Delegate any decision reversible within one week at low cost — retain only irreversible ones for direct judgment

Why This Is a Rule

Most people under-delegate because they use perceived importance as the delegation criterion: "This feels important, so I should handle it myself." But perceived importance is a poor predictor of delegation suitability. Many important decisions are also reversible — and reversible decisions are safe to delegate because a wrong choice can be corrected at low cost.

The irreversibility test replaces subjective importance with an objective criterion: can the decision be reversed within one week at low cost? If yes → delegate, because the worst-case scenario is a one-week reversal, which is almost always less expensive than your time spent on the decision. If no → retain, because a wrong delegation outcome can't be cheaply corrected.

This produces dramatically more delegation than importance-based reasoning. Tool choices, process decisions, meeting agendas, content drafts, vendor comparisons, scheduling — all reversible, all safe to delegate. The decisions that genuinely require your judgment are fewer than you think: strategic commitments, irreversible financial decisions, relationship-defining communications, and binding agreements.

When This Fires

• When deciding whether to handle a decision yourself or delegate it
• When your calendar is full of decisions that "only you" can make — most are probably reversible
• When building delegation frameworks for a team
• As a complement to Classify every decision as one-way or two-way door before deliberating — minutes for reversible, days for irreversible (one-way/two-way door) applied specifically to the delegation domain

Common Failure Mode

Retaining reversible decisions because they feel important: "I need to personally choose our new project management tool." This is a reversible decision (you can switch tools) that takes hours of your time that could be delegated. The perceived importance (it affects everyone) triggers the retention instinct, but the reversibility (tools can be changed) means delegation is safe. Even if the delegate chooses poorly, the worst outcome is switching tools — less costly than the hours you'd have spent evaluating.

The Protocol

(1) For each decision crossing your desk, ask: "If someone else makes this decision and gets it wrong, can it be reversed within one week at low cost?" (2) If yes → delegate. Don't assess importance, don't evaluate complexity, don't check if you'd do it better. The reversibility makes delegation safe regardless of these factors. (3) If no → retain for your direct judgment. Use the full decision toolkit (Classify every decision as one-way or two-way door before deliberating — minutes for reversible, days for irreversible through After 30+ journal entries, calculate your calibration — do your 70% predictions come true 70% of the time?) because the irreversibility warrants the investment. (4) Track your delegation rate: most knowledge workers should be delegating 60-80% of decisions that reach them. If you're retaining most decisions → you're likely misclassifying reversible decisions as irreversible. (5) When delegating, provide the delegate with the decision criteria (Every decision framework needs five explicit components: criteria, sequence, time budget, kill conditions, and decision rights) but not your preferred answer. Let them decide — that's the point of delegating.