Changing your mind without losing authority: how to revisit a decision before it gets expensive

A management mistake is rarely expensive on the day it is made. It becomes expensive months later, when the signs that something was off were already there and nobody dared to reopen the subject. A small decision held for too long costs more than a big one corrected in time.

Adam Grant wrote Think Again about that difficulty. Rethinking, he argues, matters more than raw intelligence, can be trained, and a manager can practice it without the team reading it as weakness.

Why reopening a decision is so hard

Grant borrows from Phil Tetlock three modes we fall into when we defend something: the preacher protects an idea as if it were sacred, the prosecutor proves the other side wrong, and the politician looks for approval. In all three we talk a lot and revise very little. The scientist, the fourth mode, treats opinions as hypotheses and decisions as experiments.

Grant sums it up in two cycles. In the overconfidence cycle, pride leads to conviction, conviction to biases, and biases to a search for validation. In the rethinking cycle, humility makes room for doubt, doubt for curiosity, and curiosity for discovery. A leadership team settled into the first one no longer sees the warning signs; it reads them in its own favor.

Practice oneTrust the method and doubt the answer

Grant calls confident humility having faith in your own ability to learn while doubting the tools you are using. A leader whose authority rests on the method can say that a new piece of data changed the reading and remain the same leader. One whose authority rests on having been right from the start is trapped in the first answer.

It takes two questions before any important decision: do I trust that my team and I can solve this, and do I trust the method we are using? If the second answer is less firm than the first, there is room to revisit.

Practice twoSeparate the opinion from the person holding it

The strongest block appears when a decision becomes part of who you are. Grant describes a totalitarian ego that filters out threatening information, and he proposes two detachments. The first separates your present self from your past self: whoever decided a year ago had different information, and saying so honestly takes the drama out of the change. The second separates your opinions from your identity, which is better anchored in values than in positions.

A director who defines herself by rigor or by care for her people can let go of a position without feeling she is letting go of part of herself. The forecasters Grant cites were more accurate the more often they updated their beliefs.

Practice threeDecide in advance what would change your mind

The most practical tool in the book for a management committee comes from a forecaster Grant describes: with every prediction he writes down the conditions under which it should hold and the conditions under which he would change his mind. Carried over to a management decision, that means writing down on the day of approval which signals would show the plan is not working, and agreeing on a date to look at them.

That way the person deciding commits to the method rather than to the outcome, and the team gets explicit permission to bring bad news, because that news already has a name. Reopening the decision stops being a defeat and becomes the fulfilment of an agreement the manager signed.

Practice fourBuild a challenge network

Almost every manager has a support network. Rethinking requires what Grant calls a challenge network: trusted critics, disagreeable and generous, who question because they care about the result. Think Again reminds us that people with less power tend to go quiet in front of the opinion of the highest-paid person in the room, so that network does not build itself; you have to ask for the criticism and protect it when it comes from the lowest rung.

It helps to call the discussion a debate, because people share more information, and to ask each proposal to explain how it would work, instead of why it is right, because that moderates extreme positions, including the boss's.

What happens to your authority

Grant admits in the epilogue of Think Again that audiences sometimes punish a leader who changes his mind as inconsistent. That is why the form matters as much as the substance, and what protects authority is fairly concrete:

• Say what changed. A new piece of data, a condition that was met, an assumption that did not hold. The team understands it was the evidence that moved the decision.
• Say what stays the same. The goal, the values and the method are unchanged; the route is what changes.
• Show moderate confidence. Grant cites that experts who admit some uncertainty come across as more credible than those who speak with total certainty.
• Share the criticism that helped you. Grant describes an experiment in which leaders who told their team about feedback that had helped them kept the openness going for far longer than those who only asked for opinions.