Lenders deposit funds into Spark and receive interest payments from borrowers (supply-side fees).

Spark is a non-custodial lending protocol built on Ethereum. Spark’s development team Phoenix Labs was founded in early 2023.

SPK tokenholders own and govern the protocol. They currently take a cut of the total interest paid by borrowers (revenue).

governance

github

twitter

discord

website

linkedin

Lending

Spark

SparkLend | Data breakdown

# Spark Total value locked Methodology

## 1. Overview

Total value locked (TVL) is defined as the total value of assets deposited within the Spark protocol ecosystem. This metric is utilized as a leading indicator for the protocol's growth and capacity to generate revenue through borrowing and savings activities. TVL is calculated as the sum of active loans and idle deposits in SparkLend, idle capital in Spark Savings, and assets under management in the Spark Liquidity Layer. To prevent double counting, assets deployed from Spark Savings into the Spark Liquidity Layer, as well as SparkLend deposits held by the Spark Liquidity Layer, are excluded from the final aggregation.

## 2. Data sources

- **Blockchain events**: ERC-20 token transfers, `Mint` and `Burn` events from `VariableDebtToken` contracts, `ReserveDataUpdated` events from `Pool` and `LendingPool` contracts, `Mint`, `Burn`, and `BalanceTransfer` events from `AToken` and `spToken` contracts, and on-chain state snapshots for Spark Savings total assets, deployed assets, and Ethena cooldown balances, covering Ethereum, Gnosis, Arbitrum, Base, and 4 other chains
- **External data**: End-of-day price feeds for deposited and borrowed assets

## 3. Step-by-step calculation

### Component A: SparkLend idle deposits and active loans

- **Purpose**: To determine the total value deposited in SparkLend, consisting of idle assets and active borrows.
- **Logic**: Idle deposits are calculated by tracking cumulative net flows of underlying assets, where inflows and outflows are extracted from ERC-20 token transfers to and from SparkLend deposit asset addresses. Active loans are calculated by tracking the scaled principal balance of variable debt assets, which is updated via `Mint` and `Burn` events from `VariableDebtToken` contracts. This scaled balance is multiplied by the variable borrow index, which is updated via `ReserveDataUpdated` events from `Pool` contracts, to determine the total active borrows. All asset amounts are normalized by decimal precision.
- **Key fields**: `amount`, `index_value`, `token_address`

### Component B: Spark Savings idle capital

- **Purpose**: To determine the idle capital held in Spark Savings vaults.
- **Logic**: Total assets for Spark Savings V1 and V2 vaults are extracted from on-chain state snapshots. For Spark Savings V2, deployed assets (representing capital allocated to the Spark Liquidity Layer) are also extracted from state snapshots. Idle capital is calculated by subtracting the deployed assets from the total assets for V2 vaults, while total assets are used directly for V1 vaults. All asset amounts are normalized by decimal precision.
- **Key fields**: `amount`, `token_address`, `version`

### Component C: Spark Liquidity Layer assets under management

- **Purpose**: To calculate the total assets managed by the Spark Liquidity Layer.
- **Logic**: Net balance changes are tracked by aggregating ERC-20 token transfers to and from designated Spark Liquidity Layer proxy and pool accounts, while non-AUM assets are excluded. For rebasing assets, such as Aave V3 and SparkLend deposit assets, balances are adjusted using the respective liquidity index updated via `ReserveDataUpdated` events from `LendingPool` and `Pool` contracts, and tracked via `Mint`, `Burn`, and `BalanceTransfer` events from `AToken` and `spToken` contracts. Ethena cooldown balances are also integrated from on-chain state snapshots. All balances are normalized by decimal precision.
- **Key fields**: `amount`, `index_value`, `token_address`

### Component D: Double-counting exclusion and USD conversion

- **Purpose**: To aggregate the components into a single TVL metric while preventing double counting and converting to USD.
- **Logic**: SparkLend deposits held by the Spark Liquidity Layer are identified and excluded from the Spark Liquidity Layer assets under management. The final TVL is calculated by summing the SparkLend idle deposits, SparkLend active loans, Spark Savings idle capital, and the adjusted Spark Liquidity Layer assets under management. Daily values are forward-filled to account for days without activity, and asset balances are multiplied by end-of-day asset prices to convert the values to USD.
- **Key fields**: `tvl`, `price_usd`, `timestamp`

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

Total value locked (TVL)

Total value locked (TVL) — value deposited into the project's applications.

SparkSPK

Total value locked

Total value locked

Competitive landscape

Chain breakdown