In April 2020, the "Opinions on Building a More Complete Market-oriented Allocation System and Mechanism of Factors" issued by the Central Committee of the Communist Party of China and the State Council formally listed data as a separate factor of production and proposed reform directions to promote the market-oriented allocation of data factors. Subsequently, in policy documents such as the "Opinions of the Central Committee of the Communist Party of China and the State Council on Accelerating the Construction of a National Unified Market", the "14th Five-Year Plan for the Development of the Digital Economy" and the 26th meeting of the Central Committee for Deepening Reform, it was further proposed to promote the smooth flow of data element resources on a larger scale, accelerate the construction of a data infrastructure system, promote the healthy and orderly development of my country's data element market, and accelerate the construction of an digital economy with data as a key element.
In the era of data elements, building an enterprise data asset management system based on data governance , providing usable and useful data, supporting enterprise business process transformation, product innovation, risk prevention and control, continuously improving enterprise data capabilities, and tapping the value of enterprise data assets has become the only way for enterprises to digitally transform.
General Manager of China Electronics Financial Information Business Analysis Division
Chairman of China Electronics Financial Information Data R&D Committee Du Xiaozheng
Data governance: the "cornerstone" of data asset management
Driven by the internal driving force of enterprise digital transformation and the external driving force of the country's strategic layout of data elements, data assetization has become unstoppable. data asset management is the most important management tool on the road to data assetization. Major financial institutions, consulting agencies, and manufacturers have invested heavily in in-depth research, forming unique data asset management development paths. Over the past decade, the data management model of financial institutions has undergone great changes, from data resource management and control to data asset valueization. However, no matter what changes are made, the availability and ease of use of data are still the prerequisite, and data governance is a key link and the basic project for realizing data asset management.
Data governance is a complete management framework for realizing data resource utilization. The industry already has a very mature governance theory system. China Electronics Jinxin believes that the core goal of data governance is to solve the problems of data availability and data security. It should not only pursue the widespread and efficient circulation of data, but also ensure data security and protect personal privacy.
Faced with the increasingly realistic demand for data value mining and assetization, data governance is particularly important under the practical constraints of difficulty in confirming the ownership of data assets. Data governance works through data model management, data standard management, data quality management , master data management, data security management, metadata management, data development management and other data resource activities to clarify the full-link responsibilities of data creation, collection, processing and application to ensure that the enterprise It improves the accuracy, consistency, timeliness and completeness of industrial data, improves data quality, ensures data security, promotes internal and external data circulation, transforms raw data into data resources, makes the data have a certain potential value, and then gradually transforms data resources into data assets through data capitalization and asset valuing. Without a data governance system as a guarantee, data will not only not be transformed into enterprise assets , but it will also easily cause enterprises to fall into the trap of "data swamp". A good data governance system will lay a solid foundation for data asset management and is the cornerstone of enterprise digital transformation under the new situation.
New situation, data governance faces new requirements
In the context of the digital economy, data is an important basic resource, and the processing and use of massive data has become the norm. China Electronics Jinxin believes that data governance, as a "project under the iceberg", is not only a "family affair" of the enterprise, but also an important and basic component of the development of the national digital economy.In the era of data elements, regulatory reporting, data services, agile development, security compliance and other aspects have put forward new requirements for data governance. Data governance will become the core driving force on the development side and server side of data, as well as in the internal use and external circulation of data.
1. Requirements for regulatory reporting
The "China Banking and Insurance Regulatory Commission Standardization Specifications for Supervisory Data for Banking Financial Institutions (2021 Edition) (EAST5.0)" puts forward higher requirements for regulatory reporting by financial institutions: data must be submitted completely, accurately, and comprehensively; reporting specification requirements must be strictly implemented; data reporting linkage must be strengthened, etc. However, the current overall status of regulatory submissions is that manual data verification requires a heavy workload and is inefficient, and the quality of submissions cannot meet the requirements of timeliness, completeness, accuracy and traceability. Through data governance, identify system breakpoint processes, improve data links, speed up the location of data quality issues, continuously improve the quality and timeliness of data submission, continue to prevent financial risks , and promote the data governance and compliance development of financial institutions.
2. Requirements for data products/services
The "Financial Technology Development Plan (2022-2025)" mentions: deepen the comprehensive application of data, continuously expand the breadth and depth of data elements in the financial industry, and create an enterprise-level data service capability center that is technology-empowered, data-driven, and business-linked. Data services are not limited to data services such as fixed reports and management cockpits, but also include value mining and in-depth analysis of massive and diverse data using joint modeling, graph computing, digital twins and other technical means. It is necessary to build a user-oriented and scenario-oriented big data knowledge graph and comprehensive analysis capabilities. Through data governance, we integrate diverse data inside and outside financial institutions to provide complete, consistent, accurate and efficient data services.
3. Requirements for agile data development
"Guiding Opinions of the General Office of the China Banking and Insurance Regulatory Commission on the Digital Transformation of the Banking and Insurance Industry" mentioned: Promote agile transformation of technology management. By establishing a basic framework for data governance, one-stop data collection, processing, operation and maintenance, and service processes (DataOps) are realized, focusing on the full-link process of collaboration from data requirement input to deliverable output, clarifying the purpose of R&D operations, and refining implementation steps. With the support of system tools, organizational models, and security risk management , data R&D operations can be integrated, agile, standardized, automated, intelligent, and value explicit, and continuously improve the collaborative efficiency between data organizations and data systems.
4. Requirements for data security compliance
The "Financial Technology Development Plan (2022-2025)" mentions: Do a good job in data security protection. Data must be secure and compliant while not affecting the sharing of data services. The key is to classify and classify data. Through data governance, classifying and controlling data with different importance and impact is an important task in the current data security management of financial institutions.
