Challenges and opportunities
Recently, we have observed two significant trends in the blockchain field:
First, with the booming development of blockchain ecosystems and applications, the activities of personal accounts on the chain have become more diverse and complex, which has left users with a more comprehensive digital footprint on the chain. Rich on-chain interactions have also brought many new opportunities, such as building on-chain credit scores based on users' asset portfolios and transaction histories. This development trend is likely to promote innovations in traditional financial practices such as non-overcollateralized lending protocols.
Secondly, as the industry matures, project owners are increasingly focusing on understanding user portraits and assessing user quality when developing targeted airdrop strategies or analyzing potential user groups. However, they face a huge challenge: EVM accounts are essentially transaction triggers, not "bank accounts" like traditional finance. The existence of this structural form limits in-depth analysis of account-level data, making it difficult for project owners to gain meaningful insights into their user groups.
These trends highlight the urgent need for advanced tools that can effectively analyze and interpret account-level data. To address this need, Hemera Protocol is developing a new generation of account-centric indexing protocols that aim to provide developers with standardized account-level data retrieval and feature development services.
This innovation is further extended to an Account-Centric Indexing (ACI) network, which ensures the decentralization of data management. Leveraging this network, Hemera provides a powerful solution for processing and analyzing blockchain data, making it easy to build a variety of user-oriented applications and suitable for a variety of use cases.
What are account-level extracts?
When an exchange occurs on the blockchain, the raw data does not appear intuitively as a "transaction". If you retrieve this data directly from a blockchain node, it may look like the blue diagram below - a complex array of information that is neither semantic nor easy to analyze. Essentially, it records who sent the transaction, where it was sent, and what was entered.
The process of data extraction is to transform these incomprehensible data into meaningful semantics. For example, in transactions, the main information we care about is the type and quantity of input and output tokens and the total value of the transaction. Therefore, a typical transaction data extraction may include key elements such as the initiator of the transaction, the token bought, the token sold, the quantity of each, and the transaction amount.
What is really important is to understand how these transactions affect the overall state of your account. For example, a transaction will change your token balance, affect your transaction volume, increase the total gas fees you pay, and increase your number of on-chain interactions.
Hemera encapsulates these account-level changes in a semantic and standardized format, called "features". The red diagram below shows how this exchange transaction affects your account status. Through the User Defined Function (UDF) module, developers can design and define these functions according to specific needs. This data extraction method presents complex blockchain data in a more intuitive way, making development work more convenient and easier to draw meaningful analytical conclusions.
How do applications leverage the Hemera Protocol?
It is very simple for applications to use Hemera for in-depth account-level analysis. Applications simply query the required functions from the ACI network. Hemera provides a range of built-in functions, including common indicators such as portfolio balances, trading volume, collateralized lending, and even social data. Developers can easily query any account of interest without writing additional code.
Hemera also provides UDF modules to meet the more personalized needs of developers. With this module, developers can create custom functions with minimal coding work. Thanks to the standardized format, these custom functions can be seamlessly integrated into existing functions, thus facilitating more comprehensive and detailed analysis.
When functions are requested, they are transmitted to Hemera's ACI network. Each indexer in the network is responsible for indexing one or more specific functions by querying the blockchain raw data. The requester then receives account-level semantic data provided by Hemera, avoiding the tediousness of directly processing complex raw data.
Using Hemera’s ACI network, applications can gain deep insights into account activity and user profiles. This not only helps make smarter decisions in the blockchain ecosystem, but also provides a more personalized user experience while simplifying the complexity of processing raw blockchain data.
How are features updated?
Functions in the Hemera ecosystem are dynamically updated based on on-chain transactions. Each indexer in the ACI network is responsible for indexing a specific function. These indexers run in real-time, constantly listening to on-chain transactions related to the functions they track.
The streamlined workflow begins with developers defining custom functions and data classes in a UDF module. The indexer then captures relevant on-chain transactions, filters the specified data, and passes it to the developer’s trigger function. The function then processes the data, updating the account-centric functions.
