You can test almost everything about your product yourself. You can't test whether it says what you meant it to say. That last one needs someone who isn't you, and most founders never get them.

Here's what I mean. You can check that the signup works. That the payment goes through. That the email sends, the page loads, the export downloads. All of that is QA you can do alone, because there's a right answer and you can see it. It either works or it doesn't.

But there's a second layer that hides in plain sight: whether the thing you built actually communicates what you intended. Does the confirmation email sound the way you'd want a stranger's first impression to sound? Does the headline say what you think it says, or what you've come to read into it after staring at it for six months? Does the result your product hands a user mean to them what it means to you? You can't QA that alone, because you're the one person on earth who already knows what it was supposed to say.

Why your own eyes can't see it

You built it from the inside. Every label, every line of copy, every default — you wrote it knowing what it meant. So when you read it back, you don't read the words on the screen. You read the intention behind them, which is invisible to everyone else.

This is why your own typos survive ten passes and a stranger catches them in ten seconds. It's not carelessness. Your brain is helpfully autocorrecting to the version in your head. The same thing happens, at a much higher level, with meaning. You see the product you meant to build. The user sees the product you actually built. Those are two different objects, and you are structurally incapable of seeing the second one, because you can't unknow the first.

The values version of this is the dangerous one. You believe your product is honest, or generous, or respectful of the user's time, because you intended it to be. Whether it reads that way to someone encountering it cold is a separate fact, and your conviction about your own intentions has no bearing on it — the same way your own enthusiasm can't validate your idea.

The surfaces where it matters most

The cost concentrates on the surfaces that travel — the ones a stranger meets without you in the room to explain.

The share card. The confirmation screen. The first sentence a new user reads. The result that gets forwarded to someone who's never seen your product. You built every one of those from inside the product's logic, in the product's language, and to you they read perfectly. To the person receiving them, they might read as cold, or confusing, or quietly off — saying something you never meant and would be horrified to learn you'd said.

I've seen a product whose proudest moment — the thing it most wanted to celebrate with the user — landed as a database message: technically correct, emotionally blank, speaking the system's language at the exact moment a human wanted to feel something. The founder couldn't see it. He'd written it. To him it said "congratulations". To everyone else it said "operation complete". It took someone who had never seen the code to notice, and once he saw it, he couldn't unsee it.

"You can QA the function by yourself. The meaning needs a stranger."

What to do with this

Get one person who has never seen the inside of your product to read the surfaces that travel, before launch, not after. Not to test whether they work — to tell you what they say. Hand them the share card, the confirmation, the first screen, the exported result, and ask one question: what does this tell you, and how does it make you feel?

Don't defend the answers. Don't explain what you meant. The moment you explain, you've contaminated the test, because you've put your intention back in the room, and the whole point was to find out what survives without it. Just listen for the gap between what you meant and what they received. That gap is the bug — the one kind you genuinely cannot find alone.

This is the same reason a second opinion is worth more than a more careful first one. It isn't that the outsider is smarter or more careful than you. It's that they can see the thing you built instead of the thing you meant, and that's a view you will never have of your own work. You can QA the function by yourself. The meaning needs a stranger.

The founders who ship something that says what they meant aren't the ones with the best taste. They're the ones who went and found the person they couldn't be.

Want more like this? Rick writes about the go/no-go decision, founder counterintuitions, and the business of building ventures worth building.

All writing