Entity-Adjusted Metrics

Raw on-chain metrics count every transaction equally, including internal transfers (an exchange moving coins between hot and cold wallets), change outputs (UTXO mechanics creating new addresses for the sender), and consolidation transactions. These inflate activity metrics and can produce misleading signals.

Entity adjustment, pioneered by Glassnode, uses clustering algorithms and heuristic analysis to group addresses that likely belong to the same entity. Common heuristics include: addresses that appear as inputs in the same transaction (common input ownership), known exchange address tagging, and behavioral pattern analysis. After clustering, internal transfers are filtered out.

According to Glassnode's methodology documentation (2023), entity adjustment reduces apparent Bitcoin transaction counts by approximately 30–40% and on-chain volume by 40–60% compared to raw figures, producing a significantly more accurate representation of genuine economic activity. Without entity adjustment, metrics like active addresses and transaction volume can be substantially inflated by exchange operations and UTXO management.

Key entity-adjusted metrics include: entity-adjusted transaction count, entity-adjusted transfer volume, entity-adjusted active entities (more accurate than active addresses), and entity-adjusted SOPR. When evaluating on-chain data for investment decisions, entity-adjusted versions should be preferred over raw metrics whenever available, as they filter out noise and more accurately reflect real user behavior and economic activity on the network.