How to analyze the relationship between enterprise digital transformation and data asset management?

Link to the original text ： Data assets are king , Analyze the relationship between enterprise digital transformation and data asset management

Video review ： Click here

Courseware acquisition ： Click here

One 、 Background introduction to data assets

With the deepening of enterprise digital transformation , Explosive growth of data volume , How to control data production cost 、 Find valuable data , Improve the data ROI, It has become a key task in the middle and later stages of enterprise digital transformation , This is also the ultimate goal of data asset management .

In the process of digital transformation, it is not plain sailing , There are many problems . How to control production cost 、 Discover the malignant cost factors 、 Quantify production costs ？ How to find value data 、 Sort out the data distribution 、 Analyze data value ？ These are all issues that enterprises need to consider on the way to realize the digital transformation of enterprises .

How to solve the above problems , The key point is to use methodology + Tool ideas to build a data asset management platform .

Two 、 Data asset management methodology

So how to give methodology for data asset management ？ We can start from the following aspects ：

1、 How to sort out data ： Metadata management

Before combing the data , First of all, we should define three goals ：

• Goal one ： Identify what data is available
• Goal two ： Understand the role of these data
• Goal three ： Convenient and unified query entry

When the goal is clear , Five steps of metadata management , Let's sort out the data ：

• First step ： Collecting metadata
• The second step ： Define meta model
• The third step ： Maintain metadata
• Step four ： Metadata quality
• Step five ： Metadata query

2、 How to improve data normalization ： Data standards + Data model

After metadata collection , The second thing to do is data standards and data models , There is no clear separation between this step and metadata management , However, when doing data asset management in general enterprises, metadata management is usually done first , Because it works faster .

There are many problems in the traditional data development method , for example ： Table names lack constraint specifications , The output level cannot be directly determined from the table name 、 Data meaning ; Field lacks comments ; The field name is not strict ,status Equal to status , Not equal to order status ; There is no data lifecycle defined lifecycle etc. .

The standard method is layered in the bottom log warehouse 、 Subject domain, etc , Establish data standards and then conduct unified model design , Then the model is reviewed and released , Improve the overall data normalization .

3、 How to improve data governance

The technology department and the business department jointly determine the data quality improvement scenarios , Starting from the business scenario , Build from top to bottom , It is not up to developers to temporarily verify what they think , Include ：

• Starting from the business scenario , Solve real business data problems , Clarify the construction objectives , Ensure that there is no deviation in the implementation strategy
• Offline research 、 collect 、 Sorting out data problems , Manage by business objectives
• Conduct rule set management according to business classification , Continuously optimize according to data quality feedback 、 Precipitation rule set

4、 How to quantify data value

Quantitative data value starts from two directions , One is data cost , One is data value .

Data cost includes construction cost and resource cost , The construction cost cannot be calculated alone , Only according to the project cost （ Central Taiwan Construction 、 Delivery and implementation ）、 Labor costs are estimated ; Resource costs can be quantified , It is also the part that we need to focus on , Including storage unit price 、 Calculate unit price, etc .

Direct and indirect benefits of data value , Direct revenue refers to some cross group data sharing 、 The scenario of data encapsulation for external sales , Directly quantifiable revenue ; Indirect income means that most of the data is used to serve the business development within the enterprise , It is not possible to quantify benefits directly , Value can only be analyzed indirectly from these two dimensions ： Data usage heat 、 Upstream and downstream data dependency .

Through the reasonable calculation of data cost and data value , We can quantify the value of data .

3、 ... and 、 Data assets product introduction

We have talked about a lot of methodology for you , Is there a product that can actually solve the problems of the enterprise , Help enterprises build a data asset management platform ？ Here we introduce the data asset products of kangaroo cloud stack —— Hucang data assets （DataAssets）.

** Hucang data asset management platform （DataAssets）,** Through metadata management 、 Data standard specification design 、 Data quality improvement 、 Data heat and cost calculation , Optimize data production costs 、 Quantify the value of data revenue , It is widely used in the middle and later stages of Lake warehouse construction .

The product feature of hucang data asset management platform is to achieve the effect of closed-loop governance through a unified platform .

The product mainly includes the following functions ：

• Access metadata for management

• Metadata query

• Data standards management

• Data model management

• Data quality management

At the same time, the product has the following values

• carding / Manage enterprise data

Comprehensive collection 、 management 、 Inventory data , Let us know and understand what data the enterprise has

• Improve data standardization

Build a data standard system , Through the automatic introduction of data model , Complete the standard model design

• Improve data quality

Configure through prior rules 、 The rules in the process are implemented 、 After the event quality report , Find problem data in time , Effectively avoid the spread of problems

• Quantify data value

By analyzing the cost of data 、 earnings , Reasonably evaluate the value of each data

Four 、 Data asset customer case

Next, let's share three actual cases of using customers , Introduce how the data asset management platform can help customers solve problems .

• Finance ： One stop metadata management platform of a payment group

• Finance ： A bank's data quality management improvement project

• manufacture ： A chip manufacturing enterprise unified data asset management platform

Open source project library address ：https://github.com/DTStack