Being wrong is a skill

Hey Warblers,

You know the fastest way to lose credibility at work? Never be wrong. Never change your mind. Project total certainty at all times. That playbook sounds safe. It's not real. It's actually the most dangerous thing you can do for your career.

Two massive layoffs hit tech this month with the same playbook. Block's Jack Dorsey cut 4,000 people, nearly 40% of his workforce, and wrapped it in total conviction: AI changes everything, smaller teams do more, "we're ahead of the curve." Days later, Atlassian's Mike Cannon-Brookes cut 1,600, framing it as choosing to adapt "thoughtfully, decisively and quickly." Two CEOs cutting thousands of people. Same move: wrap the hard call in certainty so nobody questions what you're actually unsure about.

Here's what neither said: "We hired too aggressively. Our predictions about growth were wrong. The market shifted faster than we expected, and we're making painful corrections." That would have been the honest version. Dorsey in particular had a chance to be real. Instead he declared a new era of work while analysts openly asked whether AI was the strategy or the scapegoat. Cannon-Brookes said on a podcast just five months earlier that Atlassian would hire more engineers, not fewer. No acknowledgment of the reversal.

The result in both cases: employees and investors trust the decision less, not more, because the framing didn't match reality. And that's the lesson worth carrying into your own career.

Dorsey and Cannon-Brookes made the same leadership mistake most of us make every day, just at a bigger scale. They performed certainty instead of sharing their thinking. And it cost them trust the moment reality didn't match the confidence.

There's an Amazon leadership principle that gets at exactly this tension. It's called "Are Right, A Lot." Most people read it as: have the right answer. Be the smartest person in the room. Nail the call.

That's exactly wrong.

The full text says leaders "seek diverse perspectives and work to disconfirm their beliefs." The principle isn't about having the right answer. It's about having the right process for getting to answers, which includes being visibly wrong along the way.

Here's what most people miss: the people who advance fastest at places like Amazon, Google, and Meta aren't the ones who are always right. They're the ones who are wrong out loud, learn fast, and make their thinking visible.

What certainty actually costs you

Trust isn't just competence plus reliability. There's a third ingredient most people miss: vulnerability. Without it, you're respected but not trusted. You get compliance from your team, not candor. You get head nods in meetings, not honest pushback. And you have no idea what people actually think of your judgment.

Here's what that missing ingredient costs you with the three audiences that matter most:

The deeper problem: your work is visible but your thinking is invisible. Design reviews show the output, not the 14 approaches you rejected. Strategy docs show the recommendation, not the three times you changed your mind getting there. Promo packets show results, not judgment.

The thing that actually gets you promoted is the thing nobody can see. Unless you make it visible on purpose.

Amy Edmondson's research at Harvard has shown for decades that teams where people openly discuss mistakes outperform teams that don't. Not marginally. Significantly. The teams that report more errors actually have better outcomes, because they're learning in real time rather than hiding problems until they compound. Google's Project Aristotle confirmed the same finding: psychological safety was the number one predictor of team effectiveness. Not talent. Not resources. The shared belief that you could take risks without punishment.

The tell: Watch who gets promoted into leadership roles. It's rarely the person who was right about everything. It's the person whose thinking process was trusted by the people in the room where decisions get made.

Why this is so hard

At most big companies, the performance review system punishes visible failure and rewards visible wins. You're trained to curate a highlight reel for your self-review, not a learning log. The incentive structure says: hide your wrong calls, amplify your right ones.

This creates a workplace full of people who appear certain about everything and are trusted for nothing. Their manager can't tell the difference between genuine confidence and performance confidence. Neither can the promo committee.

The paradox: the people who seem most sure of themselves are often the least trusted in calibration rooms. Because decision-makers can't evaluate judgment they've never seen in action.

The system to build

Start a "thinking log" that you share selectively. Not a public journal. A running document you reference in 1:1s, design reviews, and strategy discussions. Here's the template:

Use whichever version fits the moment. The point isn't to broadcast every mistake. It's to make your reasoning visible before the outcome is known, surface problems before someone else finds them, and share lessons after you've corrected course.

Over 90 days, this does three things:

Same work. Same outcomes. Completely different reputation.

Try this prompt in ChatGPT or Claude: After your next big decision at work, paste your reasoning into an AI chat and ask: "Play devil's advocate. What are the three strongest arguments that this decision is wrong, and what early signals would I see if it's failing?" Save the output in your thinking log. You've just built a pre-mortem in 90 seconds, and you'll have a record of your reasoning that separates decision quality from outcome quality.

Your team doesn't need you to be right about everything. They need to see how you think. Show them.

~ Warbler