website

twitter

github

Lending

Prime Numbers

# Prime Numbers Average revenue per user (ARPU) Methodology

## 1. Overview

Average Revenue Per User (ARPU) measures the average amount of revenue generated per active user on PrimeFi Lending per day. This metric is calculated by dividing daily protocol revenue by the count of daily active users. ARPU helps assess user monetization efficiency and revenue generation per user engagement on the lending protocol.

## 2. Data sources

- **XDC blockchain token transfer events**: Token transfer logs showing movement of aTokens to the protocol treasury address
- **PrimeFi Lending lending protocol events**: Parsed events from PrimeFi Lending v1 contracts on XDC including `Borrow`, `FlashLoan`, `Repay`, `Deposit`, and `Withdraw` events
- **Protocol configuration events**: `ReserveFactorChanged` and `ReserveInitialized` events from PrimeFi Lending's `LendingPoolConfigurator` contract
- **End-of-day token price feeds**: Daily USD price data for revenue conversion
- **Token metadata**: Token decimal information for proper amount normalization
- **Market dimension data**: Mapping between underlying tokens and their corresponding aToken, stable debt token, and variable debt token addresses

## 3. Step-by-step calculation

### Component A: Revenue calculation
- **Purpose**: Calculate daily protocol revenue from lending interest fees
- **Logic**: Revenue follows the AAVE v2 pattern where protocol fees are collected through aToken mints to the treasury. The process identifies token transfers where the `from` address is the null address (`0x0000000000000000000000000000000000000000`) and the `to` address is the PrimeFi Lending treasury (`0x9d96d06cb863b76b07670e98d86470407a965a9c`) on XDC. These transfers represent aToken mints that capture the protocol's share of interest fees. The transfer amounts are joined with market dimension data to identify valid aToken addresses. Reserve factor logic is applied where when a reserve factor exists and is non-zero, the full transfer amount represents protocol revenue. For tokens that don't have reserve factors or have zero reserve factors, revenue is set to zero. A special mapping converts psXDC token addresses to WXDC addresses for price lookup purposes.
- **Key fields**: `block_timestamp` for daily aggregation, `value` for transfer amounts, `token_address` for aToken identification, `to_address` for treasury verification, `reserve_factor` for fee calculation

### Component B: Daily active users calculation  
- **Purpose**: Count unique users who interact with the lending protocol each day
- **Logic**: Users are identified from five types of lending protocol events on XDC: `Borrow` events (using `user` parameter), `FlashLoan` events (using `initiator` parameter), `Repay` events (using `user` parameter), `Deposit` events (using `user` parameter), and `Withdraw` events (using `user` parameter). Each event creates a user record with classification by user type (Borrower, Repayer, Depositor, Withdrawer). The system extracts unique user addresses per day and uses HyperLogLog aggregation for efficient distinct counting across multiple dimensions including project level, chain level, user type, and product ID. This approach handles large-scale distinct counting while supporting various analytical breakdowns.
- **Key fields**: `block_timestamp` for daily bucketing, `user_address` for user identification, `user_type` for user classification, `chain` and `product_id` for dimensional analysis

### Final aggregation
- **Combining components**: Revenue and user metrics are calculated independently and then combined through division using the `calculate_arpu` macro, which matches records by date and project identifier
- **Currency conversion**: Revenue amounts are converted to USD using end-of-day price feeds, with special handling for psXDC tokens which are mapped to WXDC prices. Token decimal normalization is applied to ensure correct USD amounts
- **Time aggregation**: Both revenue and user metrics are aggregated to daily boundaries using `timestamp_trunc` to day level, ensuring consistent time alignment
- **Final calculation**: ARPU equals daily total revenue in USD divided by daily active user count, with zero-filling applied for days with missing data to maintain continuous time series

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

Prime NumbersPRFI

Average revenue per user (ARPU)

Average revenue per user (ARPU)

Competitive landscape