Organizations rely heavily on metrics to measure success. Businesses track revenue growth, schools evaluate test scores, marketers monitor website traffic, and managers assess employee performance through key performance indicators (KPIs).
However, there is a fundamental problem with metrics: once people know they are being measured by a specific number, they often start optimizing for that number rather than the actual goal.
This phenomenon is captured by Goodhart’s Law:
“When a measure becomes a target, it ceases to be a good measure.”
Coined by British economist Charles Goodhart in 1975, Goodhart’s Law explains why performance metrics often become unreliable once incentives are attached to them.
Today, the law is widely applied in business management, economics, education, healthcare, software development, and artificial intelligence.
This comprehensive guide will explore:
- The origin and meaning of Goodhart’s Law
- Why metrics become distorted
- Real-world examples across industries
- Implications for business and decision-making
- Strategies to design better measurement systems
Table of contents
The Origin of Goodhart’s Law
Charles Goodhart introduced the concept while studying monetary policy in the United Kingdom during the 1970s.
He observed that economic indicators used by governments and central banks became less useful once policymakers began targeting them directly.
Key points:
- Proposed by economist Charles Goodhart in 1975.
- Originally focused on monetary policy and economics.
- Later expanded to management, education, technology, and AI.
- Highlights the unintended consequences of performance measurement.
What began as an economic observation eventually became one of the most important principles in organizational management and decision-making.
What Goodhart’s Law Really Means
At its core, Goodhart’s Law reveals a simple truth:
A metric works well as a measurement only until people start optimizing specifically for it.
Once incentives are tied to a measurement, behavior changes.
For example:
- If teachers are judged solely by test scores, they may teach specifically for exams rather than deeper learning.
- If customer support teams are evaluated by call duration, they may rush conversations instead of solving customer problems.
- If software developers are measured by lines of code, they may write more code rather than better code.
In each case, the metric improves while the underlying objective may not.
Simply put:
People respond to incentives, and metrics can be manipulated.
Why Goodhart’s Law Happens
Several psychological and organizational factors contribute to the law.
1. Incentive Optimization
Human behavior is strongly influenced by incentives. When rewards such as promotions, bonuses, recognition, or performance ratings are directly tied to specific metrics, individuals naturally begin to focus their efforts on improving those measurable indicators.
Over time, this creates a shift in behavior: instead of prioritizing the original purpose behind the work, people concentrate on actions that improve the score being tracked. Even if those actions do not meaningfully contribute to overall quality or long-term goals, they become the most efficient path to success within the system.
As a result, the organization may see improvements in reported numbers, while the actual underlying performance or value creation may remain unchanged—or even decline.
2. Goal Substitution
Goal substitution occurs when the metric itself gradually replaces the original objective in importance.
At the beginning, the metric is only meant to be a tool for measuring progress toward a larger goal. However, once it becomes central to evaluation and decision-making, people start treating the metric as the goal itself.
Instead of focusing on the real-world outcome—such as better learning, improved customer satisfaction, or higher product quality—individuals begin focusing on improving the score, ranking, or numerical output.
This shift distorts decision-making because success is no longer defined by meaningful impact, but by performance on a narrow measurement.
3. Short-Term Thinking
Many performance metrics are designed to show quick, visible results. While this can be useful for tracking progress, it often encourages short-term optimization at the expense of long-term sustainability.
When individuals or teams are evaluated frequently, they tend to prioritize actions that improve immediate numbers, even if those actions are not beneficial in the long run. This can include cutting corners, delaying necessary investments, or focusing only on easy wins that boost performance in the short term.
Over time, this creates a pattern where surface-level improvements hide deeper structural issues. The system may appear to be performing well based on current metrics, but its long-term health, stability, or quality may gradually weaken.
4. Gaming the System
When metrics are closely tied to rewards, people often begin to identify ways to “game” the system—that is, improve the measured outcome without genuinely improving real performance.
This can involve exploiting loopholes, redefining categories, selectively reporting data, or focusing only on activities that directly influence the metric while ignoring other important but unmeasured responsibilities.
The stronger the incentives attached to a metric, the greater the motivation to find shortcuts or optimize appearances rather than substance. Over time, this behavior can significantly distort the reliability of the metric, making it less useful for decision-making and less reflective of actual performance.
Real-World Examples of Goodhart’s Law
1. Education
Schools frequently use standardized test scores to measure success.
As a result:
- Teachers may teach specifically for exams.
- Critical thinking receives less attention.
- Students focus on memorization rather than understanding.
Test scores improve, but learning quality may decline.
2. Sales Teams
Sales representatives are often rewarded based on revenue targets.
Potential consequences include:
- Aggressive selling techniques
- Overselling products
- Ignoring customer satisfaction
Revenue increases while customer trust decreases.
3. Healthcare
Hospitals commonly track performance indicators such as patient wait times.
To improve reported numbers, organizations may focus on administrative tactics rather than improving actual care quality.
The metric improves, but patient outcomes may remain unchanged.
4. Social Media Platforms
Platforms optimize for engagement metrics such as:
- Clicks
- Likes
- Shares
- Watch time
This can encourage sensational or emotionally charged content because it generates stronger engagement.
