With the popularity of big data, more and more experts and enterprises begin to agree that data analysis will become an important basis for management support in the future. However, it is not an easy task for enterprises to establish a data analysis system. During the construction process, they will encounter various special and difficult to quantify needs, especially for group companies. To this end, here is a speech on the establishment of a data analysis center shared by Yang Hao, CIO of Hunan Mengjie Home Textiles, at the Fanruan Big Data Tour, hoping to inspire fellow readers.
Two pain points hinder data management
For enterprise data management, taking Mengjie Home Textiles as an example, the number one pain point is that the data of each group adds up to far more than 500 fields. Such fields are easy to cause conflicts for group companies. For example, in a group-based heavy-duty manufacturing company, one of its steel products is made as a general material in one company, but only as a reinforcing plate in another company, and the company believes that this steel is only an auxiliary material , and accessories do not need to participate in the budget, which causes trouble in data statistics management.
Another pain point is efficiency. According to statistics, when the enterprise information system is not very powerful, production managers spend 80% of their time on business management, and the rest of the time is spent on making reports and distributing them, which is very inefficient, and with the system There are more and more interfaces, occupying a lot of resources. All these pain points hinder the data management of enterprises and must be solved by subdivision.
Building a Data Center - Branch System Management
Taking Mengjie's data center basic architecture as an example, network virtualization is used in the middle of the data center, the data center is the core, and all subsystems are imported into this data center. There are various branch systems below the center, and the communication between branch systems. For example, branch system B wants to obtain a data from branch system A, branch system B will send a data reorganization request, and branch system A will synchronize the data to the data center area. After receiving the request, the data center will synchronize the data to a virtual branch data center, and then directly distribute the data to the B branch system, and the B branch system can be used by some means.
Data visualization management operation case
Mengjie Home Textiles pays the most attention to customers, products, services and channels in terms of company operations. This is also the focus of data-based operations, which is reflected in the following cases.
1. Visual display of operational data, timely decision-making and judgment
In terms of visualization, we will use intuitive data display to reflect the operation of the company. For example, establishing a comparison of indicators in the same period, distinguishing them from yellow, red and blue, these indicators will be monitored through membership, retail, service, and store opening.
The display of daily retail data is not only a display, but through further mining, you can learn about the data, services, and store openings of sub-brands. Through the sales data, you can further drill down to the sales of each region, so as to make decisions and determine which regions can invest more energy.
For the analysis of each store, from the data display level, you can drill down to understand the store's plans, membership services, shopping guide behavior, etc. Through real-time monitoring data, once a problem occurs, it can be analyzed in time to determine which link has the situation.
The above data display and analysis are all realized through the development of the FineReport report of FanRuan. FineReport has strong data processing capabilities. More than 10,000 pieces of sales data are added every day, including inventory data. Using FineReport, the data can be run out in just 4 seconds.
2. FineReport filling function solves the problem of timeliness of information
There will be a lot of data and information collection in the usual business process, including the information of competing products. In the past, it was collected once in the city on a regular basis, then collected in the large area, and then collected on the Internet. The time consumed has long passed the expiration date. Using FineReport's reporting function, you can track and collect real-time conditions, and fill in the system and report to the database.