Validators process transactions and receive fees from users (supply-side fees).

Flow is a smart contract platform for decentralized applications (a blockchain). Flow was founded by Roham Gharegozlou, Dieter Shirley and Mikhael Naayem and launched in 2019.

FLOW holders own and govern the protocol. They currently don’t take a cut of the total transaction fees paid by users (revenue). Flow’s investors include a16z, Coatue, Coinfund, Fabric Ventures, etc.

governance

github

twitter

discord

website

Blockchains (L1)

Flow

# Flow Active users (weekly) Methodology

## 1. Overview

Flow Weekly Active Users (WAU) measures the number of unique addresses that performed transactions on the Flow blockchain within any 7-day rolling period. This metric captures user activity across Flow's dual-chain architecture, including both Flow EVM (Ethereum Virtual Machine compatibility layer) and Flow (native Flow runtime). The calculation uses HyperLogLog data structures to efficiently count unique users across rolling windows and provides dimensional breakdowns by chain, user type, and product.

## 2. Data sources

The calculation relies on several types of Flow blockchain transaction data:

- **Flow EVM transaction records**: Standard EVM-compatible transactions containing sender addresses and transaction metadata
- **Flow Cadence transaction records**: Native Flow runtime transactions with multiple user roles including payers, proposers, and proposal key addresses
- **Flow collections data**: Transaction authorization records containing arrays of authorizer addresses for each transaction
- **Transaction metadata**: Block timestamps, transaction hashes, and fee information from both Flow EVM and Flow Cadence networks

## 3. Step-by-step calculation

### Component A: Flow EVM user identification
- **Purpose**: Identifies active users on Flow's EVM compatibility layer
- **Logic**: Extracts unique sender addresses from Flow EVM transaction records. Each transaction's `from_address` field represents a user who initiated a transaction. Users are grouped by day and deduplicated to create daily active user lists. All Flow EVM users are classified with `chain` as "Flow EVM", `user_type` as "user", and `product_id` as "Flow EVM"
- **Key fields**: `block_timestamp` for daily grouping, `from_address` as user identifier, `transaction_hash` for transaction linking

### Component B: Flow user identification  
- **Purpose**: Identifies active users on Flow's native Cadence runtime across multiple user roles
- **Logic**: Extracts users from three distinct roles in Flow Cadence transactions. Payers are identified from the `payer` field in transaction records. Proposers are extracted from the `proposal_key_address` field. Authorizers are unnested from the `authorizers` array in the collections table, where each transaction can have multiple authorizing addresses. All three user types are combined and deduplicated daily. Flow users are classified with `chain` as "flow", `user_type` as "user", and `product_id` as "Flow"
- **Key fields**: `payer`, `proposal_key_address`, and `authorizers` array as user identifiers, `block_timestamp` for daily grouping

### Component C: Daily user aggregation
- **Purpose**: Combines Flow EVM and Flow users into unified daily active user counts
- **Logic**: Unions Flow EVM and Flow user data, then groups by day, chain, user type, and product to create daily unique user lists. Each day's data is converted into HyperLogLog data structures using `hll_count.init()` with precision 15 to enable efficient set operations for weekly calculations
- **Key fields**: `timestamp` truncated to day, `user_address` as unique identifier, `chain` and `product_id` for dimensional breakdowns

### Component D: Rolling window creation
- **Purpose**: Creates 7-day rolling windows for weekly active user calculation
- **Logic**: Generates a complete datespine covering all days from the earliest transaction to yesterday. For each day in the datespine, creates a 7-day lookback window using `generate_timestamp_array()` from 6 days prior through the current day. This ensures every day has a complete 7-day window for WAU calculation, even when some days have no transaction activity
- **Key fields**: `timestamp` as the target date, `joinday` representing each day in the 7-day lookback window

### Component E: Window aggregation
- **Purpose**: Collects daily HLL structures within each 7-day window
- **Logic**: Joins the rolling windows with daily HLL data to gather all unique user HLL structures that fall within each 7-day period. This creates a dataset where each target date has associated HLL structures from its 7-day lookback window, enabling weekly unique user counting across the rolling period
- **Key fields**: `timestamp` as target date, `hll` structures from contributing days within the window

### Component F: Dimensional breakdowns
- **Purpose**: Creates separate WAU calculations for different analytical dimensions
- **Logic**: Calculates weekly active users across four dimensional breakdowns. Project-level aggregation merges all HLL structures regardless of chain or product. Chain-level aggregation groups by `chain_id` to separate Flow EVM and Flow Cadence activity. User-type aggregation groups by `user_type` classification. Product-level aggregation groups by `product_id` to distinguish different Flow products. Each breakdown uses `hll_count.merge_partial()` to combine HLL structures and `hll_count.extract()` to get final unique user counts
- **Key fields**: `dimensions` indicating breakdown type, `chain_id`, `user_type`, `product_id` as grouping dimensions

### Final aggregation
- **Combining components**: All dimensional breakdowns are combined using UNION ALL to create a unified dataset with different analytical views of weekly active users
- **Currency conversion**: No currency conversion required as this is a user count metric
- **Time aggregation**: Daily rolling 7-day windows where each day represents the count of unique users active in that day plus the previous 6 days
- **Final calculation**: For each dimension and time period, the final WAU value is extracted from merged HyperLogLog structures representing the exact count of unique addresses that performed transactions within the 7-day rolling window

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

Unique revenue-generating users over a rolling 7-day window.

FlowFLOW

Active users (weekly)

Active users (weekly)

Competitive landscape

Chain breakdown