What is it?

Overfitting occurs when a model, insight, or conclusion is too closely aligned to a specific dataset and does not generalise to broader behaviour.

Overfitting reduces reliability and repeatability.

Examples in Action

• Optimising based on very small samples
  
• Over-segmenting audiences
  
• Drawing conclusions from isolated tests
  
• Excessive tuning without validation
  

Typical Outcomes / Results

• Reduced confidence in results
  
• Poor performance when scaled
  
• Misleading insights
  
• Increased risk of false positives
  

This definition reflects established analytical principles.