在《“以数据为中心的业务变革”之三种范式》中，我们提到“业务活动步骤化”和“步骤环节要素化”作为业务数字化的前两个阶段，与信息系统无关。从业务数字化第三阶段的的“要素数据规格化”开始，涉及到了信息系统和数据。尤其是到了“以数据为中心的业务变革”阶段，也就是真正的“数字化转型”阶段，数据开始登上中心舞台。
那么在数字化转型中，企业怎样做好数据治理，又怎样以数据为中心优化业务呢？
一、像管理业务一样治理数据
（一）指定数据管理的职能部门，明确组织管理体系
企业做好数据治理需要建立数据治理的管理决策体系，指定CDO（首席数据官）和归口职能管理部门，并建立相应机制。如成立数据治理领导小组，由企业分管领导作为CDO任组长，由数据管理部门领导牵头，各业务部门及IT部门专人参加，明确议事范围、决策机制、工作机制等。
部分企业将数据治理工作视为IT技术工作的一部分交给IT部门，往往会出现两方面问题：一是IT部门作为技术部门，在实际中往往突出数据治理的技术属性而忽视其管理属性，将数据战略、数据架构作为制定IT战略和IT架构工作的一部分来开展。但数据治理的价值不是IT价值或技术价值，而是业务“降本增效”价值，业务部门的缺位和管理属性的缺失将导致后续工作偏离数据治理价值目标；二是IT部门作为实施部门不具有职能管理权限，在开展数据治理工作时缺少相应的职能权力和资源，影响数据治理工作权威性和强制性；
（二）将数据治理作为一项单独的业务，而不是业务的附属
将数据治理作为一项单独的业务，其内涵是将数据治理作为企业业务集合中的重要一项，单独配备人、财、物等资源，以企业独立业务工作的相关标准要求去组织、策划和实施，而不是作为业务工作的附属。做到有战略、有规划、有计划、有制度、有流程、有激励、有监督、有检查、有总结考核。
部分企业按照所属业务板块的不同，将数据治理工作直接交给各业务部门独立负责，缺乏顶层数据战略、规划、计划以及制度、流程，会出现两方面问题：一是数据共享难。在“业务活动步骤化”、“步骤环节要素化”和“要素数据规格化”的业务数字化过程中，根据业务价值需要，业务活动中需要使用的数据很大一部分由其他业务活动产生。而在数据治理工作中业务部门各自为战会导致数据孤岛出现；二是主数据管理缺失，基础数据不一致。主数据包含企业人、财、物等基础数据，主数据建设和管理的缺失将导致企业在大量业务活动中出现由基础数据不一致导致的业务混乱现象；三是存在数据安全隐患。信息系统由IT部门统建统管后，业务部门缺乏建设数据安全所需的资源和能力。
2023年的《党和国家机构改革方案》中明确成立国家数据局，负责协调推进数据基础制度建设，统筹数据资源整合共享和开发利用，统筹推进数字中国、数字经济、数字社会规划和建设等，这是从国家层面将数据管理作为一项单独业务指定专门机构负责的新举措。企业在实践中，不是说一定要成立新的内设部门专职从事数据治理，而是要把数据治理作为一项单独的业务，按照企业业务管理的标准要求，授权某一职能部门实施独立管理。数字化转型网www.szhzxw.cn
二、像开展业务活动一样开展数据治理
（一）实现数据治理活动的计划、组织、协调、控制闭环
数据治理活动从开展数据治理评估、建设数据架构开始，对于数据标准建设、主数据管理、元数据管理、业务管理数据和时序数据管理、数据指标管理、数据质量管理、数据安全管理、数据共享与利用管理等业务，都应当按照业务管理的一般要求，建立制度、明确流程、开展运行评估和监督检查、实施奖惩、考核和总结，实现计划、组织、协调、控制全链路闭环。数字化转型网www.szhzxw.cn
（二）将数据治理融入业务管理制度、业务活动流程和业务控制要素
虽然数据治理是一项单独的业务，但数据治理的对象—数据来源于业务活动，服务于业务活动。因此在开展数据治理时，需要由数据管理部门经数据治理领导小组授权，组织各业务部门将数据治理融入各业务板块。具体包括三个层面：
1、将数据治理各总体制度的要求融入具体业务制度；
2、在业务活动的流程步骤上落实制度要求；
3、在业务步骤的各环节将数据治理的控制措施要素化。
三、与业务数字化同步实施数据治理
在《坚持业务和技术协同推动企业数字化转型》中，我们提到，企业需要同步推动业务数字化和信息系统设计选型。与此同时，结合将数据治理融入业务管理制度、业务活动流程和业务控制要素的要求，在业务数字化以单项业务活动为单位，以业务价值目标为导向，结合相关约束条件和业务活动间的协同关系，依次开展“业务活动步骤化”、“步骤环节要素化”、“要素数据规格化”的过程中，将数据治理的要求落实在业务数字化全过程中。如下图所示：
同时，数据治理作为一项单独的业务，也在此过程中，纳入企业业务体系，一并开展数据治理业务的数字化。数字化转型网www.szhzxw.cn
四、像分析数据一样优化业务
在《“以数据为中心的业务变革”之三种范式》中，我们提出了以“数字看板”、“数字工程”、“大数据和数据智能应用”为代表的“以数据为中心的业务变革”三种范式。
以“数字看板”为代表的第一种范式，以步骤定时记录的方式，将业务步骤执行过程中反映业务活动全过程、全状态信息的数据记录下来，实现业务价值流动可视化；通过显式化业务流程规则并将其数据化、规格化，在业务运行过程中发挥看板对业务的拉动机制，加速用户价值流动，并有效驱动数据更新和状态跳转；通过为各类资源建立状态参数，动态监测资源占用情况，有效暴露资源瓶颈。以价值流动数据化、数据可视化的方式，通过直接观测数据可观察业务瓶颈，有效推动业务优化。
以“数字工程”为代表的第二种范式，在装备的全生命周期内将设计、研制、生产、测试、运行中产生的各类数据在数字样机和物理实体装备之间实时传递，以实现数字样机在仿真运行过程中与物理实体装备开展迭代和同步更新完善。以实体装备数字孪生体——数字样机依托模型数据和模型规则开展仿真运算，帮助设计师在数字环境中以仿真测试的手段快速迭代和优化设计方案，降低成本和潜在风险，缩短产品研发周期。
以“大数据和数据智能应用”为代表的第三种范式，通过在业务的数据密集环节建立机器学习模型，将业务数据同时作为模型的训练样本和测试样本；或建立大数据模型，通过数据科学相关处理方法，创新知识生成方式。在互联网营销推荐、网上娱乐和阅读推荐、金融风控、公共安全、基于经营数据的决策辅助等业务领域有广泛应用。以数据作为模型的训练手段，将缺乏现有规则的业务问题转化为数据问题来解决。
数据治理是手段，运用数据解决业务需求，优化业务模式才是数据治理的根本目的。
英文翻译：
In the Three paradigms of Data-Centric Business Change, we refer to the “step of business activities”. And “step element” as the first two stages of business digitization, which have nothing to do with information systems. Starting from the “element data normalization” of the third stage of business digitization, it involves information system and data. In particular, when it comes to the “data-centric business transformation” phase, which is the true “digital transformation” phase, data begins to take center stage.
So in the digital transformation, how do enterprises do a good job of data governance. And how to optimize the business with data as the center?
Govern data like a business
