Tool/Meta-Concept/No. 0461
Heuristic
A heuristic is a practical rule that cuts the effort needed to solve a problem or make a judgment. In math and psychology, George Pólya and Herbert Simon studied how heuristics simplify search or calculation. Their accuracy depends on how well the rule fits the task and setting.
Also called Heuristics
- Evidence
- Well established
- Read
- 6 min
- Links
- 11 connections
01You've seen this when…
- in life
It’s Wednesday evening and you need dinner. You scan recipes until one takes under thirty minutes and uses ingredients already in the fridge. You stop looking.
- at work
A technician faces a machine that keeps shutting down. She checks the fault that caused the last two shutdowns before working through the full diagnostic checklist.
- out in the world
A library orders extra copies when a book’s reservation queue crosses a set threshold. Staff can handle routine purchases without reviewing every title at a meeting.
02The idea
Choosing a hotel could involve comparing every room, review, price and route to the station. A simple rule shrinks the job: consider places within a ten-minute walk, reject those below a review threshold, and take the first affordable option.
That rule is a heuristic. It saves effort by selecting what to examine, how to judge it, or when to stop. Its value comes from the relationship between the shortcut and the task. Distance is a useful cue when catching an early train. It contributes little to choosing a quiet writing retreat.
Heuristics take several forms. Satisficing sets an acceptable standard and stops when an option meets it. The availability heuristic uses ease of recall to judge frequency or likelihood. The representativeness heuristic judges by resemblance to a familiar category. Each reduces a different kind of work.
Some heuristics are deliberate rules; others operate with little awareness. A heuristic can also be implemented as an algorithm. A precise sequence of steps can still produce an approximate answer.
Heuristic names the procedure. Cognitive bias names a systematic tendency toward error. The same procedure can perform well in one setting and produce biased judgments in another.
03How to use it
- Define the task and the cost of error. Write down what a satisfactory result must accomplish. Include hard constraints, such as a budget, deadline or safety requirement. A rule for choosing lunch can tolerate errors that a rule for detecting equipment failure cannot.
- Choose a few informative cues. Look for facts that usually change the decision. For a delivery service, whether a package is perishable may matter more than its purchase price. Specify how the cues lead to an action. A fast-and-frugal tree does this through a short sequence of questions, each of which can lead to a decision.
- Test the rule against relevant cases. Try it on past decisions or a small, reversible batch. Compare it with the current method, including errors, time and effort. Check awkward cases separately: a rule that handles routine requests smoothly may miss rare, costly exceptions.
- Set a stopping rule and an escape route. Decide when the available information is enough, and which signals trigger deeper investigation. Recheck the rule when prices, customers, technology or operating conditions change.
The aim is a useful speed-accuracy trade-off. Sometimes an extra minute buys a large improvement; sometimes it changes almost nothing. Value of information helps frame that choice: how much would another check improve the decision, compared with what the check costs?
For repeated tasks, write the rule down. That makes its assumptions visible and lets other people test it.
04A worked example
Imagine a small online shop with one support agent and thirty requests waiting at 9 a.m. This is an illustrative example. The agent could estimate the financial impact, urgency and effort of every request before starting. Instead, the team tries a short triage rule: investigate failures affecting checkout first; then paid orders at risk of missing today’s shipping deadline; then handle the remaining requests from oldest to newest.
What it looks like The agent moves a checkout error ahead of an older request to change an account photo. A shipping-address correction comes next because the package leaves at noon. Several routine requests wait.
What’s actually going on The rule uses two cues that connect closely to immediate business consequences: customers’ ability to purchase and the shipping deadline. It avoids assigning a detailed score to every request. Age provides a simple fallback that keeps routine work moving. The shortcut also has blind spots: an unfamiliar complaint could signal a larger problem than its category suggests.
What made it work In this example, the team gives the agent an exception route for unusual or unclear cases. It reviews a sample of completed requests to see whether important problems were delayed. If a new kind of failure appears repeatedly, the team changes the rule. The heuristic earns continued use through those checks; its simplicity alone says little about its quality.
05When to reach for it
Heuristics can also guide the first step of a difficult problem: draw a diagram, break it into parts, or try a simpler case. That use directs exploration without committing to a final answer.
06When it misleads
- The setting changes. A familiar cue can lose its predictive value. Last year’s purchasing rule may perform poorly after supply shortages or a change in customers. Ecological rationality focuses on this fit between a rule and its environment.
- The easy cue replaces the relevant one. Memorable accidents can dominate a risk estimate even when reliable frequency data is available. Similarity to a successful hire can distract from evidence about job performance.
- Rare failures carry large costs. A shortcut that succeeds on most routine cases may still be unsuitable for safety-critical decisions. Such uses need validation, monitoring and an escalation process proportionate to the consequences.
- The rule escapes scrutiny. Intuition can feel persuasive while its accuracy remains unknown. Record predictions or decisions and check their outcomes. Include the time saved when comparing methods, as well as the errors made.
More elaborate analysis also has costs and can introduce mistakes. The practical comparison is between methods people can actually carry out with the available information and time.
07Roots
In 1945, mathematician George Pólya published How to Solve It, a book about the gap between understanding a finished solution and finding one. A student staring at a blank page needed a next move. Pólya offered prompts such as drawing a figure, recalling a related problem and solving a simpler version. He organized this guidance around understanding the problem, devising a plan, carrying it out and looking back. These were aids to discovery, with room for trial and revision.
Herbert Simon approached a related problem through decision-making. People choosing among jobs, purchases or business plans have limited information and limited computing capacity. His 1955 model of rational choice helped formalize what became bounded rationality: decision procedures shaped by those limits. Seeking an option that clears an acceptable threshold became a central example.
In the 1970s, Amos Tversky and Daniel Kahneman brought heuristics into judgment research by showing how shortcuts could produce predictable errors. Later, Gerd Gigerenzer and colleagues emphasized how simple rules can perform well when their cues match the environment. Together, these traditions made the key question concrete: which shortcut works for this task, under these conditions?
08How solid is this?
Research documents both effective heuristic decision-making and systematic errors from particular shortcuts. Performance depends on the task, available cues and comparison method; heuristics have no universal accuracy advantage.
09Connections
- Often confused with Cognitive Bias
- Can lead toSpeed-Accuracy Trade-Off
- Can follow from Bounded Rationality
- Part of Dual-Process Theory, Ecological Rationality
- Includes Representativeness Heuristic, Availability Heuristic, Fast-and-Frugal Tree, Satisficing, Peak-End Rule
- See also Value of Information
+ 1 more in the list
10Origin and sources
An older problem-solving tradition developed in George Pólya’s How to Solve It (1945), Herbert Simon’s work on bounded rationality in the 1950s, and later judgment research by Amos Tversky and Daniel Kahneman.
- [1]Pólya, G. (1945). How to Solve It: A New Aspect of Mathematical Method. Princeton University Press.
- [2]Simon, H. A. (1955). A Behavioral Model of Rational Choice. The Quarterly Journal of Economics, 69(1), 99–118.
- [3]Tversky, A., & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases. Science, 185(4157), 1124–1131.
- [4]Gigerenzer, G., & Gaissmaier, W. (2011). Heuristic Decision Making. Annual Review of Psychology, 62, 451–482.
Suggest an edit· Updated 2026-10-02