The following article is from talking about data, by Shi Xiufeng
Source: Talking about data, author: Shi Xiufeng
Introduction: In this article, we review the concept and characteristics of data assets, explain the objectives, contents and main solutions of data asset management, take stock of the current mainstream data asset management frameworks in the industry, and give the main work content of the three stages of data asset management according to the author's understanding.
01 General consensus on data assets
With regard to the definition of "data assets", the industry's usual way refers to the concept of "assets" in basic accounting: assets refer to resources formed by past transactions or events of an enterprise, owned or controlled by the enterprise, and expected to bring economic benefits to the enterprise.
For example, the Institute of Information and Communications Technology gave a similar definition in the "Data Asset Management Practice White Paper 5.0": data assets (Data Assets) refer to data resources legally owned or controlled by enterprises, recorded electronically, such as text, images, voices, videos, web pages, databases, sensor signals and other structured or unstructured data, which can be measured or traded, and can directly or indirectly bring economic and social benefits.
1. The same characteristics of data assets and assets are:
First, the enterprise legally owns and controls it. It pointed out the ownership of data assets, which must be legally owned and controlled to be the assets of the enterprise, and if they are obtained illegally, they cannot be counted as the assets of the enterprise, and may also bring legal risks to the enterprise. This is clearly stipulated in Article 47 of the National Data Security Law, which draws a red line for the management and trading of enterprises' data, and if it crosses the line, it may be punished 10 times the penalty obtained by illegal transactions.
Second, it is expected to bring direct or indirect economic benefits to enterprises. It points out the value characteristics of data assets, the data that can bring direct or indirect economic and social benefits to the enterprise is the asset of the enterprise, and the data that cannot generate value is not only not an asset, but also wastes the resources of the enterprise and belongs to the cost of the enterprise.
2. There are also differences between data assets and assets, such as:
First, reproducibility. The replicability of data makes data assets simple and can be shared infinitely, which makes the wide utilization and value release of data assets produce unlimited imaginable space.
Second, value uncertainty. Everyone knows that data is valuable, but the value it can generate in different enterprises and different business scenarios is different, and it is difficult to measure it in the form of currency. Regarding the valuation of data assets, there are some research results in the industry, such as the cost method, the income method and the market method, but these theories still lack practical tests and need to be further improved.
Third, virtuality. Data assets do not have a physical form, can not be seen, can not be touched, which is very similar to intangible assets (technology patents, goodwill, etc.), but also different. Intangible assets are non-monetary long-term assets of an enterprise. Data assets are monetary, and this monetary nature is valuable in a particular environment, and once it is out of the environment, it may be worthless. Like Bitcoin, it's valuable in the "mining farm" that hosts it, but if you take it out and put it in another network environment, Bitcoin is just an unrecognizable string of strings.
The above is the individual's naïve perception of the same and different points of data assets and assets. There are also experts who study the characteristics of data assets from the perspective of economics, such as: data assets are not applicable to diminishing marginal utility, price elasticity, data assets are non-exclusive private assets, etc. These definitions are too professional, and the author does not understand them deeply enough, so I will not get an axe here.
02 What is the data asset management?
There is also a general consensus about data assets: not all data is data assets, and those that do not bring value and benefits to the enterprise are just data resources. Data asset management is the process of how to turn data resources into (transform) data assets.
Baidu Encyclopedia defines: "Data asset management (DAM) is a set of business functions that plan, control and provide data and information assets, including developing, executing and supervising plans, policies, programs, projects, processes, methods and procedures related to data, thereby controlling, protecting, delivering and improving the value of data assets." ”
I prefer this definition, which has a pleasing point is that it integrates the concepts of data governance and data management, for example, "plans, policies, programs" are generally considered to be the content of data governance, focusing on planning; while "projects, processes, methods and procedures" is the content of data management, focusing on implementation. If we insist on "finding differences" between the three, the author believes that data asset management is more committed to solving the three major problems in data governance/management, namely: difficult to confirm data rights, difficult to protect security, and difficult to assess value.
1. Ownership of data
First, the data is stored in electronic form, and the electronic data has the characteristics of easy deletion, easy to change, easy to copy, easy to disseminate, non-exclusive, etc., which brings great difficulties to the confirmation of data rights. For example, if a data set of A is transferred to B and B is processed and processed, is this processed data set attributable to A or to B?
Second, unlike physical assets, data will not only not be exhausted in the process of circulation, but will be used more and more, which will also increase the difficulty of determining the right to confirm data.
