How to Use Randomness to Make Better Decisions
There is a specific kind of frustration that comes from being stuck between two options that seem equally good — or equally bad. You have weighed the pros and cons. You have talked it through. You have come back to it the next morning. Nothing has changed. This is decision paralysis, and it is one of the most common and draining cognitive experiences people have.
One of the oldest and least appreciated tools for breaking this kind of deadlock is also one of the simplest: a coin flip. Or, more broadly, deliberate randomness. Used thoughtfully, random selection is not an abdication of decision-making — it is a specific technique with genuine psychological value.
The coin flip as a preference detector
The most well-known use of randomness in decision-making is not to make the decision for you, but to reveal what you actually want. Here is how it works: flip a coin and assign each option to a side. The moment the result comes up, pay attention to your gut reaction. If you feel relieved, you have your answer — the coin confirmed what you already wanted. If you feel disappointed, you have your answer too — the coin just showed you what you were hoping it wouldn't say.
This technique is popular because it works. The reason it works is that for decisions where you genuinely feel torn, your rational analysis has often produced a genuine tie — the options really are close to equivalent on the factors you can measure. What the coin flip surfaces is the information that lives below the analytical layer: your emotional preference, your intuition, the thing you want that you haven't fully articulated yet.
Psychologist Barry Schwartz, in research on what he calls the "paradox of choice", found that when people are confronted with too many options of roughly equal value, the deliberation process itself becomes the problem. The cognitive cost of continuing to weigh options that are genuinely close in value is high, and the expected benefit of finding the marginally better option is low. At some point, deciding randomly and moving on is the better strategy.
When randomness is genuinely the right answer
There is a category of decision where randomness is not just a tiebreaker but actually the optimal approach. This is when the options are genuinely equivalent, when the stakes of choosing suboptimally are low and when the cost of continued deliberation is high.
The classic example is where to eat lunch. If you and a colleague are debating between two nearby restaurants and you have been discussing it for ten minutes without resolution, the expected cost of continuing to debate is already higher than the expected benefit of eventually choosing the marginally better option. Flip a coin, go to lunch, use the time you saved on something that matters.
More seriously, randomness has genuine value in business decisions involving genuine uncertainty. When you cannot predict which of two strategies will perform better, and you have limited resources to test both, random selection between them eliminates the bias introduced by advocacy, overconfidence and sunk-cost thinking. It also produces a cleaner result for later analysis: you can compare outcomes without wondering whether the chosen option was favoured by hidden bias in the selection process.
Breaking group deadlock
Decisions made by groups are particularly susceptible to specific dysfunctions: dominant personalities overriding quieter ones, anchoring on whoever spoke first, circular discussions that return to the same positions repeatedly. Random selection can interrupt these patterns.
One technique is to use a random number generator to assign decision weights randomly, then vote using those weights. This breaks the correlation between seniority or assertiveness and decision influence. Another approach is to generate a shortlist randomly — if you have ten candidate options, use a random process to reduce them to three, then deliberate only among the finalists. Reducing the decision space first makes the remaining deliberation more tractable.
Randomness as an antidote to bias
Human decision-making is subject to a long list of well-documented biases. Recency bias makes us weight recent information too heavily. Confirmation bias leads us to seek evidence for what we already believe. Status quo bias makes us favour the current situation over alternatives, even when the alternatives are objectively better. Familiarity bias makes us prefer things we recognise.
Random selection, by design, bypasses all of these. When you randomly choose which option to trial, which candidate to interview first or which market to enter, you remove the influence of these biases from the selection itself. This does not make the random choice better in any fundamental sense, but it makes it more defensible and more likely to surface opportunities that your biases would have screened out.
Randomness in creative work
Writers, designers and musicians have used deliberate randomness as a creative technique for decades. The technique of using random prompts, images or words to disrupt fixed thinking patterns is well-established in creative practice. The value is not the random input itself, but the way it forces you off the mental path you were already on.
If you are stuck in a creative project, generating a random constraint — work in a random colour, start from a random point in the text, develop the idea from a random angle — can break the loop. The constraint itself may not survive into the final work, but the act of responding to it often produces something that would not have emerged from unconstrained effort.
Use the Random Number Generator to generate a number in any range, or the Coin Flip for a binary choice. Both are useful tools for breaking decision deadlock cleanly and quickly.
When not to use randomness
Randomness is not appropriate for decisions with genuinely large stakes that are asymmetrically distributed — where the options are not actually equivalent and more analysis would reveal a clearly better choice. It is also not appropriate as a substitute for research, due diligence or professional advice. And it should not be used as an excuse to avoid accountability: "the coin told me to" is not a responsible answer when a real decision with real consequences was required.
Used properly, randomness is a tool for specific situations: genuine ties, low-stakes choices, creative deadlock and the removal of bias from processes where human judgement introduces unwanted variance. In those situations, it is genuinely useful. Outside them, the coin should stay in your pocket.
This article discusses decision-making techniques for general informational purposes. For significant financial, medical or legal decisions, consult a qualified professional.