Definition. The net growth of unique entities in the network, defined as the difference between new entities and 'disappearing' entities (entities with a zero balance that had a non-zero balance at the previous timestamp).

Technical. Entities are clusters of addresses estimated to be controlled by the same actor, identified through advanced heuristics and Glassnode's proprietary clustering algorithms. Entity-based metrics rely on statistical and data-science methods that are refined over time. The series is therefore mutable: its established history is stable, but recent data points may revise as clustering improves. For methodology, see our article on account-based metrics. The computation requires statistical information from several days and is therefore only available with a lag of one week.

Notes. For more information, see our article on how many entities hold Bitcoin.

This is the Point-in-Time (PiT) variant of Entities Net Growth. PiT metrics are strictly append-only and their history is immutable. The historic data does not necessarily reflect the best current knowledge, but the information at the time when a data point was first computed. PiT metrics are ideal candidates for applications in model backtesting and related quantitative purposes. Read our article on PiT metrics for more information.