IoTeX is the blockchain platform for Real-World AI. Since 2017, its foundational infrastructure has delivered verified, real-time data from the real world to AI systems and decentralized applications. Powering 100+ projects and 40M devices across mobility, robotics, energy, health, and more, IoTeX enables developers to build next-gen AI models and applications that deliver real-world impact. The IoTeX tech stack provides the data, identity, and verification layers to fulfill the surging demand for real-world data in the new AI era. Trusted by partners like Google, Samsung, IEEE, ARM, and Nordic Semiconductor, IoTeX is uniquely positioned to capture the multi-trillion-dollar AI and data economy.

website

twitter

github

discord

Blockchains (L1)

IoTeX

# IoTeX Active addresses (weekly) Methodology

## 1. Overview

Weekly active addresses measures the number of unique addresses that initiated at least one transaction on the IoTeX blockchain within a rolling 7-day window. This metric provides insight into IoTeX network adoption and user engagement by tracking distinct addresses that demonstrate on-chain activity. The calculation uses HyperLogLog probabilistic counting to efficiently handle deduplication across time windows while maintaining accuracy for large-scale blockchain data.

## 2. Data sources

The metric relies on the following raw data sources:

- **IoTeX blockchain transaction records**: Complete transaction history from the IoTeX blockchain, including transaction metadata such as sender addresses, timestamps, and transaction hashes
- **Transaction metadata**: Block timestamps and transaction identifiers used for temporal grouping and deduplication

## 3. Step-by-step calculation

### Component A: Transaction sender extraction

- **Purpose**: Identifies all unique addresses that initiated transactions on the IoTeX blockchain
- **Logic**: Extracts transaction sender addresses from IoTeX blockchain transaction records. Each transaction record contains a `from_address` field representing the address that initiated the transaction. Addresses are normalized to lowercase format to ensure consistent identification across the blockchain. Each sender is classified with metadata including chain identifier ('iotex'), market sector ('blockchains_l1'), business segment ('blockchain'), and user type ('Transaction sender'). The transaction timestamp and hash are preserved for downstream temporal aggregation.
- **Key fields**: `from_address` (normalized to lowercase as `user_address`), `block_timestamp`, `transaction_hash`, chain and classification metadata

### Component B: Daily active user aggregation

- **Purpose**: Creates daily snapshots of unique transaction senders
- **Logic**: Groups transaction sender data by day using timestamp truncation to day granularity. For each day, identifies all unique addresses that sent at least one transaction, creating a distinct list of daily active users. The aggregation maintains chain, user type, and product classification metadata while focusing on the `user_address` field as the primary identifier for deduplication.
- **Key fields**: `timestamp` (truncated to day), `user_address` (renamed to `user_dau`), `chain`, `user_type`, `product_id`

### Component C: Weekly active user calculation with HyperLogLog

- **Purpose**: Calculates weekly active users using a rolling 7-day window with efficient probabilistic counting
- **Logic**: Creates HyperLogLog sketches for each day's unique users to enable memory-efficient deduplication. Generates a complete datespine covering all days from the first available data point to the current date. For each target day, constructs a 7-day rolling window including the target day plus the 6 preceding days. Joins this rolling window with the daily HyperLogLog sketches, then merges the HLL sketches across the 7-day period and extracts the final approximate count of unique users. This approach handles missing days gracefully and provides efficient deduplication across time boundaries. The calculation produces breakdowns across multiple dimensions: project-level ('project_dbt_tag'), chain-level ('chain'), user type-level ('user_type'), and product-level ('product_id').
- **Key fields**: `timestamp`, `user_wau` (extracted from merged HLL sketches), `dimensions` (breakdown category), dimension-specific identifiers

### Final aggregation

- **Combining components**: Filters the weekly active user results to include only 'chain' and 'project_dbt_tag' dimensional breakdowns, which provide the most relevant aggregation levels for the IoTeX ecosystem analysis
- **Field transformation**: Renames the `user_wau` field to `active_addresses_weekly` to align with standard metric naming conventions and clearly indicate that the metric represents unique blockchain addresses
- **Time boundaries**: Uses daily timestamp boundaries with 7-day rolling windows, where each day's metric represents unique addresses active in that day plus the 6 preceding days
- **Final calculation**: The final metric value represents the approximate count of unique IoTeX addresses that initiated at least one transaction within the 7-day rolling window, calculated using HyperLogLog probabilistic counting for computational efficiency while maintaining high accuracy for cardinality estimation

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

Unique active addresses over a rolling 7-day window.

IoTeXIOTX

Archived

Active addresses (weekly)

Active addresses (weekly)

Competitive landscape