Two-Way Door Decision
Two-Way Door Decision: Reversible decisions should be made fast, at the lowest level of the organization, with minimal process. Irreversible decisions warrant slow, careful deliberation. The most common failure is treating reversible decisions as if they were irreversible.
What Is the Two-Way Door Framework?
Jeff Bezos first articulated this distinction in his 2015 Amazon shareholder letter, where he described two types of decisions:
"There are decisions that are consequential and irreversible or nearly irreversible — one-way doors — and these decisions must be made methodically, carefully, slowly, with great deliberation and consultation. If you walk through and don't like what you see on the other side, you can't get back to where you were before. We can call these Type 1 decisions. But most decisions aren't like that — they are changeable, reversible — they're two-way doors. If you've made a suboptimal Type 2 decision, you don't have to live with the consequences for that long. You can reopen the door and go back through."
The framework addresses a specific organizational pathology: as companies grow, they tend to apply heavy process — committees, approvals, extensive analysis — to all decisions regardless of reversibility. This is appropriate for One-Way Door decisions and harmful for Two-Way Door decisions. The result is a large, slow organization that fails to move with the urgency of a startup on decisions that are actually low-risk.
The Two-Way Door framework is not just a classification tool — it's an argument for organizational design. Two-Way Door decisions should be delegated to small teams or individuals who are close to the information, with authority to decide and execute without senior approval. Only One-Way Door decisions should require escalation.
How It Works
For any significant decision you're facing, ask:
Question 1: If I make this decision and it turns out to be wrong,
can I reverse it or correct course with acceptable cost?
If YES → Two-Way Door:
- Make the decision quickly, at the lowest level of authority
that has sufficient information
- Use a bias toward action: gathering more information has
diminishing returns when the decision is reversible
- Set a time limit: "we'll decide by end of week"
- Launch, observe, adapt
If NO → One-Way Door:
- Slow down; gather more input and information
- Involve more stakeholders whose buy-in you'll need
- Conduct more thorough analysis
- Document the decision and rationale
- Consider what would need to be true for you to be confident
Examples:
Two-Way Door: testing a new onboarding flow, changing
a pricing tier, trying a new meeting format, hiring a
contractor for a project, launching in a new city
One-Way Door: acquiring a company, shutting down a product
line, relocating headquarters, making a key executive hire
at the VP level or above, signing a 10-year lease
Real-World Examples
Example 1: Amazon's Experimentation Culture
Amazon runs thousands of A/B tests simultaneously across its website and apps. Each test — a different checkout button, a new product recommendation algorithm, a revised email format — is a Two-Way Door decision. The test can be reversed instantly if it performs poorly. The cost of failure is the performance of one variant on one traffic segment for one test period.
This recognition allowed Amazon to build an experimentation culture where teams are empowered to test and learn without escalation. No senior approval is needed to run a well-instrumented A/B test, because the test is inherently reversible. The organization treats this as the category it is: Two-Way Door.
The contrast: when Amazon decided to acquire Whole Foods for $13.7 billion in 2017, the decision process involved years of due diligence, multiple executive reviews, and board approval. That is a One-Way Door decision and received appropriate process.
Example 2: Startup Pivot Assessment
A startup CEO is deciding whether to pivot from a B2C consumer app to a B2B enterprise product. The team has been discussing this for months without deciding.
Applying Two-Way Door analysis: is this reversible? Partially. Building enterprise features takes time. Reorienting the sales motion takes time. But if the pivot fails — if enterprise customers don't convert or churn quickly — the startup could realistically pivot back within 6–9 months, having learned something valuable.
The decision is closer to a Two-Way Door than it appears. The CEO's conclusion: run a 90-day enterprise sprint with two dedicated sales resources and a limited feature build. If three enterprise customers don't sign LOIs in 90 days, return to B2C. This time-boxed experiment converts what felt like an irreversible strategic pivot into a reversible test. The framework changes the action from "deliberate indefinitely" to "test quickly."
Example 3: Organizational Decision Hygiene
A 200-person software company has grown to the point where most decisions — pricing changes, new hires below VP level, feature launches, marketing campaigns — require VP or C-suite approval. Meeting calendars are full; decisions take weeks; teams are frustrated.
