Liquidity providers deposit funds into Biswap and receive trading fees from traders (supply-side fees).

Biswap is a non-custodial exchange protocol built on BNB chain. Biswap was founded by an anonymous team and launched in 2021.

BSW tokenholders own and govern the protocol. They currently take a cut of the total trading fees paid by traders (revenue). Biswap’s investors include Binance Labs.

github

twitter

etherscan

website

Exchanges (DEX)

Biswap

# Biswap Active users (weekly) Methodology

## 1. Overview

The Biswap weekly active users (WAU) metric measures the number of unique users who performed at least one trade on the Biswap decentralized exchange within any 7-day rolling window. This metric provides insight into the protocol's trading activity and user engagement across both Biswap v1 and v3 versions on the BSC blockchain. The calculation uses HyperLogLog probabilistic data structures for efficient cardinality estimation across large user sets.

## 2. Data sources

The calculation relies on several types of blockchain data from the BSC network:

- **Swap event logs**: Parsed `Swap` events from both Biswap v1 pair contracts and Biswap v3 pool contracts, containing token amounts, pool addresses, and transaction details
- **Function call traces**: BSC transaction traces for Biswap v3 `swapY2X` and `swapX2Y` function calls, used to determine swap directions and extract recipient addresses
- **Transaction records**: BSC blockchain transaction data containing user addresses (`from_address`) who initiated the swap transactions
- **Event metadata**: Block timestamps, transaction hashes, and other blockchain metadata for temporal organization

## 3. Step-by-step calculation

### Component A: Trade event processing
- **Purpose**: Extracts and standardizes all swap events from both Biswap protocol versions
- **Logic**: The system processes swap events from Biswap v1 pair contracts and Biswap v3 pool contracts separately. For v1 swaps, it extracts `amount0In`, `amount0Out`, `amount1In`, and `amount1Out` fields from `Swap` events and calculates net token changes by taking differences between input and output amounts. For v3 swaps, it extracts `amountX` and `amountY` from the `returnValues` field of `Swap` events. All events are joined with BSC transaction records to obtain the actual user address (`from_address`) who initiated the transaction. Events are filtered to include only successful transactions from the protocol launch date of May 23, 2021.
- **Key fields**: `block_timestamp`, `transaction_hash`, `user_address`, `token0_amount`, `token1_amount`

### Component B: Swap direction determination  
- **Purpose**: Determines the direction of v3 swaps to correctly apply amount calculations
- **Logic**: For Biswap v3, the system examines function call traces to identify whether each swap used the `swapY2X` or `swapX2Y` function. It matches swap events to function calls using `transaction_hash` and `trace_address`. For `swapY2X` functions, it applies direction multipliers of -1 for X direction and 1 for Y direction. For `swapX2Y` functions, it applies direction multipliers of 1 for X direction and -1 for Y direction. These multipliers are then applied to the `amountX` and `amountY` values to calculate the correct token amounts for each swap.
- **Key fields**: `directionX`, `directionY`, `user_address`, `recipient`

### Component C: Daily active user aggregation
- **Purpose**: Identifies unique users who traded on each day
- **Logic**: The system groups all processed trade events by day using timestamp truncation to day granularity. For each day, it extracts the unique set of user addresses who performed at least one trade, classifying all users as 'Trader' type. The aggregation includes chain information (BSC), business classification ('Exchange AMM'), and protocol version details. This creates a daily snapshot of all unique trading users.
- **Key fields**: `timestamp`, `timestamp_grain`, `chain`, `user_type`, `user_dau`

### Component D: HyperLogLog sketch creation
- **Purpose**: Converts daily user sets into probabilistic data structures for efficient processing
- **Logic**: For each day's unique user set, the system initializes HyperLogLog sketches with 15-bit precision using the `hll_count.init` function. This converts the exact user sets into probabilistic sketches that can be efficiently merged and processed while maintaining cardinality accuracy. The sketches are grouped by timestamp, chain, user type, and product dimensions.
- **Key fields**: `hll` (HyperLogLog sketch), `timestamp`, `chain_id`, `user_type`, `product_id`

### Component E: Rolling window construction
- **Purpose**: Creates 7-day rolling windows for WAU calculation
- **Logic**: The system generates a complete datespine covering all days from the earliest trade date to yesterday. For each day in the datespine, it creates a 7-day window by generating timestamps for the current day and the previous 6 days using timestamp arithmetic. This creates join relationships between each day and its corresponding 7-day window. The windows are then joined with the daily HyperLogLog sketches to create overlapping 7-day user activity windows.
- **Key fields**: `timestamp` (window end date), `joinday` (individual days in window)

### Component F: WAU aggregation by dimension
- **Purpose**: Calculates weekly active users across different dimensional breakdowns
- **Logic**: For each 7-day window, the system merges HyperLogLog sketches from all days in the window using `hll_count.merge_partial`, then extracts the final cardinality estimate using `hll_count.extract`. This provides the estimated count of unique users active within each 7-day period. The calculation is performed separately for four dimensional breakdowns: project-level (aggregating all activity), chain-level (by blockchain), user type-level (by user classification), and product-level (by specific products). Each breakdown groups the data differently while maintaining the same core WAU calculation logic.
- **Key fields**: `user_wau` (weekly active user count), `dimensions` (breakdown type), `chain_id`, `user_type`, `product_id`

### Final aggregation
- **Combining components**: The four dimensional breakdowns are combined using union operations to create a comprehensive dataset with multiple perspectives on weekly active users
- **Currency conversion**: Not applicable for this user count metric
- **Time aggregation**: Uses rolling 7-day windows with daily granularity, where each day represents the end of a 7-day period
- **Final calculation**: The final WAU value for each day represents the estimated number of unique users who performed at least one trade within the 7-day window ending on that date, calculated using HyperLogLog cardinality estimation for computational efficiency

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

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

BiswapBSW

Active users (weekly)

Active users (weekly)

Competitive landscape