Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Classify an engineering decision by its practical reversibility and the consequences of getting it wrong. A high-impact choice that is costly or nearly impossible to undo is Type 1 and merits deliberate review. A choice with a credible, low-risk rollback path is Type 2 and can usually be made quickly, with a clear plan for detecting and correcting problems.
What Type 1 and Type 2 mean
Jeff Bezos introduced the distinction in his 2015 Amazon shareholder letter, using one-way and two-way doors as metaphors for how difficult it is to return to the previous state.
As an Amazon Associate I earn from qualifying purchases.
| Decision type | Practical test | Appropriate process |
|---|---|---|
| Type 1 | Consequential and irreversible or nearly irreversible; reversing it would be difficult, expensive, slow, or disruptive. | Decide methodically, with careful deliberation and consultation. |
| Type 2 | Changeable, with a credible way to reverse or correct it if it goes badly. | Decide quickly, often through a responsible person or small group with good judgment. |
Bezos reiterated in his 2016 shareholder letter that teams should not use one decision process for every choice and should be able to correct bad decisions promptly. The letters offer a qualitative management principle, not an engineering-specific standard or a formula for assigning labels.
How to classify an engineering decision
- State the decision and its scope. Specify what will change, which systems or users are affected, and what is outside the decision.
- Write down the realistic rollback path. Identify how to restore the prior state, how long it would take, who must coordinate, and whether data, compatibility, or customer impact would complicate recovery.
- Assess the cost of being wrong. Consider the likely blast radius, safety or regulatory exposure, external commitments, and the consequences for dependent services. Separately consider the cost of waiting; urgency does not make a hard-to-reverse decision reversible.
- Look for a smaller commitment. A prototype, limited experiment, staged rollout, feature flag, or compatibility boundary may make correction more feasible. Count it as a real rollback path only if the team can use it in practice and the effects during the trial are acceptable.
- Match review to the risk. Use broader consultation and deliberate review for high-consequence decisions that are difficult to undo. For a genuinely reversible choice, a lightweight decision by the responsible person or a small group may be enough.
- For a Type 2 decision, set the correction trigger. Name the signal that would prompt rollback or another change, and identify who has authority to act.
- Reassess when circumstances change. New dependencies, external adoption, accumulated data, or additional commitments can make reversal harder; reclassify the decision if the original assumptions no longer hold.
Judge practical reversibility, not the label
Technical reversibility is not the same as practical reversibility. An API change may be easy to revert in a code repository but costly after external customers have built integrations against it. A database migration may have a documented reverse script, yet restoring the old state could be risky after new writes or dependent changes have accumulated.
#1 Best Overall
The reverse can also be true: an important architecture choice may become easier to correct if it is introduced incrementally or separated behind a clean compatibility boundary. These examples apply the door metaphor to engineering; they are not examples or a checklist supplied in the shareholder letters.
Common classification mistakes
- Treating every major technical choice as Type 1. A decision can be important yet still have a safe, credible rollback path. If so, the team may be able to learn through a bounded change rather than wait for exhaustive agreement.
- Calling a choice Type 2 because it is reversible in theory. If recovery requires lengthy coordination, loses data, disrupts customers, or depends on conditions that may not hold, the practical cost of reversal may be high.
- Skipping monitoring after a quick decision. A Type 2 process depends on noticing when the choice is failing and being able to correct it. Decide what to watch and who responds before the change is made.
- Keeping a classification after the risk changes. Once a choice creates dependencies or outside commitments, the cost of undoing it may rise. The process should reflect the current situation, not the original label.
What the framework does—and does not—claim
Bezos argues that applying a heavyweight process to reversible decisions can slow teams and inhibit experimentation. That is his management rationale, not empirical proof in the letters that faster Type 2 decisions improve engineering outcomes. The letters do not provide numeric thresholds, an exhaustive list of engineering decisions by type, or a validated scoring system. Teams can use practical questions about rollback cost, recovery time, detection, blast radius, data and interface reversibility, external commitments, and consultation burden without treating those questions as a formal score.
Quick Recap
Rank #4
Rank #3
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problems