A new COO maps decisions across the One-Way Door / Two-Way Door axis. She finds that approximately 80% of decisions being escalated to VPs are genuinely reversible Two-Way Door decisions. She creates an explicit delegation policy: decisions with reversibility ≥6 months and financial impact less than $250K require only team-lead approval. VPs are escalated to only for decisions with lasting structural impact.
Decision speed for the two-way door category improves by 70%. Senior leader time is freed for the decisions that actually require their input. Team engagement improves as individuals feel trusted to act without constant supervision.
When to Use It
✅ Before any significant decision — ask the reversibility question first to determine the appropriate level of deliberation.
✅ When organizational decision-making has become slow — audit whether decisions are being over-processed because they're miscategorized as irreversible.
✅ When trying to instill a bias toward action in a team — use the framework to create explicit permission to move on reversible decisions without excessive approval.
✅ When evaluating risk tolerance for experiments — convert apparently risky decisions into time-boxed, reversible tests.
❌ As a rationalization for rushing irreversible decisions. The Two-Way Door framework is not an argument for speed in all situations — only for situations where the decision is genuinely reversible. Misclassifying a One-Way Door as a Two-Way Door to justify moving fast is the failure mode.
Model Combinations:
| Combine with | Effect |
|---|---|
| Pre-mortem | For One-Way Door decisions, run a pre-mortem before committing |
| Regret Minimization | Use regret minimization for One-Way Door decisions where values and identity are at stake |
| Satisficing | For Two-Way Door decisions, satisficing (good enough to act) is more appropriate than optimization |
Common Misuses and Limitations
Misuse 1: Treating everything as Two-Way Door to justify moving fast. The discipline of the framework is in honest assessment of reversibility. Calling an irreversible decision reversible to avoid deliberation is the most dangerous failure mode.
Misuse 2: Applying perfect reversibility as the test. Almost no decision is perfectly reversible — some cost is always incurred in changing course. The relevant test is whether reversal is feasible with acceptable cost, not whether it's free.
Limitation — reversibility changes over time: Some decisions start as Two-Way Doors and become One-Way Doors as commitments accumulate. A pricing change is easily reversible early; after customers have based business decisions on the price for a year, it becomes harder to change. The framework should be applied at the time of the decision, not statically.
Related Models
Precommitment: The deliberate use of One-Way Door behavior to create accountability and follow-through.
Satisficing: The appropriate decision process for Two-Way Door decisions — "good enough" is the right standard when you can correct course.
FAQ
How do I tell whether a decision is a One-Way or Two-Way Door in practice?
Ask two questions: (1) What is the cost of reversing this decision if it's wrong? Include direct costs (unwinding contracts, redoing work, lost time) and indirect costs (reputation, relationship damage, lost momentum). (2) What is the time horizon before reversal becomes impractical? Decisions that can be reversed within weeks or months with moderate cost are Two-Way Doors; decisions with permanent or multi-year consequences are One-Way Doors. When uncertain, err toward treating decisions as One-Way Doors — the cost of over-deliberating a Two-Way Door is lower than the cost of under-deliberating a One-Way Door.
Does this framework apply to personal decisions as well as organizational ones?
Yes. Taking a new job is usually a Two-Way Door (you can leave within a year if it's wrong). Accepting a marriage proposal is a One-Way Door (even if divorce is possible, the consequences of error are severe). Having children is definitively a One-Way Door. The framework helps calibrate how much deliberation a personal decision warrants.
Where did Bezos originally describe this framework?
Jeff Bezos described the One-Way Door / Two-Way Door distinction in his 2015 Amazon Annual Shareholder Letter, which is available free at ir.aboutamazon.com. He has also described it in interviews and it is developed in Colin Bryar and Bill Carr's Working Backwards (2021).
Apply This Model with AI
Describe the decision you're facing in MindMax. The AI will help you assess its reversibility, classify it as One-Way or Two-Way Door, and recommend the appropriate level of deliberation and process.
🚀 Apply the Two-Way Door framework in MindMax →
Further Reading
- Jeff Bezos, Amazon 2015 Annual Shareholder Letter — The primary source; available free at ir.aboutamazon.com.
- Colin Bryar and Bill Carr, Working Backwards (2021) — Provides operational detail on how Amazon applies this framework in practice.
This page is part of the MindMax Mental Models Knowledge Base.