website

twitter

Blockchains (L1)

Plasma

# Plasma Active addresses (weekly) Methodology

## 1. Overview

Weekly active addresses measures the number of distinct addresses that sent at least one transaction on the Plasma blockchain during a 7-day rolling window period. This metric answers the business question: "How many unique users are actively transacting on Plasma each week?" The calculation uses a rolling 7-day window approach with HyperLogLog data structures for efficient distinct counting, and treats each transaction sender address as an active user address.

## 2. Data sources

The calculation relies on the following raw data sources:
- **Plasma transaction records**: Raw blockchain transaction data containing transaction hashes, block timestamps, and sender addresses (`from_address`)
- **Chain availability configuration**: System configuration data that determines which time periods and chains are available for analysis

## 3. Step-by-step calculation

### Component A: Transaction sender extraction
- **Purpose**: Extract and standardize all transaction sender addresses as the foundation for user activity analysis
- **Logic**: Process raw Plasma blockchain transaction records to identify each transaction sender address along with the transaction timestamp. Cast and standardize data types for consistency. Add metadata classifications including chain identifier ('plasma'), market sector ('blockchains_l1'), business segment ('blockchain'), and user type ('Transaction sender'). Each transaction sender represents a distinct user address that will be counted in activity metrics.
- **Key fields**: `block_timestamp` for timing analysis, `from_address` as the user identifier, `transaction_hash` for transaction tracking

### Component B: Daily active address calculation  
- **Purpose**: Calculate the count of distinct addresses that were active (sent transactions) each day
- **Logic**: Group transaction records by day using timestamp truncation to day granularity. For each day, count distinct sender addresses to determine how many unique addresses were active. This creates daily snapshots of address activity that serve as the foundation for weekly aggregation. Apply incremental processing to efficiently update only recent data changes.
- **Key fields**: `timestamp` truncated to day level, `user_address` (renamed from `from_address`) as the distinct address identifier, `user_dau` as the daily distinct count

### Component C: HyperLogLog sketch creation
- **Purpose**: Create efficient data structures for distinct counting across time windows
- **Logic**: Convert daily distinct address lists into HyperLogLog sketches using `hll_count.init` function. These probabilistic data structures allow for efficient set operations when calculating rolling windows. Group by timestamp, chain, user type, and product dimensions to create separate sketches for different analysis breakdowns.
- **Key fields**: `hll` containing the HyperLogLog sketch for efficient set operations, dimensional grouping fields for analysis breakdowns

### Component D: Rolling window construction
- **Purpose**: Build 7-day rolling windows for weekly active address calculation
- **Logic**: Create a complete datespine covering all days in the analysis period to ensure no gaps in the time series. For each day in the datespine, generate a corresponding 7-day window that includes that day plus the previous 6 days. Use `generate_timestamp_array` to create these date ranges efficiently. Join the datespine windows with daily HLL sketches to associate each window with the relevant address activity data.
- **Key fields**: `timestamp` as the window end date, `joinday` for the days within each 7-day window, matched HLL sketches from corresponding days

### Component E: Weekly distinct counting
- **Purpose**: Calculate distinct addresses across each 7-day rolling window
- **Logic**: For each 7-day window, merge the HyperLogLog sketches from all days in that window using `hll_count.merge_partial`. Extract the final distinct count using `hll_count.extract` to get the number of unique addresses that were active at any point during the 7-day period. This accounts for addresses that may have been active on multiple days within the week, counting them only once.
- **Key fields**: `user_wau` containing the weekly distinct address count extracted from merged HLL sketches

### Component F: Dimensional breakdown generation
- **Purpose**: Create weekly active address metrics across different analysis dimensions
- **Logic**: Generate separate breakdowns for multiple analysis levels including project-level ('project_dbt_tag'), chain-level ('chain'), user-type-level ('user_type'), and product-level ('product_id'). Each breakdown aggregates the HLL sketches differently - project-level combines all chains and products, chain-level maintains chain separation, etc. Combine all dimensional breakdowns using UNION ALL to create a comprehensive dataset with multiple analysis perspectives.
- **Key fields**: `dimensions` indicating the breakdown level, `chain_id` for chain-specific analysis, `product_id` for product-specific analysis, `user_type` for user classification analysis

### Final aggregation
- **Combining components**: Select weekly active user counts from the dimensional breakdown results, filtering to only include 'chain' and 'project_dbt_tag' dimensions for the active addresses metric
- **Currency conversion**: Not applicable for address counting metrics
- **Time aggregation**: Uses 7-day rolling windows where each day represents the distinct address count for that day plus the previous 6 days
- **Final calculation**: Rename the `user_wau` field to `active_addresses_weekly` to reflect blockchain-specific terminology where active users are equivalent to active addresses. Apply the finalization macro to standardize the metric schema and fill missing time periods with zero values. Filter results using chain availability rules to exclude periods or chains that are not available for analysis.

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

Unique active addresses over a rolling 7-day window.

PlasmaXPL

Active addresses (weekly)

Active addresses (weekly)

Competitive landscape