Logos52
wiki / Decision Making / Changing Decisions

Changing Decisions

Concept updated 2026-09-11

Changing Decisions

Changing a decision is revising an earlier choice when the facts have changed. The method is to write the original reasons for the choice next to the facts now known, then ask whether the same choice would be made today. Staying with the decision is not automatically better than revising it. The choice to stay has to be justified against the new facts.

Core takeaways

  • Persistence is useful when the decision still fits current goals, values, and priorities. Rigidity is staying for some other reason.
  • The question to ask is whether the decision still makes sense with the information now available.
  • Later decisions can be better than earlier ones because they can include more information. The time a decision has been in place is not evidence for it.
  • Four biases produce a stay that is no longer a choice: status quo, sunk cost, loss aversion, and identity lock-in.
  • A seven-question review compares the cost of staying with the cost of changing and ends with the condition that would force another change.
  • Revising a decision is a skill. At the moment of a change, write down what would cause the next one.

Reasons to stay and reasons to change

The question comes up at the moment when it is unclear whether to continue, revise, or stop. Persistence is useful when the decision still aligns with goals, values, and priorities. Without that condition, every choice to stay would count as a failure.

Rigidity appears when staying is driven by something else: changing feels embarrassing; effort has already been invested; the current path is familiar; loss feels more painful than future upside; or consistency is being confused with good judgment. Embarrassment and identity are the two causes people recognize least often. Invested effort is the sunk cost bias. Familiarity is the status quo bias. Loss that feels larger than the upside is loss aversion.

The useful question is: does this still make sense with the information I have now?

Changing can be better when new information changes expected value; when the downside is larger than first understood; when the original assumptions were wrong; when the opportunity landscape has changed; when the decision no longer fits current goals; or when staying mostly protects comfort rather than purpose. The last reason applies even when no new information has arrived.

Later decisions can be better than earlier ones because later decisions can contain more information. A decision does not become more likely to be right because it has been in place for a long time. New facts change the value of the options that remain. The length of time already spent on a decision is not evidence for it. “We have already come this far” is not a reason to stay. Whether changing improves position or expected value belongs to Positional Decisions and Expected Value.

Four biases that keep a decision in place

Four checks identify a decision to stay that is no longer a deliberate choice.

  • Status quo: staying because the decision is already in motion, not because it is better.
  • Sunk cost: would this be chosen again if time, money, or identity had not already been invested? Identity counts as something invested.
  • Loss aversion: giving the pain of changing more weight than the benefit of adapting. In laboratory studies the ratio is near 2:1. Larger ratios that are often repeated are not what the studies found.
  • Identity lock-in: protecting an image of consistency rather than the goal.

Noticing one of these four while it is happening is a Self-Regulation problem.

The seven-question review

The review asks seven questions: what was the original decision; what information changed; which assumptions no longer hold; what is the cost of staying; what is the cost of changing; which option improves future position; what would force another change. The answer to the last question is the condition that would end the new decision. Judging the process against the outcome after the change belongs to Good Decisions.

Revising as a skill

Revising a decision when the facts change is a skill. A revision is not a failure. At the moment of the change, write down what would cause the next change. Use the same sheet of reasons, with a future date. Kolbs Experiential Cycle is the review of the old choice after the change: what happened, what rule came out of it, and what changes as a result.

How to practice this

  1. Take one decision you are currently staying with. Write its original reasons on a page. Next to them, write the facts you know now. Notice whether you would make the same choice today.
  2. Ask whether the decision still fits your current goals, values, and priorities. If it does, staying is persistence. If it does not, notice what else is keeping you there.
  3. Run the four checks on that decision: status quo, sunk cost, loss aversion, identity lock-in. Notice whether you would choose it again if no time, money, or identity had been invested.
  4. Answer the seven review questions in order, from the original decision to what would force another change. Notice how the cost of staying compares with the cost of changing.
  5. If you change the decision, write down at that moment what would cause the next change. Use the same sheet of reasons, with a future date.
  6. After the change, review the old choice. Note what happened, what rule came out of it, and what changes as a result.

Open questions

Which current commitments should have explicit change criteria?

What signs distinguish persistence from rigidity in the study system?

Sources

  • Arkes & Blumer 1985: sunk cost. People stay on a path because of what they have already spent on it.
  • Samuelson & Zeckhauser 1988; Kahneman, Knetsch & Thaler 1991: status quo. People stay because the decision is already in motion.
  • Kahneman & Tversky 1979: loss aversion. The typical laboratory coefficient is near 2:1. A 3–10× range is not what the study found.
  • Duke, Quit (2022); CNBC 2023-06-11 interview: quitting can be the skilled move. Grit without an exit test keeps a person on a decision too long.

Linked from

Decision MakingThe first form has been measured. People disproportionately keep the option they already have, and the tendency grows with the number of alternatives. The fix, a four-bias check and a seven-step change review, is on Changing Decisions. The second form comes from how information aJudging a Decision by Its ProcessA process review asks six things. What was known at the time? What was uncertain? What downside was protected? What option improved the position? That fourth question is covered on Positional Decisions and Expected Value. What would change my mind? That fifth question continues oPositional Decisions and Expected ValuePast two or three predicted replies, the variance grows too large to work with. Stop simulating the whole decision tree. Take the move that stays strong whatever the reply, and that you could afford to repeat. Score it by the options, information, and ruin it leaves for the nextThe Uncertainty-Opportunity TradeoffThe current path is easier to price because it has a record. Its trend is the estimate. The same approach producing flat or worsening results while the effort behind it rises is a path that cannot be sustained. The trend also says when that point arrives. In the model's teachingLoss AversionWhen the decision is live, your first feeling favours the option where nothing is lost. That feeling is not neutral information about which option is better. It systematically misses a gain that was never priced. A decision made from fear and avoidance is a decision that feeling
1 more
ICS Program Map- Decision-making → Decision Making · judging a decision by its process · Changing Decisions · Choice Throttling · Decisional Delays · Positional Decisions and Expected Value