Skip to main content
At its core, onchain analytics is simply data analytics applied to blockchain data. Like any data-driven field, it builds upon the foundational principles of data analytics, also known as business analytics, which has evolved over decades across industries from retail to healthcare to finance. Data analytics traditionally encompasses four main types or levels of sophistication, each answering a different type of question:
  • **Descriptive **analytics asks “what happened?” and summarizes historical data, like tracking daily transaction volumes or identifying the largest token holders.
  • **Diagnostic **analytics asks “why did it happen?” and examines patterns to understand causes, such as exploring why a protocol’s total value locked suddenly dropped.
  • **Predictive **analytics asks “what might happen?” and uses statistical models to forecast outcomes, like estimating future network congestion based on historical patterns.
  • **Prescriptive **analytics asks “what should we do?” and recommends specific actions, such as suggesting optimal times to execute large trades or advising on protocol parameter adjustments.
These four categories provide a useful framework for understanding the different ways blockchain data can be analyzed and the varying levels of insight they provide. While much of onchain analytics today remains focused on descriptive and diagnostic work, the field is rapidly advancing toward more sophisticated predictive and prescriptive applications as tools mature and more historical data becomes available.