website

twitter

Blockchains (L2)

Humanity Protocol

# Humanity Protocol Active addresses (weekly) Methodology

## 1. Overview

Weekly active addresses measures the number of unique addresses that initiated at least one transaction on the Humanity blockchain within a rolling 7-day period. This metric treats transaction senders as "active addresses" and uses a rolling window approach to capture weekly activity patterns. The calculation employs HyperLogLog algorithms for efficient distinct counting across large datasets while maintaining accuracy.

## 2. Data sources

The calculation relies on these primary data sources:
- **Humanity blockchain transaction records**: Raw transaction data from the Humanity blockchain containing transaction metadata, sender addresses, and timing information
- **Transaction sender addresses**: The `from_address` field identifying the wallet or contract that initiated each transaction  
- **Block timestamps**: Precise timing data for when each transaction was included in a block, used for daily and weekly aggregation boundaries

## 3. Step-by-step calculation

### Component A: User identification and daily aggregation
- **Purpose**: Identify unique active users per day from raw transaction data
- **Logic**: Extract transaction sender addresses from Humanity blockchain transaction records. Each transaction record includes the sender address (`from_address`), transaction hash, and block timestamp. The component classifies all transaction senders as "Transaction sender" user type and assigns them to the 'blockchains_l2' market sector. Daily boundaries are created by truncating block timestamps to day granularity, then counting distinct sender addresses within each day to produce daily active user counts.
- **Key fields**: `from_address` as the user identifier, `block_timestamp` for temporal grouping, and derived fields for chain classification and business segmentation

### Component B: HyperLogLog sketch creation  
- **Purpose**: Convert daily user counts into efficient data structures for weekly aggregation
- **Logic**: Transform daily distinct user address counts into HyperLogLog sketches using a precision parameter of 15. Group the daily data by timestamp, chain, user type, and product dimensions. Each day's unique user addresses are initialized into HLL sketches that can later be merged efficiently to calculate weekly metrics without double-counting users who were active on multiple days within the week.
- **Key fields**: `timestamp` (day-truncated), `chain_id`, `user_type`, `product_id`, and the resulting `hll` sketch containing the distinct user addresses

### Component C: Rolling window generation
- **Purpose**: Create a complete 7-day rolling window framework for weekly calculations
- **Logic**: Generate a datespine covering all days from the earliest data through yesterday. For each day in the datespine, create associations with the previous 6 days to form complete 7-day windows. This uses timestamp arrays to generate all date combinations, ensuring that weekly calculations don't skip days with missing data. The window structure enables consistent weekly metrics even when some days have no transaction activity.
- **Key fields**: `timestamp` for the target date and `joinday` representing each day within the 7-day lookback period

### Component D: Weekly aggregation with HLL merging
- **Purpose**: Calculate weekly active users by merging daily HLL sketches within 7-day windows  
- **Logic**: Join the rolling window framework with the daily HLL sketches to combine user data across 7-day periods. Within each window, merge the HLL sketches using HLL merge functions to calculate the distinct count of users who were active on any day within that week. This approach efficiently handles users who were active on multiple days without double-counting them. Extract the final count from the merged HLL sketch to produce the weekly active user metric.
- **Key fields**: `timestamp` for the week ending date, dimensional grouping fields, and `user_wau` as the final weekly distinct user count

### Component E: Dimensional breakdowns
- **Purpose**: Provide weekly active user metrics across different analytical dimensions
- **Logic**: Calculate weekly active users across four different dimensional views: project-level aggregation (project_dbt_tag), chain-level aggregation, user type segmentation, and product-level breakdowns. Each breakdown maintains the same HLL merging logic but groups data differently. The project_dbt_tag dimension aggregates all data without chain/product filters, while chain-level preserves chain_id grouping, user_type maintains user classification, and product_id segments by specific product categories.
- **Key fields**: `dimensions` field indicating the breakdown type, relevant dimensional identifiers, and `user_wau` counts for each segment

### Final aggregation
- **Combining components**: All dimensional breakdowns are combined using union operations to create a comprehensive dataset with different analytical views of weekly active users
- **Currency conversion**: No currency conversion is required as this metric counts addresses/users rather than monetary values
- **Time aggregation**: Weekly boundaries are defined as 7-day rolling periods, with each day representing the end of a week that includes the previous 6 days
- **Final calculation**: The metric renames `user_wau` to `active_addresses_weekly` to align with blockchain terminology where active users are conceptually treated as active addresses. The calculation applies chain availability filters and uses standardization macros to ensure consistent output formatting across all Token Terminal metrics

*This methodology was automatically generated by Token Terminal's methodology analyzer.*

Unique active addresses over a rolling 7-day window.

Humanity ProtocolH

Active addresses (weekly)

Active addresses (weekly)

Competitive landscape