Third, production data does not necessarily own data. This is easier to understand, we will produce a lot of data every day on the Internet, but whether it is our personal social data or e-commerce consumption data, it is actually not controlled by producers (users), but is regarded as a core asset by Internet companies. Internet companies will even unknowingly, through the analysis and mining of the data we produce and then profit from us.
2. Data security issues and personal privacy protection issues
Recently, the author was targeted by scammers, and almost every day I can receive two or three calls from overseas, saying that I still have a balance on a P2P platform that has not been refunded, so I can add their QQ group. In fact, the author has what P2P platform, a long time ago to "cut the meat off the car", app has long been uninstalled.
There is no doubt that the author's personal information has been illegally leaked!
Every time I see a phone call from abroad, my heart is like a hundred thousand "grass mud horses" galloping...
Some people say: "Under the Internet, in front of big data, each of us is "running naked", where is the privacy? Now that I think about it, this sentence is not unreasonable.
I don't know if you have encountered this situation, a few days ago on the T treasure search for a product, but when you open P duoduo, found that P duoduo home page actually appeared in the T treasure search for the product. I believe that TBao will not share data generously with competitors, so how does P Duoduo get this data? Was it the internal ghost in the T treasure that leaked the data, or did P Duoduo break through the T treasure database? The answer is unknown, but it's safe to say that privacy breaches are nothing new.
When you shop on Taobao, post on Weibo, brush videos on Douyin, order food on Meituan, hail a taxi on Didi... inadvertently, there is no personal privacy.
You see the world in the Internet, and the world looks at you on the Internet!
The "Data Security Law" and "Personal Information Protection Law" have been officially implemented, and these damn bad businesses or malicious individuals are still so rampant! The author appeals: the crackdown on personal privacy data protection and illegal data transactions should be strengthened and severely cracked down!
3. Difficulty in assessing the value of data assets
Data value has timeliness, variability, data may depreciate over time (may also increase in value); data value has infinity, unlike other physical assets, data assets can be used indefinitely, so its value is difficult to estimate; data value has a scenario, in different scenarios the value of data played is different, even if the relevant data is used differently, different people use the value of its production is not the same; data value is also uncertain, the legal supervision of data, Issues such as privacy also have a substantial impact on the value of data.
This is an interesting question, and I want to write a long article to talk to everyone. Those who are interested can have a "share, like, watch" triple combo at the end of the article, so stay tuned.
03 Data Asset Management Reference Framework
At present, many enterprises and research institutions have formed some very good results based on their own practices and research, let's take a look at it together.
1, Shintsuin
The reference framework for data asset management given by the INTDC in the "Data Asset Management Practice White Paper 5.0" is as follows:
Source: Data Asset Management Practice White Paper 5.0
The white paper proposes that achieving raw data to data assets requires two steps:
(1) Data resources. It mainly focuses on data governance, with the goal of improving data quality and ensuring data security, ensuring the accuracy, consistency, timeliness and integrity of data, and promoting the internal and external circulation of data. The activities and functions at this stage are mainly: data model management, data standard management, data quality management, master data management, data security management, metadata management, data development management, etc.
(2) Data assetization. It mainly focuses on expanding the application scope of data assets, explicitly standardizing the costs and benefits of data assets, and forming a benign feedback closed loop between the data supply side and the data consumer side. Including data asset circulation, data asset operation, data value evaluation and other activities.
2. China Southern Power Grid
In December 2021, China Southern Power Grid Corporation released the "White Paper on Data Asset Management System of China Southern Power Grid", which has a great highlight, which for the first time put forward the concept of "power data elements". Power data elements refer to data resources invested in the production and operation links of electric energy production, storage, transmission, trading, consumption, etc., and are integrated with other production factors and continuously iterated to improve the efficiency of electric energy production and consumption, including data, data models, data products, data services and other forms。
Source:China Southern Power Grid Data Asset Management System White Paper
The data asset management system proposed in the white paper is guided by the external environment such as national policies, laws and regulations, industry norms, and ecological development, and is mainly composed of six modules of data strategy, data governance, data operation, data circulation, organizational guarantee, and technical support, with a total of 36 management functions and 8 links in the whole life cycle of data assets, which covers the data asset management work field of China Southern Power Grid Company relatively completely by clearly defining the positioning and internal relationship of various functional activities.
At the same time, the data asset management system also clarifies the specific work that needs to be carried out around each link of the whole life cycle of data assets, puts forward various detailed management requirements in a targeted manner, ensures that the implementation process is accurate and in place, and is committed to achieving the company's high-quality development, "three-business transformation", world-class enterprises, and "double carbon goals" of the company's strategic goals.