The real-time indexing and updating process ensures the timeliness and accuracy of functional data. By focusing on transactions that specifically affect its assigned functions, the indexer is able to efficiently process large amounts of blockchain data and transform it into meaningful information at the account level.
How do I create a UDF?
To illustrate the process of building a custom UDF, we take OpenSea related functions as an example:
Function definition: Developers first define specific functions (such as NFT transaction volume, NFT transaction collection) and data classes. For example, the "Opensea_Order" class may include variables such as "order_hash" (string), "offerer" (string), "recipient" (string), and "offer" (dictionary).
Triggers and function development: Next, developers create triggers, specifying which transactions or log events change the state of the defined function. This step ensures that the indexer effectively captures relevant on-chain transactions and filters the data of interest. Developers can then write custom logic to update the function value.
Query execution: Finally, the developer runs the indexer and the new UDF and processes the historical data. The feature data is then stored in the database, from which the developer can easily retrieve account-level data or query feature values through the standard REST API.
For a more detailed guide on implementing UDFs, see our documentation: https://docs.thehemera.com
In essence, each function is a dynamic entity that is continuously updated based on on-chain activity. The role of UDF is to interpret the raw transaction data, create appropriate triggers, and use these triggers to keep the function continuously updated.
Hemera’s unique positioning in data analytics
To understand Hemera’s unique position in the data space, it can be compared to existing data analysis tools and protocols. The following highlights the main points of difference:
Hemera's core advantage lies in its account-centric data extraction
Dune's Spellbook also focuses on data extraction, abstracting high-level data patterns from raw data, but it is not account-centric.
The Graph, while a popular indexing protocol, is centered around smart contract events. It is great for decoding smart contract-related data, but lacks in extracting account-centric data.
Debank has the ability to extract account-level data, but it mainly focuses on DeFi assets, which limits its applicability to diverse on-chain activities (such as on-chain game records).
Hemera offers unparalleled flexibility in data processing
With Hemera, users can easily manipulate the data they need with minimal coding.
The Graph allows users to build subgraphs for smart contract data retrieval, but restricts deeper data operations.
Debank provides a fixed API to query asset information, but lacks programmability.
Hemera's data update method is proactive and real-time
The Hemera Indexer continuously monitors the blockchain, updating functionality in real time via a “push” mechanism triggered by transactions.
In contrast, Dune updates data in a "pull" manner, requiring users to obtain all data from the database to update their own data.
In the field of Web3 applications, Hemera is superior in composability and integration
Hemera provides curated data in standard formats, simplifying database construction and maintenance.
Dune provides a comprehensive on-chain data query terminal within its platform, but the cost of building independent applications is high.
Debank’s fixed API structure limits it to processing only asset-related data.
Looking ahead, as the industry develops and matures, we believe that user portraits will be critical for user-facing applications. In this evolving environment, Hemera's account-level, multi-dimensional data will play an indispensable role in making user targeting more accurate and effective. In addition, we foresee that the synergy between AI models and on-chain data will increase day by day, and Hemera's semantic extraction capabilities will help large language models understand and process data more efficiently.
Hemera's vision is clear and explicit: We are committed to building Hemera into a comprehensive data center for the entire Web3 industry. By significantly lowering the threshold for access and utilization, we will provide services to developers, researchers, marketing teams, end users, AI systems and other users. Hemera's ultimate goal is to become the preferred platform for all users who want to fully tap the value of blockchain data, allowing blockchain ecosystem users to easily access important resources and benefit.
Discover Hemera Products:
Website: https://thehemera.com/
Develop applications based on Hemera ACI network:
Documentation: https://indexer-docs.thehemera.com/
Github: GitHub - HemeraProtocol/hemera-indexer at pre-release/v0.3.0
Visit the SocialScan Agent Store:
Website: https://socialscan.io/home
Follow us on Twitter:
Hemera: https://x.com/HemeraProtocol
SocialScan: https://x.com/socialscan_io
