How this is built
The method
A verdict without a method is just an opinion. Here is the rule applied to all 178 cards.
One dominant channel, never three
Every product is filed under a single primary channel. The dominant channel is the one without which the growth would not have happened, not the one bringing the most traffic today. Secondary channels are listed separately and capped at two.
The verdict is relative to you, not to them
It always answers the same question: can a solo founder or a small team, with no media budget and no pre-existing audience, run this play today? That is why Duolingo's play is not copyable even though the method is public.
- COPYABLE
- You could start this on Monday. No budget, no audience, no permission needed.
- HARD
- Reproducible, but it will cost you real money, real months, or a skill you may not have.
- UNFAIR EDGE
- It worked because of something they already had, or a window that has closed. Copying it will not reproduce it.
The costs are yours, not theirs
Time, money and time-to-signal describe what running this play costs you, starting today. Not what it cost them back then. Their numbers are history; yours are the decision.
So when a card says twenty-four months, it does not mean they took twenty-four months. It means you will need twenty-four months of audience before this play does anything for you.
Evidence before judgement
Every channel has observable signals: a sitemap far too big for the team, a hashtag fed by third parties, continuous presence in the Meta Ad Library, a living subreddit, a product name carrying the generic keyword. The channel is inferred from those signals, not from what the team says in an interview.
The confidence level is on every card
Confirmed (92 cards): the team said it publicly, or the signals leave no room for doubt. Read from signals: consistent with what is publicly observable, but nobody confirmed it. Best guess: a working hypothesis, useful to think with, not to repeat as fact.
That column exists because a database of 178 cards presented as fully verified would be a lie. An inferred card is still useful to think with. It is not usable to make a claim about a real company.
What this site does not do
It does not publish revenue numbers that are not public. Traction is an observable proxy, never an invented estimate.
It does not claim the listed channel is the only one they used. It claims it is the one that carried the growth.
It does not promise that copying a play produces the same result. Context, timing and execution usually matter more than the method.