Using a data warehouse to spend less _ MySQL

When you walk into the boss's office and face corporate decision makers, do you know what to say to them? How can we sell our ideas to decision makers? How can they accept your opinion? Suggestions for implementing data warehouse

You know that you need a data warehouse and have made up your mind: your competitors have predicted market trends earlier than you, other companies are also more responsive than you in the Web-driven market, and you are not enough to catch up with them. In addition, data queries and report requests from various departments almost dragged your business system down. All of this shows that you really need a data warehouse.

When you enter the boss's office and face corporate decision makers, do you know what to say to them? How can we sell our ideas to decision makers? How can we let them accept your opinion and invest in this intangible project without a clear budget or a clear degree of benefit? Below are some suggestions for you.

Return on Investment

When you present your data warehouse implementation plan to decision makers, do not use the return on investment as a bargaining chip for investment. Because you have never determined what the data warehouse will do for the growth of production efficiency or market understanding, you cannot provide a return on investment plan. Without any data warehouse experience, you cannot quantify the benefits it brings. In fact, the primary reason for implementing a data warehouse is to develop the enterprise's business intelligence capabilities, which will teach you how to clearly measure the growth of market understanding. That is to say, you can develop a tool that can determine your own value at the time of application and funding.

Even if you have developed a product with data analysis and performance monitoring capabilities, you still cannot describe what you ultimately want to measure. Why? Because higher-level business performance measurements are based on more detailed and lower-level measurements, these are the fields in which you will be responsible for data analysis at the department level. Before the product is formed, they do not know what to do.

It sounds so mysterious, but these mysterious numbers are indeed useful. You can say the following content, and I will tell you how to say it. Although no specific data is attached, the logic is strong and cannot be refuted.

Advantages of data warehouse

1. multi-level trend analysis. Your financial staff, sales staff, and marketing staff will be able to define and analyze market trends in a multi-level manner, covering a wide range to the entire market, small to the market of a region or any defined range. They can also control the prediction accuracy, because both the quality and measurement accuracy of data sources can be controlled.

2. enterprise-wide performance monitoring. The same level of analysis capability can be applied to departments, business units, and the entire enterprise. You can develop and continuously improve to measure the performance of the entire enterprise.

3. user-defined, user-controlled reports. This is especially important because it sounds like a trend that follows the world. In fact, we have not exaggerated its capabilities at all, and this is exactly what you need.

Your case

Now let's take a look at your report system. Taking order quantity as an example, the system sends a large number of reports to different personnel in different departments. The problem is that no one can process data including orders, order history, and data from different databases (such as customer tables), so they cannot summarize the required information. Where is the problem? First, because the report is static to a large extent and the required information is mostly from different databases, multiple requests are required to obtain the report, resulting in high operation costs.

Now you can let decision makers choose to invest one by one to implement individual functions or to implement all functions at a time. You must let them know clearly that the ultimate goal of data warehouse implementation is to give your user group a unique application, which can replace several previous programs. Investment in a large program is enough to implement a data warehouse, and the functions it brings will be more powerful. This is the enterprise's pursuit of benefits.

Another advantage of a data warehouse is that it can implement a system called the general manager Information System (EIS. EIS is an application that provides any information that managers need in decision reports by category. In general, the manager will not deal with the problem of data analysis encountered by other users, and they will not process data aggregation for four or five data sources. They only need to obtain the most accurate information in the simplest way. For such users, data warehouses can make their dreams come true.

Who are the beneficiaries?

If you cannot clearly describe this, the decision makers will not pay for it. Data Warehouse is a powerful strategic tool that can greatly enhance the competitiveness of enterprises. If you cannot propose a specific amount of investment, it can be said that a very conservative figure of 0.1 million US dollars, and the maximum investment will certainly not reach seven digits.

So What equipment do you need to buy? If you are currently using an ERP platform (sap r/3, Oracle, PeopleSoft), you can purchase the corresponding development kit, the tools can help you extract and load data into programs and develop data mining and analysis programs. If you do not have an ERP platform yet, you need to purchase a storage device and use the extract-transform-load (ETL) software to import data to the data warehouse, and the software used for data mining and analysis (Online Analytical Processing, or OLAP ). In addition, you also need to conduct business training for employees so that they can be competent for data analysis, and also help employees change their thinking model of transaction processing, this allows them to smoothly develop reports (further discussions will be made in subsequent articles ).

Finally, you need to make a summary. this is what policymakers need to hear: through data warehouses, your user team can do things with less money.