(1) Specify the functional departments of data management and clarify the organizational management system
To do a good job in data governance, enterprises need to establish a management decision-making system for data governance, designate a CDO (chief Data Officer). And a centralized functional management department, and establish a corresponding mechanism. For example, the data governance leading group is set up, which is led by the corporate management leader as the CDO leader, led by the data management department. And attended by the business departments and IT departments to clarify the scope of discussion, decision-making mechanism, and working mechanism.
Some enterprises regard data governance work as a part of IT technical work and entrust IT departments with two problems: First, as a technical department, IT departments tend to highlight the technical attributes of data governance while ignoring its management attributes in practice, and carry out data strategy and data architecture as part of the work of formulating IT strategy and IT architecture. However, the value of data governance is not the value of IT or technology, but the value of business “cost reduction and efficiency”. The absence of business departments and management attributes will lead to subsequent work deviating from the value target of data governance. Second, the IT department, as the implementation department, does not have functional management authority and lacks corresponding functional power and resources when carrying out data governance work, which affects the authority and compulsion of data governance work;数字化转型网www.szhzxw.cn
(2) Treat data governance as a separate business, rather than an adjunct to it
The connotation of data governance as a separate business is to take data governance as an important part of the enterprise business set, separately equipped with human, financial, material and other resources, according to the relevant standards of the enterprise independent business work to organize, plan and implement, rather than as a subsidiary of the business work. There is a strategy, a plan, a plan, a system, a process, an incentive, a supervision, an inspection, a summary assessment.
Some enterprises according to the different business segments, the data governance work directly to each business department is independently responsible for the lack of top-level data strategy, planning, planning and systems, processes, there will be two problems: one is difficult to share data. In the business digitization process of “business activity step”, “step element” and “factor data normalization”, according to the needs of business value, a large part of the data needed in business activities is generated by other business activities. In the data governance work, the business departments will fight against each other, which will lead to data silos.数字化转型网www.szhzxw.cn
(3) The second is the lack of master data management.
And the basic data is inconsistent. Master data contains basic data such as human, financial and material. The absence of master data construction. And management will lead to business chaos caused by inconsistency of basic data in a large number of business activities. Third, there are data security risks. When the information system is built and managed by the IT department. The business department lacks the resources and capabilities required for data security.
