every player got a different random word, once a day, "wouldn't have caught on the way it has" - because "it's one puzzle, and everybody is solving it." The ration was collective. Two million people were solving the same word on the same day, which turned a private puzzle into a shared appointment - and a shared appointment is the only kind an audience can keep together. It caught on in New Zealand first, he has said, because its Twitter crowd is small and densely connected - "if one person shares something, a lot of others are likely to see it" - and one early adopter there began typing her results out as emoji squares, the shareable grid Wardle later built into the game itself. Scarcity made the game an appointment. The shared word made the appointment a meeting.
Sixty-six quiet days
Why an appointment outgrows a buffet is not a marketing finding. It is a habit finding. In 2010, Phillippa Lally and colleagues at University College London published a study in the European Journal of Social Psychology titled "How are habits formed: Modelling habit formation in the real world." They had 96 volunteers choose an everyday behavior - something to eat, drink or do - and repeat it daily in the same context, after breakfast for instance, for 12 weeks, rating each day how automatic it felt. The curves rose steeply at first, bent, then flattened toward a plateau. Among the 39 participants whose curves fit cleanly, the median time to reach 95 percent of that plateau was 66 days, and the range ran from 18 to 254.
Sixty-six days of showing up before the thing stops taking effort. Not one of those days feels like growth while it is happening - that is exactly the point. The compounding is invisible at the daily grain and total at the monthly one, which is why teams that judge by the burst never see it happen. The study even carries a mercy for the streak-obsessed: "Missing one opportunity to perform the behaviour did not materially affect the habit formation process." Skip a day. The curve does not care. The new word is still there in the morning.
So here is my opinion, plainly. Rationing is not a constraint you put on an audience. Rationing is the audience. Unlimited supply asks nothing of a visitor and schedules nothing for them to answer to. A ration gives the visit a shape - it ends, and it comes back, and it will be here tomorrow - and somewhere in those quiet weeks the return stops being a decision. The sites that grow the way Wordle grew are not the ones holding the most content. They are the ones with a tomorrow: one puzzle, one drop, one edition, changing on a rhythm a person learns without trying. Wardle's best line explains why strangers took to it the way his family did. "I think people have an appetite for things that transparently don't want anything from you." One word a day transparently wants almost nothing. The smallness of the ask is the honesty, and the honesty is the mechanism.
What this means on Bid Spotlight
I run a paid-rank leaderboard, so let me be exact about where this lands on a site that sells attention. A bid buys the crowd; that has never been hidden here. What no bid can buy is day sixty-six. The nightly top-3 edition goes out to our social followers once a day, which gives the board its own rhythm, and the board under it moves when money moves - but the reader who comes back tomorrow morning comes back for one reason only: something on your side of the click gave them a reason to. So the question before any bid is not "how many visitors will this slot send." It is "does my site have a tomorrow?" If it does, a position in front of a browsing crowd is a first look that compounds, because the crowd you met on day one is still solving puzzles on day sixty-six. If it does not, you are renting crowds, and the rent only ever buys day one.
Build something with a tomorrow, then claim your spot in front of today's crowd.