The feedback your idea can survive.

You did the work. You didn't just sit in your own head or ask your mum — you went out and talked to 10 people about the idea. You took notes. And somehow, every single conversation left the idea exactly as intact as it started. No wounds. No rethink. Just a stack of nods.

That's not luck, and it's not proof you're right. You chose feedback your idea could survive.

How the selection happens

When you're attached to an idea, you don't go hunting for the test most likely to kill it. You do the opposite, usually without noticing. You ask the people most likely to be kind. You ask the questions least likely to land on the soft spot. And you frame the whole thing in a way that already assumes the idea is good — "how would you use this?" instead of "would you use this — and if the problem's real, why haven't you already solved it?"

Each of those choices is small and reasonable on its own. Together, they build a process that cannot return a no. It's the same closed loop as grading your own work — only now it's wearing the costume of openness. You can point at your notes. You did customer research. You just quietly designed it so the customer couldn't disappoint you.

The tell

Here's how you catch it: if every conversation confirmed you, you weren't testing anything. You were selecting. A real test has the capacity to surprise you, which means a real research process produces at least some findings you didn't want. If yours never changes your mind, it isn't measuring the idea — it's just collecting reassurance and filing it under 'evidence'.

The questions you're steering around

You already know which questions you're avoiding, because they're the ones whose worst answer would actually hurt:

• Would you pay for this today — and how much?
• What would stop you from using it?
• Who's already solving this for you, and why would you switch?
• How are you handling this right now — and is it actually a problem worth money, or just a mild annoyance?

These are the questions that can hurt, which is exactly why they're the ones worth asking. The soft question protects the idea. The sharp one tests it. And you can feel, before you even ask, which is which — the flinch is the signal.

So the discipline isn't "get more feedback". You can get a hundred more nods and learn nothing. It's to deliberately seek the feedback you'd least like to hear, from the person most likely to give it to you straight, and to ask the one question whose worst answer scares you. The founders who actually learn something aren't the ones who asked the most people. They're the ones who asked the question they were afraid of.