In 2023, the “Party and State Institutions Reform Plan” clearly established the National Data Bureau, responsible for coordinating. And promoting the construction of data infrastructure systems, coordinating the integration and sharing of data resources. And the development and utilization of data resources, coordinating and promoting the planning. And construction of digital China, digital economy, digital society, etc., which is a new measure from the national level to designate data management as a separate business to be responsible for specialized agencies. In practice, it is not necessary to set up a new internal department to engage in data governance. But to take data governance as a separate business. And authorize a functional department to implement independent management in accordance with the standard requirements of enterprise business management.数字化转型网www.szhzxw.cn
Conduct data governance as a business activity
(1) To achieve the planning, organization, coordination and control of data governance activities closed loop
Data governance activities start from data governance evaluation and data architecture construction. For data standard construction, master data management, metadata management, business management data and time series data management, data index management, data quality management, data security management, data sharing and utilization management. And other services, the general requirements of business management should be followed. Establish a system, clarify the process, carry out operational evaluation and supervision and inspection, implement rewards and punishments, assessment. And summary, and realize the planning, organization, coordination and control of the full link closed-loop.
(2)Integrate data governance into business management systems, business activity processes and business control elements
While data governance is a separate business, the object of data governance – data that originates from and serves business activities. Therefore, when carrying out data governance, it is necessary for the data management department to be authorized by the data Governance Leading group. And organize all business departments to integrate data governance into each business segment. Specifically, it includes three levels:
- Integrate the requirements of the overall system of data governance into specific business systems;
- Implement the system requirements in the process steps of business activities;
- Factor the control measures of data governance in each link of the business step.
- Implement data governance simultaneously with business digitization
In “Adhering to Business and Technology Collaboration to Promote Enterprise Digital Transformation”, we mentioned that enterprises need to promote business digitalization and information system design and selection simultaneously. At the same time, combined with the requirements of integrating data governance into business management system, business activity process. And business control elements, business digitalization takes individual business activities as the unit, takes business value goal as the orientation . And combines relevant constraints and synergies between business activities. In the process of “business activity step”, “step element” and “factor data normalization”. The requirements of data governance are implemented in the whole process of business digitization. As shown in the picture below:
(3)Data governance into business digitization
At the same time, data governance, as a separate business, is also incorporated into the enterprise business system in this process, and the digitization of data governance business is carried out.
Optimize your business as you analyze data
In “Three paradigms of Data-Centric Business Change”, we put forward three paradigms of “data-centric business change”, represented by “digital Kanban”, “digital engineering”, “Big data and data intelligence application”.数字化转型网www.szhzxw.cn
The first paradigm, represented by “digital Kanban”, records the data reflecting the whole process.
And state information of business activities during the execution of business steps in the way of step periodic recording, and realizes the visualization of business value flow; Through explicit business process rules and their datatization and standardization, Kanban can exert the pulling mechanism of business in the process of business operation, accelerate the flow of user value, and effectively drive data update and state jump.
By establishing status parameters for various resources, dynamic monitoring of resource occupancy can effectively expose resource bottlenecks. By means of value flow digitization and data visualization, business bottlenecks can be observed through direct observation of data, and business optimization can be effectively promoted.数字化转型网www.szhzxw.cn
The second paradigm, represented by “digital engineering”, transmits all kinds of data generated during design, development, production, testing.
And operation in real time between digital prototype and physical equipment during the whole life cycle of equipment, so as to realize the iteration and synchronous update and improvement of digital prototype and physical equipment during simulation operation. The digital twin of physical equipment – digital prototype relies on model data and model rules to carry out simulation operations, helping designers to quickly iterate and optimize design schemes by means of simulation testing in the digital environment, reduce costs and potential risks, and shorten product development cycle.
The third paradigm, represented by “big data and data intelligent application”, establishes a machine learning model in the data-intensive part of the business.
And uses the business data as both training and test samples of the model; Or build a big data model and innovate the way of knowledge generation through data science related processing methods. It is widely used in Internet marketing recommendation, online entertainment and reading recommendation, financial risk control, public safety, decision aid based on business data and other business fields. Using data as the training means of the model, business problems that lack existing rules are transformed into data problems to solve.数字化转型网www.szhzxw.cn
Data governance is a means, using data to solve business needs, optimize business model is the fundamental purpose of data governance.
本文由数字化转型网（www.szhzxw.cn）转载而成，来源于 数字化老兵；编辑/翻译：数字化转型网小汤圆。
免责声明: 本网站(https://www.szhzxw.cn/)内容主要来自原创、合作媒体供稿和第三方投稿，凡在本网站出现的信息，均仅供参考。本网站将尽力确保所提供信息的准确性及可靠性，但不保证有关资料的准确性及可靠性，读者在使用前请进一步核实，并对任何自主决定的行为负责。本网站对有关资料所引致的错误、不确或遗漏，概不负任何法律责任。
本网站刊载的所有内容(包括但不仅限文字、图片、LOGO、音频、视频、软件、程序等) 版权归原作者所有。任何单位或个人认为本网站中的内容可能涉嫌侵犯其知识产权或存在不实内容时，请及时通知本站，予以删除。