The metric succeeds even if user experience suffers.
Goodhart’s Law in Business and Management
1. Employee Performance Metrics
Employees naturally adapt to whatever management measures.
Poorly designed metrics can encourage:
- Quantity over quality
- Speed over accuracy
- Activity over results
2. KPI Management
Key Performance Indicators are valuable tools, but they can become misleading when treated as ultimate goals.
Examples include:
- Increasing website traffic instead of customer value.
- Maximizing production volume instead of product quality.
- Focusing on quarterly profits while neglecting long-term growth.
3. Organizational Culture
When metrics dominate decision-making, employees may optimize numbers instead of solving meaningful problems.
This can weaken innovation, collaboration, and ethical decision-making.
Goodhart’s Law in Software Development
Software engineering offers many examples of metric distortion.
1. Lines of Code
More code does not necessarily mean better software.
Developers may write unnecessary code to appear productive.
2. Bug Resolution Metrics
Teams may close tickets quickly rather than fully resolving underlying issues.
3. Velocity Tracking
Agile teams may inflate story points if velocity becomes a performance target.
4. Deployment Frequency
Teams may release minor changes simply to increase deployment counts.
The metric improves while customer value remains unchanged.
Goodhart’s Law vs. Campbell’s Law
Goodhart’s Law is often compared with Campbell’s Law.
Goodhart’s Law:
“When a measure becomes a target, it ceases to be a good measure.”
Campbell’s Law:
“The more a quantitative indicator is used for decision-making, the more likely it is to be distorted and corrupted.”
While similar, the two laws focus on different aspects of measurement.
- Goodhart’s Law focuses on optimization pressure.
- Campbell’s Law focuses on corruption and unintended consequences.
Together, they explain why poorly designed measurement systems frequently fail.
Strategies to Mitigate Goodhart’s Law
1. Use Multiple Metrics
Avoid relying on a single measurement.
Balanced scorecards reduce opportunities for gaming.
2. Measure Outcomes, Not Activities
Focus on actual results rather than proxy indicators whenever possible.
3. Review Metrics Regularly
Metrics that worked in the past may become ineffective over time.
Regular evaluation helps maintain relevance.
4. Include Qualitative Assessment
Not everything important can be measured numerically.
Customer feedback, peer reviews, and expert judgment provide valuable context.
5. Align Incentives Carefully
Ensure rewards support the broader objective rather than the metric itself.
6. Watch for Unintended Behavior
Monitor whether people are optimizing for the measurement rather than the desired outcome.
7. Encourage Long-Term Thinking
Sustainable success often requires balancing multiple goals rather than maximizing a single number.
Real-Life Application Tips
- Managers: Evaluate performance using both quantitative and qualitative measures.
- Business Leaders: Review KPIs regularly to ensure they still reflect organizational goals.
- Teachers: Focus on learning outcomes, not just test scores.
- Product Teams: Measure customer value alongside engagement metrics.
- Individuals: Avoid defining personal success through a single measurement.
Key Takeaways
- Goodhart’s Law states that when a measure becomes a target, it ceases to be a good measure.
- Metrics often become distorted when incentives are attached to them.
- The law applies across business, education, healthcare, technology, and AI.
- Overreliance on a single metric encourages gaming and unintended consequences.
- Balanced measurement systems reduce the risk of distortion.
- Effective decision-making requires looking beyond the numbers.
Frequently Asked Questions (FAQ)
1. What is Goodhart’s Law in simple terms?
Goodhart’s Law states that when a specific measurement is used as a target, people start optimizing for that number instead of the actual goal. As a result, the metric loses its effectiveness as a true indicator of performance.
2. Why does Goodhart’s Law happen?
It happens because people naturally respond to incentives. When rewards, promotions, or evaluations depend on a metric, individuals focus on improving that number—even if it means ignoring the real objective or finding shortcuts.
3. Can Goodhart’s Law be avoided completely?
It cannot be completely avoided, but it can be reduced. Using multiple metrics, focusing on real outcomes instead of proxy indicators, and combining quantitative data with qualitative evaluation can help reduce its impact.
4. Where is Goodhart’s Law commonly seen?
It is commonly seen in business KPIs, education systems (like standardized testing), healthcare performance tracking, social media engagement metrics, and even in artificial intelligence systems where reward optimization is used.
5. What is the difference between Goodhart’s Law and Campbell’s Law?
Goodhart’s Law explains how metrics become unreliable when used as targets. Campbell’s Law adds that the more a metric is used for decision-making, the more likely it is to be manipulated or corrupted.
Goodhart’s Law reminds us that metrics are tools, not goals.
Measurements help us understand performance, but they become unreliable when people begin optimizing specifically for them. Once a metric becomes the target, it often loses its ability to represent the underlying objective accurately.
By understanding Goodhart’s Law, individuals and organizations can design better incentive systems, make smarter decisions, and focus on meaningful outcomes rather than misleading numbers.
Whether in business, education, software development, healthcare, or everyday life, Goodhart’s Law serves as a powerful reminder that what gets measured is not always what truly matters.
