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

Bancor is a non-custodial exchange protocol built on Ethereum. Bancor was founded by Galia Benartzi, Guy Benartzi, Eyal Hertzog, & Yudi Levi and launched in 2016.

BNT tokenholders own and govern the protocol. They currently take a cut of the total trading fees paid by traders (revenue). Bancor's investors include DeFiance Capital, ParaFi Capital, Mechanism Capital, etc.

governance

github

twitter

discord

etherscan

website

Exchanges (DEX)

Bancor

# Bancor Active users (daily) Methodology

## 1. Overview

The Bancor daily active users (DAU) metric measures the number of unique addresses that executed trades on the Bancor protocol each day. This metric provides insights into protocol adoption and user engagement by tracking trading activity across Bancor's decentralized exchange aggregator functionality. The calculation uses HyperLogLog data structures for efficient distinct user counting across multiple dimensional breakdowns.

## 2. Data sources

- **Bancor protocol events**: Parsed Ethereum event logs from Bancor V1 protocol contracts, including `BancorNetwork_TokensTraded_event` and `Converter_Conversion_event` logs containing trade execution data
- **Ethereum transaction records**: Transaction data providing user addresses, timestamps, and transaction metadata for trade execution context
- **Chain availability data**: Configuration data defining supported blockchain networks and their availability windows

## 3. Step-by-step calculation

### Component A: Trade event extraction
- **Purpose**: Extract all trading activity from Bancor V1 protocol on Ethereum
- **Logic**: Collects trading events from two contract versions covering `BancorNetwork_TokensTraded_event` (v2.1) and `Converter_Conversion_event` (v3) logs. Joins these events with Ethereum transaction records to obtain user addresses from the `from_address` field. Filters trades to those occurring after June 16, 2020. Normalizes native token addresses by converting the native token identifier `0xeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee` to WETH address `0xc02aaa39b223fe8d0a0e5c4f27ead9083c756cc2`
- **Key fields**: `block_timestamp` for daily aggregation, `user_address` for distinct counting, `transaction_hash` and `log_index` for event identification, `chain` identifier, token addresses and amounts for trade context

### Component B: Daily user identification  
- **Purpose**: Identify unique users per day and classify them by type
- **Logic**: Truncates block timestamps to daily boundaries using day-level granularity. Creates distinct records for each unique user address per day. Classifies all users as 'Trader' type since the metric focuses on trading activity. Includes dimensional attributes like `chain`, `market_sector_id` (exchanges), `business_segment_id` (dex_aggregator), and `product_id` for breakdown analysis
- **Key fields**: `timestamp` truncated to day, `user_dau` containing unique user addresses, `user_type` set to 'Trader', dimensional fields for aggregation breakdowns

### Component C: HyperLogLog aggregation
- **Purpose**: Efficiently count distinct users across multiple dimensions using probabilistic data structures
- **Logic**: Initializes HyperLogLog structures with precision 15 for each combination of timestamp, grain, chain, user type, and product. Groups data by these dimensions to prepare for distinct counting. Uses HLL for scalable distinct user estimation across large datasets
- **Key fields**: `hll` containing HyperLogLog structures, dimensional grouping fields for aggregation boundaries

### Component D: Dimensional breakdowns
- **Purpose**: Create multiple views of DAU data across different analytical dimensions
- **Logic**: Generates four separate dimensional views by merging HLL structures and extracting distinct counts. Creates breakdowns by project level (all dimensions null), chain level (chain populated), user type level (user_type populated), and product level (product_id populated). Uses `hll_count.merge_partial` to combine HLL structures and `hll_count.extract` to get final distinct counts
- **Key fields**: `dimensions` indicating breakdown type, `chain_id`, `user_type`, `product_id` for dimensional analysis, `user_dau` containing distinct user counts

### Final aggregation
- **Combining components**: Unions all four dimensional breakdowns (project, chain, user_type, product_id) into a single result set using UNION ALL operations
- **Time aggregation**: Uses daily timestamp boundaries with 'day' granularity for consistent time-based analysis
- **Final calculation**: Applies the `finalize_metric_user_dau` macro with 'fill_zeros' strategy to standardize schema, handle missing days by filling with zero values, and apply consistent formatting. Filters final results using chain availability filters to ensure data quality across supported networks

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

Unique revenue-generating users over a fixed 24-hour window.

BancorBNT

Active users (daily)

Active users (daily)

Competitive landscape

Chain breakdown