Four keys to hone the "sharp tool" of data governance
Under the trend of digital transformation, external supervision and internal data use have put forward more efficient, more accurate, more complete and more compliant requirements for data governance. How can enterprises seize the requirements of the new situation and carry out their own data governance? China Electronics Finance has long been committed to the practice and research in the field of data for financial institutions, and has gradually accumulated a complete set of methodologies for data asset management. Looking at the development process of data governance and analyzing the construction path of data governance, according to the recently released "China Electronics Jinxin Data Governance White Paper", continuous investment in data governance can start from the following four directions:
1. Overall planning, comprehensive layout Data governance, solving overhead problems
Data governance has been practiced in the domestic financial industry for more than 20 years and has accumulated a lot of successful experience. However, when enterprises carry out data governance work, they must plan and implement it in conjunction with their own business strategic development, IT planning, and data management status.The system-driven solution is built from the top down and can help enterprises achieve an understanding and consensus on the overall picture of data governance, which is conducive to promoting follow-up work. The specific practice process needs to be combined with the actual situation, focusing on actual data quality issues, continuously improving data quality and even business quality, and ultimately realizing the value of data.
The data governance system generally includes: data governance guarantee mechanism, data governance platform, and data governance activities. In addition to top-level strategic support, the data governance guarantee mechanism also includes data governance organizational structure, data governance management methods, and data governance management processes. The data governance platform is the carrier for implementing data governance activities. It generally consists of a data standard management module, a data quality management module, an metadata management module, and a data governance portal. Data governance activities include: data standard management activities, data quality management activities, metadata management activities, data security management activities, life cycle management activities, data model management, data development management, etc.
2. Standards first, determine the direction of data access and improvement, and solve the starting point problem
With the general trend of domestic database migration in recent years, through data standards sorting, implementing data standards in domestic database is also a good starting point for data governance. Data standards-driven service solutions generally include the following activities: current situation research, standard design, standard mapping, standard execution and standard management.
When starting a data governance project with data standards as the starting point, you must wait for the opportunity, such as cooperating with the new/renovated source system, migration of master data construction data platform, etc., to ensure the actual implementation effect of data governance. Data standard management must establish a preservation mechanism for data standards. At present, enterprises generally have three modes of data standard preservation management: beforehand, during the process, and afterward. Prior management and control involves intervention during the demand analysis stage of project/system construction, allowing system construction to be designed around data standards at the beginning. In-process management and control refers to the management and control during the project/system development process. With the help of tools such as data research and development workstations, data standards are implemented during the data development process. Post-event management is to conduct audits during the project launch phase, explain the reasons for non-compliance with standard requirements, and update the data standards if the standards need to be updated.
3. Local improvements, seeking the best cost-effectiveness of data governance, and solving efficiency problems
Local improvements can be based on improving high-priority data quality issues or meeting urgent regulatory requirements to partially promote data governance and seek the best cost-effectiveness of data governance.
The quality of data is the most direct reflection of the quality of data governance. Therefore, financial institutions generally launch special data quality improvement projects to meet regulatory requirements, and also use data quality work to continuously improve the enterprise-level data governance system.
Data quality management work has been carried out in financial institutions for nearly 20 years. The key is to do two tasks well, one is data accountability, and the other is quality assessment and evaluation. When taking responsibility for data, the responsibilities of the data owner department, data development department, and data entry department can be comprehensively considered. The quality assessment evaluation can be designed from the following dimensions: evaluate the overall data quality of the customer to form an evaluation system (data subject, owner, quality dimensions, etc.); evaluate the work of each R&D team in the quality rectification process to form an assessment system (problem rectification rate of the month, on-time feedback rate, etc.); evaluate the deployed quality rules to form an assessment system (effective). Finally, based on this, high-priority data quality issues are determined and promoted.
In addition, China Banking and Insurance Regulatory Commission "Notice on Carrying out Special Governance Work on Supervisory Data Quality" puts forward the overall requirements of "raising awareness and consolidating responsibilities; highlighting key points and addressing both symptoms and root causes; strengthening rectification and improving mechanisms", and establishes a one-year special governance work on regulatory data quality, covering all banks and insurance financial institutions. This is also a good starting point for data governance for commercial banks.
4. Upgrade and transformation, with the implementation of data governance in the new generation of IT construction, solve the timing problem
Taking the construction of the new generation core project group as an opportunity, through data standard construction, data development management, data asset operation and other governance activities, the entire process of data governance is integrated into the life cycle of IT projects: planning requirements, design and development, testing and online, operation and maintenance and other stages. In the planning requirements stage, data governance personnel participate in system requirements and architecture reviews, and make suggestions that are consistent with data governance. In the design and development stage, data models are constructed with reference to data standards through integrated development tools to ensure the consistency of standards implementation. During the test and launch phase, the data governance team will review and launch the DDL, and it will only be allowed to go online if it meets the governance requirements. In the operation and maintenance stage, effective management and efficiency of data resources are achieved; more data assets are gradually introduced to increase the proportion of data assets under control.
In the era of data elements, giving full play to resource advantages and releasing the maximum value of data cannot be separated from the "refining into gold" behind data governance. China Electronics Jinxin's data governance consulting system and Yuanqi data asset platform products aim at data asset accumulation and data value creation, using big data technology, AI technology and data security technology to create a data intelligence base that integrates data management and control platform, data middle platform and AI platform. It combines industry precipitation and data governance to achieve data-driven business operations and lean management, complete the conversion of business digitization, data capitalization and asset valueization, assist financial institutions in rapid data capitalization, and comprehensively support the digital transformation of enterprises.
