Making Better Decisions Faster: Start Now or Wait?

Every day you make roughly 35,000 decisions, from the trivial (which socks) to the consequential (which job offer). Most run on autopilot, as they should. But the important ones — the ones you replay at 3am — tend to trigger one of two failure modes: paralysis (researching the seventeenth review, polling eleven friends, deciding to “sleep on it” for the sixth night) or impulse (deciding in a rush and rationalising later). The question “should I start making better decisions faster now, or wait until I’m ready?” contains its own answer, because decision-making is a skill improved only by deciding. This guide covers the science of deciding well — and quickly — without the recklessness.

Why Waiting Rarely Improves Decisions

The intuition says more time equals better decisions. The research says: up to a point, then sharply diminishing returns. For most decisions, 80% of the available information arrives in the first 20% of the research time; the rest is diminishing-returns detail that mainly serves to feel thorough. Worse, extended deliberation has active costs: decision fatigue (the quality of decisions degrades across a day of deciding — the famous parole-board studies, whatever their methodological debates, capture a real phenomenon), opportunity cost (the weeks spent deciding are weeks not acting), and preference distortion (overthinking can actually worsen choices — Dijksterhuis’s “unconscious thought” research found that for complex decisions, brief distraction outperformed prolonged conscious deliberation, because the conscious mind overweights salient-but-trivial factors).

The pro move is not “decide instantly” but time-box the decision: assign each decision a deliberation budget proportional to its reversibility and stakes, then decide when the budget expires.

The 10-10-10 Rule and the Reversibility Test

Two frameworks handle 90% of decisions. Jeff Bezos’s reversibility test (from his Amazon shareholder letters): sort decisions into Type 1 (one-way doors — hard to reverse: quitting a job, buying a house, having a child) and Type 2 (two-way doors — easily reversible: trying a new restaurant, a holiday destination, most purchases). Type 2 decisions should be made fast and with ~70% of the information — Bezos’s explicit threshold, on the grounds that waiting for 90% costs more in speed than it gains in accuracy, and most Type 2 decisions can be corrected. The great tragedy of decision-making is treating two-way doors like one-way doors: agonising for weeks over choices that could be unmade in an afternoon.

Suzy Welch’s 10-10-10: for the decision in front of you, ask how you will feel about it in 10 minutes, 10 months and 10 years. This recruits the long-term perspective that short-term emotion hijacks. The job offer that thrills for 10 minutes but looks thin at 10 months; the difficult conversation dreadful for 10 minutes but liberating at 10 years. It takes ninety seconds and cuts through more fog than a week of pros-and-cons lists.

Killing Analysis Paralysis: The Practical Toolkit

Set a decision deadline

“I will decide by Friday 5pm” converts open-ended agonising into a project with an end date. Parkinson’s Law applies to deliberation: research expands to fill the time available. Constrain the time and the research miraculously becomes sufficient.

Define “enough information” in advance

Before researching, write down what would decide it: “I’ll choose the supplier with the best 3-year total cost among the top three quotes.” Then stop when the criterion is met. Without a pre-defined stopping rule, research becomes a hobby.

Limit the options

The paradox of choice (Barry Schwartz’s research) is real: beyond ~3–5 options, more choice increases paralysis and decreases satisfaction (because the imagined perfect option haunts whichever you pick). Pre-filter ruthlessly to three finalists, then decide among those. Satisficing — choosing the first option that meets your criteria — beats maximising (seeking the absolute best) on both speed and long-term satisfaction.

Consult, don’t poll

Ask two or three people with relevant experience for their reasoning — not their vote. Polling eleven friends produces eleven anxieties and zero clarity, plus social pressure toward the consensus. You want perspectives that surface considerations you missed, not a referendum.

Pre-mortem the finalists

For each finalist option, imagine it is a year later and the choice failed spectacularly — then explain why. This prospective hindsight (Gary Klein’s pre-mortem technique) surfaces risks that optimistic deliberation suppresses, in about ten minutes. The MindTools decision-making toolkit covers pre-mortems and related techniques in depth. It is the fastest risk-analysis tool in existence.

Deciding Faster Without Deciding Worse

Speed and quality are not opposites when the speed comes from better process rather than less thought:

• Separate the decide from the doubt. Schedule a 20-minute “worry appointment” for post-decision second-guessing — and forbid it outside that window. Most post-decision anxiety evaporates when given a contained slot.
• Use your values as a filter. Decisions aligned with pre-clarified values (“I prioritise time with family over maximising income”) nearly make themselves. The slow decisions are usually values conflicts in disguise — resolve the values question once, and dozens of future decisions accelerate.
• Make small decisions tiny. The 35,000 daily micro-decisions should consume ~zero deliberation: standardise them (same breakfast, capsule wardrobe, default routes). Every micro-decision automated preserves decision quality for the ones that matter.
• Decide in your peak state. Important decisions deserve your best brain: morning, rested, fed, calm. Never decide big things exhausted, hungry, or at 11pm — the research on ego depletion may be contested, but nobody does their best thinking depleted.

Two adjacent skills compound decision quality: escaping the comparison trap removes the “but what will people think” distortion that slows so many decisions, and balancing full-time work with a side project is decision-making under real constraints, practised daily.

When to Slow Down (The Exceptions)

Faster is not always better. Slow down for genuine Type 1 one-way doors (the decision is hard to reverse AND high-stakes), when you are emotionally flooded (decide nothing important in the grip of strong emotion — sleep on it, literally once, then decide), when key information is imminently arriving (the survey results land Thursday — waiting three days is rational; waiting three months is not), and when the decision affects others significantly (consultation is respect, not dithering — but time-box the consultation too).

The skill is calibrating speed to reversibility: fast for two-way doors, deliberate for one-way doors, and honest about which is which. Most people have the calibration backwards — agonising over restaurant choices, rushing career moves.

Conclusion

Should you start making better decisions faster now or wait? Now — because the skill improves only through reps, and the frameworks above (reversibility test, 10-10-10, decision deadlines, pre-defined stopping rules, three finalists, pre-mortems) are usable in the next decision you face, today. Time-box the deliberation, satisfice rather than maximise, separate deciding from doubting, and reserve your slow careful thinking for genuine one-way doors. A year of faster, well-processed decisions compounds into a life that moves — while the waiters are still on review seventeen, deciding to sleep on it for the seventh night.

Frequently Asked Questions

How do I stop second-guessing after I decide?

Schedule the doubt: a 20-minute “worry appointment” the next day, outside which second-guessing is banned. Remind yourself that post-decision regret is largely imagination — you are comparing reality against a fantasised alternative, not a real one. And for Type 2 decisions, remember reversibility: most choices can be adjusted.

What if I make the wrong decision?

Distinguish wrong process from wrong outcome: a good decision process can still produce a bad outcome (luck exists), and that does not invalidate the process. Review the process, extract the lesson, adjust the system. People who treat every bad outcome as proof they “can’t decide” train themselves into paralysis; people who review processes get better.