**Most decisions receive far more stress than their consequences justify.**
Some decisions genuinely require days or even weeks of careful consideration.
But most decisions don't.
The mistake is applying roughly the same decision process to decisions with very different consequences. We often deliberate over minor operational experiments as though they carry the same consequences as major strategic commitments.
The consequences aren't the same.
Better decision making begins by asking one question:
**How difficult would this decision be to undo?**
A decision is reversible when it can be changed or undone at relatively low cost. Irreversible decisions, by contrast, are costly, difficult, or sometimes impossible to undo.
The degree of reversibility should determine the level of scrutiny, the speed of action, and **who should hold the decision-making authority.**
And that is the point where a decision-making problem becomes a question of decision architecture.
## Reversible vs. Irreversible Decisions: What's the Difference?
A **reversible decision** gives you the ability to change course at relatively low cost. You can decide, observe what happens, learn, and adjust.
Common examples include:
* Testing a new meeting format
* Changing a workflow
* Testing a small marketing initiative
* Piloting a process with one team
* Testing the delegation of a responsibility
* Testing a pricing variation
An **irreversible decision** creates consequences that cannot easily or cheaply be reversed.
Selling a company, signing a major long-term contract, making a large capital commitment, or risking permanent reputational damage are examples of decisions that may be difficult to reverse.
The principle is simple:
**As reversibility decreases, scrutiny should increase. As reversibility increases, authority should move toward the person closest to the relevant information.**
This is the foundation of what I call the **Decision Reversibility Framework.**
## Why Do We Overthink Reversible Decisions?
A decision may be reversible in reality while feeling permanent in the moment.
A manager considers changing a process and imagines everything that could go wrong.
A leader considers delegating authority and worries about losing control.
An entrepreneur considers launching an experiment and wants more evidence before committing.
Instead of acting, uncertainty creates a demand for more information.
One more conversation.
One more spreadsheet.
Another perspective.
Another layer of approval.
Eventually, the decision-making process costs more than making the wrong reversible decision would have cost.
The result is **decision latency**: the unnecessary gap between knowing enough to act and actually making the decision.
One slow decision may be insignificant. Hundreds of slow decisions compound.
And when reversible decisions repeatedly travel upward for approval, senior leadership becomes a bottleneck.
This does not necessarily make the organization safer.
It may have merely centralized uncertainty and delay.
## How the Decision Reversibility Framework Works
Before solving the decision itself, classify it using five questions.
### 1. How Reversible Is the Decision?
Don't simply ask whether reversal is possible.
Instead, determine the real cost of reversing course.
A meeting format can usually be changed easily. A company-wide technology platform may also be reversible in theory, but reversing it could create significant financial and operational costs.
Decisions aren't simply reversible or irreversible; reversibility exists on a continuum.
As reversal becomes more costly, the case for deeper deliberation becomes stronger.
### 2. What Is the Real Downside?
Separate consequences from discomfort.
A failed experiment might create two weeks of additional work.
A failed strategic bet can create serious financial, legal, operational, or reputational consequences.
Ask yourself:
**If we're wrong, what do we really lose?**
The decision making framework answer determines how much uncertainty you should tolerate.
When the downside is small, you can move faster.
When the downside is significant, greater scrutiny is appropriate.
### 3. When Will Reality Tell You Whether You're Right?
Sometimes acting generates better information than analyzing further.
A sales-script experiment can start producing feedback in a matter of days.
You can change a meeting format for a month and measure whether it improves productivity.
Fast feedback allows you to replace this process:
**Analyze → Analyze → Analyze → Decide**
and replace it with:
**Decide → Test → Learn → Adjust**
With reversible decisions, action is more than execution.
**Action becomes a source of information.**
### 4. Can We Limit the Downside?
A new decision doesn't need to be rolled out everywhere at the same time.
Start the workflow with a single team.
Offer the product to a small customer segment.
Run the policy for 30 days.
Delegate authority within clear boundaries.
Instead of trying to eliminate uncertainty before acting, reduce the consequences of being wrong.
**High uncertainty doesn't always require more analysis. Sometimes it requires a smaller bet.**
This can turn seemingly high-risk decisions into manageable experiments.
### 5. Who Should Really Own This Decision?
Many decision frameworks fail to ask this question.
If the downside is contained and the decision is easy to reverse, why should it travel to the top of the organization?
