Summary: Over the past year, more and more enterprises are considering or starting data governance projects. As someone who has worked in this field for many years, I am often asked: isn’t data governance a concept that has existed for many years? Why are a lot of enterprises mentioning suddenly? Is it old wine in new bottles? What is the relationship between data center and data center? In this article, I will help you to clarify these questions through three core questions, and more importantly, three questions that an enterprise must consider before starting a data governance project.
Ask: purpose, why want to manage
This is an issue that must be addressed before a data governance project can be launched. Governance is not governance for governance’s sake; governance itself is worthless.
We believe that the value of governance lies in building a good data of reliable quality, safe and controllable, and convenient service within the enterprise, so that the value of data can be released. In other words: The ultimate purpose of data governance is to release data value and lay the foundation for the release of data value.
With the acceleration of the digital transformation process in all walks of life, the foundation of business digitalization is becoming increasingly solid. Data-driven business or data-driven operation is no longer the patent of traditional head enterprises and large Internet companies, and more and more enterprises are embracing data and consuming data. Therefore, the demand for data governance shows an increasingly strong trend.
Under the strong appeal of data asset governance, enterprises must reach a unified understanding from top to bottom before starting the project: the ultimate purpose of data governance is to drive data consumption and release data value.
Q: What is the goal of governance
This is a question that needs to be refined and clearly answered before a data governance project can be launched. Data governance doesn’t happen overnight. We need to be clear about what our long-term goals are, but also what our short-term project scope and goals are.
It is not difficult to understand the purpose of data governance. The periodic goals of data governance projects can be set from the release of data value of the ultimate purpose, or from the perfection of the data system. In general, the latter is a common immediate target.
Of course, those who have been in the data field for years know that data governance is not a quick fix. Generally, we carry out work around the four key areas of “cost, quality, safety and service” and set goals. Goal setting can use the following ideas.
- Production economy refers to the cost aspect, which aims at controlling and even saving costs. The actual operation needs to be disassembled to the cost control of business lines or different fields, such as sales business lines, market launch lines, etc., as well as basic data lines and flow log data lines.
- Reliable quality refers to the quality aspect, which aims at reducing data quality problems and narrowing data quality risk exposure. The actual operation needs to be disintegrated into application scenarios or different data levels, such as commodity data and sales data, etc., as well as the data quality involved in executive data kanban and driving business operation.
- Controllable security refers to the security aspect, which aims at meeting compliance requirements and preventing data leakage. The actual operation needs to be disassembled into different data types or processes, such as privacy data and regulatory data submission. Often, the capacity building of security system is an important work.
- In terms of convenient service, it aims at providing abundant data and driving data consumption. In actual operation, it needs to be disintegrated into different business lines, data types and service scenarios, such as marketing and promotion, business decision-making, etc.
From the perspective of specific work implementation, cost is the focus of data to a certain volume, quality is the most important and challenging area, security is the foundation, and service is the key area that needs to pay high attention to and invest in creating data value.
Therefore, before the start of the project, it is necessary to make clear which or which areas the project scope focuses on (e.g., cost and quality), what the governance scope is (e.g., customer data, regulatory submission scenarios, traffic log line data), and what the core objectives are (e.g., In 3 months, the project went online, XXX cost was saved, the number of P0 data quality failures was 0 in 6 months, XXX data was put on the shelves, and the data service system capacity was built to provide services for XXX and XXX users). In short, the domain, scope, and core goals must be selected and set before starting a governance project.
Three questions: method, how to implement governance
How should specific data governance be implemented? This is the key problem to ensure the success of data governance projects, involving the design of personnel organization and power and responsibility, the design of process specifications, tool selection of three aspects of the basic work, but also around the target setting of the field, scope and core goals to carry out work.
- Due to data quality problems caused many reasons, there may be technical data construction and development, may have provided input has flaws, business level may have organizational mechanism is not sound management level lead to no propulsion, etc., so the data quality and management in parts, determine the quality of the water and exposure to risk, And formulate the data quality management plan of the whole link before and after the event.
- Release data value is the ultimate goal of data management project and so on the one hand, through to the enterprise all the data inventory and management of the whole domain data gathers data on the supply side, systematic to organize data, and abundant data information, on the other hand strengthen operational training and promotion, establish good asset retrieval and be sure to experience, and access, analysis, such as data service link, Expand the demand side consumption of data assets. At the same time, it is essential to systematically present the asset capability and asset valuation.
- In terms of data security, the key is to classify and classify data after data identification, and adopt different authorization policies for data with different privacy and security levels. Whether from the source of data collection, authorization, or from the external introduction of cooperation, compliance is a very important topic in the field of data security today, so the relevant process mechanism and capacity building can also be an important part of the implementation.
- In terms of data cost, it focuses on discovering and dealing with the waste of data storage and computing, and realizes cost saving by analyzing, setting up governance items and starting governance.
Therefore, before starting a governance project, the basic governance implementation path must be defined, including organizational assurance, process specification, tool improvement, and related areas, such as quality: Comprehensive assessment, quality risk control of data construction within the scope of the project from the perspective of full link, establishment of data quality failure system to improve response capacity, and focus on value export, inventory and operation promotion of data assets, so that data can be used.
Write in the last
As a data person who has been working in this field for many years, I have been reflecting on why data governance has been promoted for so many years. There are few successful cases of governance projects in the past. Now, when it is mentioned again, the project has a much higher probability of success, mainly for the following reasons:
- In the wave of digital transformation, data governance has become more objective, which is long-term capacity building rather than campaign projects.
- Data centralization and data governance are combined. The unified aggregation of data provides the foundation for data governance. The systematic and standardized data construction makes data governance come before the event, and full-link data governance rather than point-and-post data governance
- A group of people who truly understand data and have practiced large-scale data construction and data operation provide services for data governance, actual combat precipitation rather than armchair thinking
Ali cloud data asset governance solutions, is based on the extension of the data platform, with the platform methodology, so that all enterprises have an enterprise-level good data!
Ali Cloud data center is the only commercial output of Alibaba data Center. With data Center methodology as the core, it builds an intelligent big data system of “fast, accurate, complete, unified and universal”.
The product matrix of Ali Cloud Data center takes Dataphin as the base and Quick series as the business scene:
- Dataphin, intelligent data construction and governance
- Quick BI, data visualization analysis
- Quick Audience, one-stop consumer operation and management
- Quick Tracking, global behavioral insight
- Quick Stock, smart goods operation
- Quick Decision, a digital engine for risk control decisions
At present, we are exporting a series of solutions, including general data solution, retail data solution, financial data solution, Internet data solution, etc.
The original link
This article is the original content of Aliyun and shall not be reproduced without permission.