Kill Your Darlings, Save Your Product

It begins with a sharper question than any business plan: Am I willing to rethink?

In a world moving faster than we can debug, Adam Grant’s Think Again is less a book and more a survival manual for the intellectually alive. It’s a compass for founders who don’t want to calcify. Who’d rather evolve than entrench.

We’re wired, he says, to play three roles: The preacher, defending sacred ideas. The prosecutor, picking holes in someone else’s. The politician, chasing approval over truth.

But rarely the scientist. The one who forms hypotheses, runs tests, welcomes disconfirmation. The one willing to toss out an idea when the data—or the users—say it's time.

Grant’s map has three terrains: self, others, and systems.

First: the self. The mind is a minefield. And certainty, it turns out, is often the biggest bomb. He revisits the Dunning-Kruger effect—how the most confident people are often the least competent. And how real learners don’t strut. They loop. They unlearn. Rethink. Rewire.

The goal isn’t to never be wrong. It’s to be wrong less—because you’re humble enough to keep adjusting. That’s the mental agility every builder needs.

Then: others. Persuasion isn’t about piling up facts until someone breaks. Grant shows how effective rethinking happens through questions, not lectures. Through motivational interviewing. Through empathy over ego.

Think of vaccine skeptics shifting their stance—not because someone yelled louder, but because someone listened longer. That’s the lever. Invite someone into rethinking, and they might surprise you. Corner them, and they’ll harden.

Finally: systems. Teams, startups, entire cultures collapse not because they’re stupid— —but because they stop questioning. They turn rituals into rules. Debate into danger. Groupthink into gospel.

Grant argues for something braver: Cultures of curiosity. Workplaces that reward “I don’t know” over fake certainty. Teams that debate to improve, not to dominate.

For founders, it’s gold:

• Treat opinions like versioned code—ready to patch or delete.

• Invite dissent into your standups.

• Design onboarding to reward curiosity, not compliance.

• Lead with humility, not bravado.

You can hear the echoes in Marc Andreessen’s path—getting knocked flat by Loudcloud, only to rethink, reboot, and help birth Netscape. Or in indie hackers who scrap product lines, rewrite business models, and ship based on user truth over founder myth.

Think Again doesn’t ask you to surrender conviction. It asks you to earn it—again and again.

To let go of your best idea when it stops working. To make peace with being wrong. To outgrow your own logic.

Tomorrow, when you're tempted to cling to certainty— Pause. Ask: What if the thing I’m defending isn’t the thing that’s true? And what would it look like to build from that?

Because your strongest edge as a founder might not be what you know. It might be your willingness to let it go.

And start again. Clearer. Better. Wiser.