3. PwC & Huawei & CHINA AUTOMOBILE data
In the "White Paper on The Realization of Data Assets and Business Value of Car Companies" jointly released by PwC, Huawei and China Automobile Data, a very good data asset management framework is also proposed to help car companies establish a data asset management system, tap the value of data assets, and transform traditional car companies into data service enterprises.
Source: White Paper on Data Assets and Business Value Realization of Car Companies
The data asset management framework mainly involves three parts:
(1) Data governance, including: enterprise data strategy, governance of organizational talents, standardization of business processes, to provide basic guarantee for data-driven business operations;
(2) Unified data assets, including: data asset catalogs, data standards, enterprise-level data models, data distribution, data maps, etc., providing design guidance from data generation, lake entry, connection, and application throughout the life cycle.
(3) Data operation: Provide data operation mechanism and responsibilities, establish operational indicators (such as data service construction cycle, data demand response cycle, etc.), and ensure the continuous benign operation of data management.
4. Data asset management system oriented to value realization
The following system framework comes from a paper by the big data journal BDR entitled "Construction of Data Asset Management System Oriented to Value Realization".
This diagram is very interesting, the amount of information is very large, it divides data asset management into 4 levels, 3 stages:
Source: Big Data Research (BDR), authors: Li Yufei, Liu Haiyan, Yan Shu
On the basis of the data asset management system oriented to value realization, the data asset management carried out by enterprises is usually divided into three stages: pre-stage, mid-term and late stage, and the focus and output of each stage in terms of safeguard measures, data management functions, technology platforms and data operations are different to ensure that enterprises plan data asset management as a whole, promote data asset management in an orderly manner, and gradually release the value of data assets.
Data operation refers to the data value-added activities driven by data applications and services that enterprises already have good data asset management capabilities, usually in the later stage of data asset management, to release a steady stream of data value to drive the continuous updating of enterprise data asset management strategic planning, management functions and technology platforms.
The data management function includes eight management functions: data standard management, data quality management, metadata management, master data management, data model management, data sharing management, data security management, and data value management. Data management functions focus on different stages of data asset management.
The technology platform is a powerful tool to assist enterprises to efficiently carry out data asset management, mainly including data collection, data storage, data processing and data analysis. In the medium-term phase, the results of the previous period of data asset management (such as data standards, data models) were brought together, and data quality audits and metadata management analysis were carried out. Provide data analysis and mining and data service sharing in the later stage.
Safeguard measures usually include five aspects: strategic planning, organizational structure, institutional system, audit mechanism and training and publicity. Safeguard measures are the basis for ensuring the orderly development of enterprise data asset management, starting in the early stage of data asset management and running through the later stage of data asset management.
04 Steps for data asset management
Data asset management is a systematic thing, involving more content, more complex, from the above data asset management framework is not difficult to find this. Therefore, as a high-level application of data governance, data asset management is not achieved overnight, and the goal of data asset management needs to be gradually realized. From the perspective of implementation and landing, the author believes that data asset management can be divided into three steps (or three stages) in general: data asset inventory, data asset management, and data asset operation.
1. Inventory of data assets
To create the basic ability of enterprises to identify and inventory data assets, the main goal of this stage is to "find out the bottom of the family", clarify what data the enterprise has, what do they "look like", where?
(1) Clarify the goals and scope of data combing and inventory
(2) From the perspective of business, the data is sorted out from the business domain: business domain - business theme - entity data - data attributes
(3) From the perspective of IT, data inventory is carried out from the information system: information system - table / view - data field - data dictionary
(4) Establish data standards, including data asset coding, data classification and grading, and data asset catalog
(5) Establish the mapping relationship between data assets from the business perspective and data assets from the IT perspective, and provide query views from two perspectives
2. Data asset management
To build the management capability of enterprise data assets, the main goal of this stage is to form a working environment for managing and using data assets within the enterprise to ensure that enterprise data assets can be managed and landed.
(1) Data asset management platform
(2) Data assets are collected into the lake, data standards are landed, and data quality is audited
(3) Data asset authorization and access control
(4) Establish a data asset management organization system to ensure that relevant matters are responsible for special personnel
(5) Establish a data asset management process system to ensure the safe and controlled and compliant use of data
3. Data asset operation
To build the operational capability of enterprise data assets, the main goal of this stage is "data value", truly realize the empowerment of data for business, ensure the good use of enterprise data assets, and enhance the competitiveness of enterprises.
(1) Data exchange and sharing
(2) Data analysis and mining
(3) Valuation of data assets
(4) Data asset trading and circulation
(5) Data asset operation monitoring and audit
Transferred from the public number: ERP House