Often, it doesn't.
The person closest to the relevant information should usually have greater authority to act.
That changes the leadership question from:
**"What should I do?"**
to:
**"Should this decision actually belong to me?"**
That is a question about power.
## A Decision Architecture Matrix
Use reversibility and cost of error to determine the appropriate decision process.
| Decision Type | Reversibility | Cost of Error | Best Response | | |
| ---------------------- | ------------- | ------------- | -------------------------------- | ----------------------------------------------------- | ------------------------------------------- |
| **Experiment** | High | Low | Move quickly and test |
| **Delegated Decision** | High | Moderate | Set guardrails and delegate |
| **Strategic Bet** | Low | Moderate–High | Increase analysis before committing |
| **Critical Decision** | Very low | High | Slow the process and apply deeper scrutiny |
The mistake is using the highest level of scrutiny for almost every decision.
More analysis doesn't necessarily create better decisions.
Often, it simply delays action.
## When Should You Make the Decision?
If a decision can be reversed and failure is manageable, waiting for perfect information is usually unnecessary.
A practical rule is to move once you have sufficient information to make an informed choice and enough protection to absorb a mistake.
Jeff Bezos has famously described many decisions as reversible "two-way doors" and argued that decisions can often be made with roughly 70% of the information you wish you had.
The deeper principle is this:
**Reversible decisions don't require certainty. They require guardrails.**
If you can test cheaply, receive feedback quickly, and reverse course without serious damage, additional analysis may be less valuable than action.
Experimentation then becomes part of how the decision is made.
## Replace Approvals With Guardrails
For leaders, this is where the framework becomes especially consequential.
Suppose employees routinely bring decisions to you before taking action.
Your first conclusion might be:
**They need to stop asking for permission.**
But that hesitation may actually be rational.
If people don't know where their authority begins and ends, asking permission protects them.
The solution isn't simply encouraging confidence.
Change the architecture.
Instead of controlling each decision individually, establish clear boundaries of authority:
"You have authority to resolve customer issues below this financial limit."
"You can pilot process changes within your team for up to 30 days."
"You may run experiments as long as they remain within legal, security, and brand boundaries."
The boundaries of authority are now explicit.
Instead of constantly seeking permission, people can exercise judgment within an established decision territory.
**An approval system controls individual decisions.**
**Architecture determines how decisions are made repeatedly.**
One creates dependency. The other creates scale.
## A Two-Minute Test for Better Decisions
Before you start debating your next decision, give yourself two minutes to classify it.
Start with five questions:
**1. Can it be reversed?**
Can we change course without serious damage?
**2. What is the downside?**
Define the actual downside.
**3. How quickly will we know?**
Estimate when the decision will begin producing evidence.
**4. Can we limit the first commitment?**
Pilot, limit, stage, or test the commitment.
**5. Where should the authority sit?**
Move decision authority toward the best available information.
Then match the decision to the appropriate response:
**Easy to reverse + low cost of error → Decide.**
**Reversible + uncertainty → Experiment.**
**Easy to reverse + someone else has better information → Delegate.**
**Hard to reverse + high downside → Deliberate.**
Most decisions become easier once you stop treating them equally.
## Decision Speed and the Architecture of Power
The goal isn't to make every decision quickly.
That would confuse speed with judgment.
Better decision making means matching the process to the consequences.
Some decisions deserve research, dissent, scenario planning, and careful scrutiny.
Many others deserve a test.
The framework also reveals something deeper about leadership.
A leader whose approval is required for everything may appear to hold enormous power.
From a systems perspective, it may indicate the opposite.
The system has become dependent on your intervention.
Real organizational power isn't measured by the number of decisions that require a leader's approval.
It is demonstrated by whether the organization can repeatedly make good decisions **without requiring that leader's constant intervention.**
That is why reversibility matters at the organizational level, not just the personal one.
It determines where authority should sit.
It determines how quickly information becomes action.
It influences whether an organization is built around distributed judgment or centralized permission.
So before asking:
**"What should we decide?"**
ask:
**"How reversible is this decision—and who really needs the power to make it?"**
Most decisions aren't improved by adding more stress.
They need better architecture.
*The Architecture of POWER* reveals how authority, decision rights, incentives, information, and invisible organizational systems shape where power really sits.
**Power isn't just who gets to decide. It's how the system decides.**