website

twitter

discord

Blockchains (L2)

Soneium

# Soneium Active addresses (weekly) Methodology

## 1. Overview

Soneium weekly active addresses measures the distinct number of unique addresses that initiated at least one transaction on the Soneium blockchain within any 7-day rolling window. This metric answers the business question "How many unique users are actively transacting on Soneium each week?" The calculation uses a sophisticated rolling window approach with HyperLogLog sketches to efficiently count distinct addresses across overlapping weekly periods while maintaining accuracy for large-scale blockchain data.

## 2. Data sources

- **Soneium transaction records**: Raw blockchain transaction data containing sender addresses, timestamps, and transaction hashes from the Soneium network
- **Transaction sender data**: Extracted from the `from_address` field of each transaction, representing the initiating wallet address
- **Blockchain timestamps**: Block-level timestamps used to establish daily and weekly time boundaries for activity measurement

## 3. Step-by-step calculation

### Component A: User identification and classification
- **Purpose**: Identifies and classifies active blockchain users from transaction data
- **Logic**: Extracts the sender address from each Soneium transaction using the `from_address` field. Each unique sender address represents a blockchain user, classified specifically as "Transaction sender" user type. The component captures the block timestamp, transaction hash, and categorizes the data with Soneium chain metadata including market sector as "blockchains_l2" and business segment as "blockchain"
- **Key fields**: `block_timestamp` for timing, `from_address` as user identifier, `transaction_hash` for transaction reference, `chain` set to "soneium"

### Component B: Daily active user aggregation  
- **Purpose**: Calculates daily distinct active users as the foundation for weekly calculations
- **Logic**: Groups transaction data by day using timestamp truncation to day granularity. Counts distinct user addresses per day, maintaining separate breakdowns by chain, user type, and product dimensions. Creates the base dataset of daily active users with consistent timestamp grain and dimensional attributes
- **Key fields**: `timestamp` truncated to day, `user_address` for distinct counting, `timestamp_grain` set to "day", `chain`, `user_type`, `product_id`

### Component C: HyperLogLog sketch creation
- **Purpose**: Creates memory-efficient data structures for distinct counting across time windows
- **Logic**: Converts daily user lists into HyperLogLog sketches using 15-bit precision for efficient storage and merging. Groups sketches by timestamp, chain, user type, and product dimensions. These sketches enable accurate distinct counting when merged across multiple days without storing full user address lists
- **Key fields**: `hll` sketch containing compressed user data, `timestamp`, `chain_id`, `user_type`, `product_id`

### Component D: Rolling window construction
- **Purpose**: Builds 7-day rolling windows for weekly active user calculation
- **Logic**: Creates a complete datespine covering all days from minimum timestamp to current date. For each day in the datespine, generates "joindays" representing the current day plus the preceding 6 days, creating overlapping 7-day windows. This approach ensures consistent weekly calculations even when specific days have no activity data
- **Key fields**: `timestamp` as the window end date, `joinday` representing each day within the 7-day window

### Component E: Weekly distinct user calculation
- **Purpose**: Calculates distinct users across 7-day rolling windows using HLL merging
- **Logic**: Joins the rolling window structure with daily HLL sketches, matching each joinday to its corresponding daily sketch. Merges HLL sketches across all 7 days in each window, then extracts the distinct count from the merged sketch. This process maintains accuracy while efficiently handling overlapping user activity across multiple days within each window
- **Key fields**: `timestamp` for the weekly period, `hll` merged sketches, extracted distinct counts

### Final aggregation
- **Combining components**: Creates multiple dimensional breakdowns of weekly active users including project-level (project_dbt_tag), chain-level, user-type-level, and product-level aggregations. Each breakdown groups and merges HLL sketches appropriately for the dimension, then unions all breakdowns into a single comprehensive dataset
- **Currency conversion**: No currency conversion required as this metric counts addresses rather than monetary values
- **Time aggregation**: Uses 7-day rolling windows where each day represents the end of a weekly period. Missing days are filled with zeros using the fill_zeros strategy to maintain consistent time series data
- **Final calculation**: Renames the `user_wau` field to `active_addresses_weekly` and applies chain availability filtering. The final metric represents the distinct count of unique addresses that initiated at least one transaction within any 7-day period ending on the specified timestamp

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

Unique active addresses over a rolling 7-day window.

Soneium

Active addresses (weekly)

Active addresses (weekly)

Competitive landscape

Chain breakdown