Nobody says no at a kickoff.
Three people say the date is realistic, two think it doesn’t solve the problem, one sees a huge technical rewrite, but no one speaks up. It costs nothing to know that on day one and a lot to find out on day sixty.
Daylight is one card, sent before the work starts, so you know where the room actually stands.
One linkA minute to answerNo accountNobody sees anyone else’s answers
There is rarely one reason.
Someone outranks the room. A dependency nobody wants to reopen. A suspicion the feature gets cut anyway. Half the meeting read the PRD in the meeting. And some doubts only arrive once you see the thing written down. Answering privately means you get the real read, not the polite one.
Three questions tell the story
Current state
How good is this today?
Everyone's read of the current state — usually the answer that surprises people most.
The success bar
How good does it need to be by the date?
The bar. Not everyone is working to the same one, and the card is where that shows.
Chance of success
Will this project be successful?
A percentage, scored later against what actually happened.
The card never averages. It shows every answer, and the median through them — 6 on its own tells you nothing; four people at 3 and five at 9 tells you the team is not seeing the same project. You can add up to five doubts, and a doubt you write yourself is shown back in your exact words.
What comes back
What comes back is the team’s real confidence in the project, the open questions and dependencies nobody raised out loud, and a chance to reset while that is still cheap. Later the card is graded against what actually happened.