Some people treat decisions the same way. They spend three weeks deliberating a software purchase and three weeks deliberating a market entry. Everything gets the same level of analysis, the same amount of worry, the same paralysis.

The Reversible vs. Irreversible framework introduces one distinction that eliminates most decision paralysis: can this decision be undone? If yes, it’s reversible. Decide fast. Move. Learn the outcome. Adjust. If not, it’s irreversible. Slow down. Analyze. Stress-test. Get it right before committing. That’s the entire framework. One question. Two categories. Completely different decision protocols for each.

Where it comes from:

• Jeff Bezos formalized this as Amazon’s decision-making philosophy, calling reversible decisions “two-way doors” and irreversible decisions “one-way doors.” He wrote about it in his 1997 shareholder letter and has referenced it repeatedly since.
• His argument: Most companies slow down as they grow because they apply one-way-door analysis to two-way-door analysis. Every choice gets run through committees, reviews, approvals. But most decisions are two-way doors: you walk through, see what’s on the other side, and walk back if you don’t like it. Only a small percentage are one-way doors where you can’t return. Treating every decision like a one-way door creates organizational paralysis.
• The concept itself predates Bezos. Military decision-making has always distinguished between “commitment decisions” (deploying forces, which is hard to reverse) and “shaping decisions” (sending scouts, which is easily reversed). Venture capitalists distinguish between decisions that burn capital permanently and decisions that preserve optionality. Options theory in finance is built entirely on the value of keeping doors open.
• The main idea: identify which type of door you’re facing, then match your decision speed to the door type.

Why it matters?

• For executives, this framework is liberating. Most of the decisions that keep an executive up at night are actually reversible. Hiring a contractor, testing a new channel, launching a pilot, trying a new pricing tier. She’s spending emotional and analytical resources on decisions that could be made in an hour, tested in a week, and reversed in a month.
• This framework helps her see which decisions are actually one-way doors and which ones she’s treating as one-way doors because of anxiety, perfectionism, organizational culture, or fear of being wrong publicly.
• Learning to identify which doors are actually two-way, and giving yourself permission to walk through them fast and walk back if needed, is not just an executive tool but also a nervous system tool.

The key disciplines: 

• Default to reversible. Most decisions are more reversible than they feel in the moment. Hiring someone feels irreversible but you can part ways. Launching a product feature feels irreversible but you can remove it. Choosing a niche feels irreversible but you can expand later. The question isn’t “can this be undone perfectly” but “can this be undone at an acceptable cost?”
• Identify what makes a decision truly irreversible. Some decisions genuinely cannot be undone or can only be undone at catastrophic cost. Selling a company. Signing a ten-year lease. Publicly announcing a merger before due diligence. Sharing a confidential information. Burning a key relationship. These deserve full analysis.
• Match the process to the door type. For two-way doors: small teams or individual decides, fast timeline, bias toward action, plan to evaluate and reverse if needed. For one-way doors: broader input, more analysis, scenario planning, pre-mortem, stakeholder mapping. Two-way doors need speed, not frameworks.
• Watch for false one-way doors. These are decisions that feel irreversible but aren’t. They’re the most common source of organizational and personal paralysis. The feeling of irreversibility comes from ego (if I reverse this, I’ll look indecisive), from sunk cost (we’ve already invested so much), from identity (this choice defines who I am), or from catastrophizing (if this goes wrong, everything collapses). The main job is to test the feeling against reality.
• Watch for hidden one-way doors. The reverse also happens: decisions that seem easily reversible but actually aren’t. Hiring the wrong senior leader is theoretically reversible but practically devastating because of the damage done before you realize the mistake. Launching a half-baked product is theoretically reversible but the reputation damage may persist. These deserve more analysis than their apparent reversibility suggests.

In practice, decisions aren’t purely reversible or irreversible. They sit on a spectrum. The relevant dimensions are:

• Cost of reversal. Can you undo it for free, or does undoing it cost time, money, reputation, or relationships?
• Speed of reversal. Can you undo it in a day, a month, or a year?
• Completeness of reversal. Can you fully return to the starting point or does some consequences persist even after reversal?
• Information gained. Even if you reverse the decision, did you learn something valuable from the experience that you couldn’t have learned any other way?

The mechanics. A decision with low reversal cost, fast reversal speed, full reversal completeness, and high information gain is deeply two-way. Decide in minutes. A decision with high reversal cost, slow reversal speed, incomplete reversal, and low information gain is deeply one-way. Bring out the heavy thinking and decision-making frameworks. Most decisions fall somewhere in between, and the analysis is about figuring out where on the spectrum they sit.

How organizations use it:

• Amazon structures its entire decision-making culture around this distinction. Two-way door decisions are pushed down to small teams who can decide and act without senior approval. One-way decisions get escalated to senior leadership with full analysis.
• Military operations distinguish between reconnaissance (reversible: send scouts, gather information, withdraw if needed) and commitments (irreversible: launch the offensive). The entire concept of “probing,” small, reversible moves that generate information, is built on this framework.
• Startups implicitly use this when they talk about MVPs and iteration. Launch something minimal, see how the market responds, adjust. Every iteration is a two-way door. The one-way door is raising a large funding round or signing an enterprise contract with specific deliverables.

Common pitfalls:

1. Treating all decisions as one-way to avoid accountability. If every decision requires full analysis, nobody can be blamed for being slow. This is organizational fear disguised as rigor.
2. Treating all decisions at two-way to avoid analysis. The opposite failure, moving fast on everything, including decisions that genuinely require careful thought. This is recklessness disguised as agility.
3. Confusing emotional weight with irreversibility. A decision can feel enormous and still be reversible. Choosing which client to approach first feels like it defines your entire positioning. It doesn’t. You can approach a different client next month. The emotional weight is real but it doesn’t change the door type.
4. Ignoring the cost of not deciding. Sometimes the cost of analysis exceeds the cost of being wrong. If you spend three months analyzing a reversible decision, you’ve lost three months of learning that only comes from action. The cost of delay is invisible but real.