Understanding Goodhart’s Law of Metrics

Goodhart’s Law: When a Measure Becomes a Target

British economist Charles Goodhart once observed that "When a measure becomes a target, it ceases to be a good measure." This idea, now known as Goodhart’s Law, explains why organizations, leaders, and even entire industries struggle when they optimize for numbers instead of real outcomes.

Metrics are essential for tracking progress, but when people focus too much on hitting specific numbers, they often game the system, distort incentives, and sacrifice long-term value for short-term gains.

This issue is particularly damaging in high-stakes industries where true success cannot always be neatly quantified.

Why This Happens: The Dark Side of Metrics-Driven Thinking

1. Optimizing for the Metric Instead of the Outcome

When numbers become the primary goal, people start chasing the metric rather than improving the underlying reality.

Examples:

• Employee Engagement Scores: If a company sets a target to improve engagement survey results, managers might pressure employees to rate them highly—rather than actually improving work culture.
• Sales Quotas: If a sales team is judged purely on the number of deals closed, they might push low-quality deals that fall apart later, just to meet short-term targets.
• Social Media Growth: If content teams are incentivized by likes and views, they may resort to clickbait and sensationalism rather than meaningful engagement.

By optimizing for the number rather than the real goal, organizations undermine the very thing they were trying to improve.

2. Distorted Behaviors and Incentives

When the wrong incentives are in place, people will do whatever it takes to hit their numbers—even if it leads to unethical, counterproductive, or meaningless outcomes.

Examples:

• Short-Term Stock Gains Over Long-Term Health: Public companies focused on quarterly earnings often cut research & development to boost short-term stock prices, weakening long-term innovation.
• Hospital Efficiency Metrics: Hospitals focused on reducing average length of patient stays might discharge people too soon, increasing the risk of complications.
• Customer Service Metrics: Call centers that measure average call time might rush through customer interactions, reducing actual service quality.

Metrics should inform strategy, not drive it. Otherwise, people start playing the system rather than improving actual performance.

When Metrics Are Useful

Metrics are not inherently bad, but they must be used as indicators rather than rigid targets. Instead of saying, “We need to hit X number,” a better approach is:

• What are we truly optimizing for?
• Are we improving the real outcome, or just manipulating the metric?
• Are we balancing multiple measures to prevent gaming?

A holistic approach ensures that numbers serve as a guide, not a goal in themselves.

The Problem with Over-Reliance on Metrics in Mental Health & Executive Performance

1. Vanity Metrics vs. True Impact

It’s tempting to use simple, quantifiable metrics to measure success. But these often fail to reflect real transformation. 

Examples:

• Session Completion Rates: A psychologist or coach might have high client retention, but if the client still makes poor leadership decisions under pressure, what was the real impact?
• Client Satisfaction Scores: Just because a client feels good about a session doesn’t mean they actually developed resilience, improved decision-making, or changed behavior.

Short-term improvements in a survey do not measure long-term success and transformation.

2. Reductionism: Oversimplifying Complex Human Factors

Business leadership, decision-making, and psychological resilience are multi-layered, complex constructs. Trying to distill them into a single metric creates a flawed, oversimplified view of progress.

Examples:

• Burnout Score Reductions: Measuring improvement by a 20% decrease in reported burnout ignores qualitative factors like enhanced judgment, mental agility, or emotional regulation, which aren’t easily captured in a number.
• Leadership Confidence Scores: If a coaching program is judged only by self-reported leadership confidence, participants may learn to give the “right” answers rather than developing true self-awareness and adaptability.

Numbers should not replace deeper qualitative assessments of progress.

3. Gaming the System: When Metrics Become the Goal

When mental health or leadership performance is measured too rigidly, professionals start optimizing for the metric rather than real progress.

Examples:

• Therapy & Insurance Coverage: Many insurance companies require therapists to show symptom reduction scores to justify continued coverage. This pushes therapists to prioritize short-term symptom relief over deeper, long-term change.
• Executive Coaching Programs: If success is measured by increased self-reported confidence, clients may adjust their answers rather than actually evolving as leaders.

For real impact, we must resist the urge to game the system and instead focus on actual transformation—even if it’s harder to quantify.

How to Avoid the Goodhart’s Law Trap

1. Use Metrics as a Guide, Not the Goal

• Instead of “We need a 90% engagement score,” ask, “What workplace conditions drive genuine engagement?”
• Instead of “Close X number of deals,” ask, “Are we building sustainable customer relationships?”

2. Balance Quantitative and Qualitative Measures

• Combine hard metrics with subjective, long-term indicators of success.
• Conduct deep-dive assessments instead of relying on surface-level numbers.

3. Optimize for the Real Outcome, Not the Shortcut

• Focus on behavioral and organizational change, not just reported scores.
• Recognize when a metric is being gamed rather than driving real progress.

Final Thought: The Right Way to Use Metrics

Goodhart’s Law reminds us that numbers alone cannot define success. Whether in business, leadership, or mental health, the most effective strategies balance measurable progress with deeper qualitative insights.

Metrics should serve as a compass, not a destination. The key is to optimize for the true outcome, not just the number.

In the end, real success isn’t what the data says—it’s what actually changes.