In This Article
The British colonial government in India wanted fewer venomous cobras slithering through Delhi’s streets, so they offered a bounty for every dead snake brought in. Clever locals started breeding cobras just to kill them for the reward money. When officials discovered the scam and canceled the program, the breeders released their now-worthless snakes into the wild — making the cobra problem worse than before.
This disaster perfectly illustrates Goodhart’s law: when a measure becomes a target, it ceases to be a good measure. Named after British economist Charles Goodhart, this principle reveals why your best-intentioned metrics often backfire spectacularly.
Why Good Measures Go Bad
Think of metrics like a thermometer. A thermometer works great for measuring room temperature because it has no incentive to lie. But imagine if your salary depended on showing a specific temperature — you might hold a lighter under it or stick it in the freezer.
That’s exactly what happens with organizational metrics. The moment you tie rewards, punishments, or reputation to a number, people start gaming the system. They optimize for the metric instead of the underlying goal the metric was supposed to represent.
Goodhart’s Law Explained Examples in Action
Schools provide the clearest example of this phenomenon. When standardized test scores became the primary measure of educational success, teachers began “teaching to the test.” Students learned to fill in bubbles correctly but lost critical thinking skills. Art, music, and creative writing disappeared from curricula because they weren’t measured.
The underlying goal — quality education — got sacrificed for higher test scores.
Healthcare’s Dangerous Games
Hospitals face similar pressures with wait-time metrics. When emergency departments get penalized for long wait times, some staff learned to game the system: they’d quickly move patients from the waiting room to hallway gurneys, technically “seeing” them faster while providing no actual care improvement.
Even more troubling: some hospitals avoided treating the sickest patients because they took longer and hurt the metrics. The measure meant to improve patient care actually made it worse for those who needed help most.
Social Media’s Engagement Trap
Facebook, Twitter, and YouTube optimize for “engagement” — likes, shares, comments, and time spent on platform. This seems logical: more engagement means happier users, right?
Wrong. The algorithms discovered that outrage, controversy, and divisive content generate the most engagement. Your angry uncle’s political rant gets more comments than your thoughtful article about climate solutions. The platforms accidentally incentivized toxicity while trying to measure satisfaction.
The Business World’s Metric Obsession
Corporate America runs on Key Performance Indicators (KPIs), and Goodhart’s law explains why so many companies struggle with unintended consequences.
Sales teams measured purely on revenue might offer unsustainable discounts or sign unprofitable contracts. Customer service reps evaluated on call duration might rush customers off the phone without solving their problems. Software developers tracked on lines of code written might create unnecessarily complex, buggy programs.
Wells Fargo’s fake account scandal represents Goodhart’s law at its most destructive. Employees were measured and rewarded based on new account openings, so they opened millions of unauthorized accounts for existing customers. The metric designed to measure growth instead measured fraud.
The Amazon Warehouse Warning
Amazon measures warehouse workers’ productivity down to the second — packages picked per hour, time between tasks, even bathroom breaks. While this drives efficiency, workers report feeling like robots, leading to high turnover and union organizing efforts.
The company optimized for short-term productivity metrics but may have created long-term problems with worker satisfaction and retention.
Government Policy and Unintended Consequences
Politicians love measurable targets because they sound decisive and scientific. But Goodhart’s law explains why so many well-intentioned policies backfire.
The Soviet Union famously measured nail factory success by weight produced, so factories made fewer, heavier nails instead of the variety actually needed. When they switched to measuring quantity, factories made tiny, useless nails.
Modern examples abound: police departments measured on arrest numbers might target minor offenses instead of serious crimes. Welfare programs with strict income cutoffs create “benefit cliffs” where earning slightly more money results in losing thousands in assistance, trapping families in poverty.
Personal Goals Gone Wrong
Goodhart’s law doesn’t just affect organizations — it can sabotage your personal goals too.
Fitness trackers measuring steps might encourage you to pace around your house instead of doing strength training or yoga. Weight loss goals might lead to crash dieting instead of sustainable lifestyle changes. Productivity apps counting tasks completed might push you toward busy work instead of meaningful projects.
The key insight: measuring something changes behavior around that thing, often in ways that defeat the original purpose.
Defending Against Goodhart’s Law
Smart organizations use multiple metrics that balance each other out. Instead of just measuring customer service call duration, track both speed AND customer satisfaction scores. Instead of only counting sales revenue, monitor customer retention and profitability too.
The best approach often involves periodic metric rotation — changing what you measure before people can fully game the system. Like a vaccine, exposure to different measures builds organizational immunity to any single metric’s weaknesses.
Some companies deliberately keep certain metrics private or use them only for diagnosis, not rewards. This preserves their usefulness as genuine feedback tools.
The Measurement Paradox
Goodhart’s law creates a fundamental paradox: the act of measurement changes what’s being measured. It’s like quantum physics for organizations — observation affects outcomes.
This doesn’t mean we should abandon metrics entirely. Numbers remain essential for understanding reality and making decisions. But Goodhart’s law reminds us to stay humble about what our measures actually capture and vigilant about their unintended effects.
The next time someone proposes a new KPI, bonus structure, or government program, ask yourself: “What would happen if people started optimizing specifically for this metric?” The answer might save you from breeding your own organizational cobras.
Frequently Asked Questions
What exactly is Goodhart’s law in simple terms?
Goodhart’s law states that when a measure becomes a target, it ceases to be a good measure. Essentially, once people know they’re being evaluated on a specific metric, they’ll find ways to improve that metric that don’t necessarily improve the underlying thing it was supposed to measure.
Can you give a real-world example of Goodhart’s law explained?
A classic example is when schools focus too heavily on standardized test scores. Teachers start “teaching to the test” rather than providing well-rounded education, students become good at test-taking but lose critical thinking skills, and the original goal of quality education suffers despite higher scores.
How can businesses avoid falling victim to Goodhart’s law?
Use multiple, balanced metrics rather than focusing on just one. Combine quantitative measures with qualitative feedback, rotate metrics periodically, and keep some measurements private for diagnostic purposes only. Most importantly, regularly ask whether your metrics are driving the behaviors you actually want.
Is Goodhart’s law the same as the cobra effect?
The cobra effect is actually a specific example of Goodhart’s law in action. While Goodhart’s law describes the general principle that measures become less useful when they become targets, the cobra effect specifically refers to solutions that make problems worse — like the British bounty program that increased Delhi’s snake population.
Does Goodhart’s law mean we shouldn’t use metrics at all?
No, metrics are still essential for understanding performance and making decisions. Goodhart’s law simply warns us to be thoughtful about how we design and use metrics. The key is using them as tools for insight rather than rigid targets, and being aware of how measurement might change behavior in unexpected ways.