The characteristics of the implementation of master data are large workload and the need for more human resource and material resource. Thus, reasonable project management is more important. Project management is the various management activities which are conducted that uses all kinds of knowledge, skills, methods and tools to meet or exceed the requirements and expectations of the project's relevant parties. The implementation of information system via the method of project management is the effective planning, organization, coordination, leading and control on the whole process of information system implementation under the constraint condition of limited resources. The project implementation management implemented by the master data management system generally includes aspects such as project planning, project organization and coordination, project schedule and risk control.
Project planning includes task decomposition and defining task relationships. The master data project organization should establish the project leadership committee and the project management office to coordinate resources, logistics supports and the overall control on solutions; afterwards, establishing the project team according to business types. Project audit and control act a critical role on whether the whole system can be implemented within the budget on time. The audit and control of the information system are generally realized through activities such as the supervision and inspection on the implementation conditions of work plans, the analysis and solving on the reasons of plan delay. Project risk control reduces the probability of risk events and minimize the loss when risk occurs mainly through risk identification, risk analysis, risk response planning formulation and risk monitoring.

After the master data management system has been implemented and run for a period of time, an overall evaluation on the system should be performed. System performance and running effects are evaluated according to the records of user's feedbacks and operation situations to see if it has reached the design target, and further improvement advises are put forward. Finally, the results of the review are formed to be written system evaluation reports. 

System evaluation generally includes: system operation, master data quality evaluation, system performance evaluation, system management level evaluation and system economic evaluation.

 

In order to avoid significant problems as much as possible, and avoid affecting the normal business operation severely, test running should be performed before the system goes into operation, and be spread after the success.

The purpose of test running is to simulate the match degree between check operation functions and demands, and to find out potential problems and errors of software. It evaluates the feasibility of the system before submission, and verifies the stability and robustness of the system, and accumulates valuable experiences for the formal operation of the system. 

The main task of the test running includes making the plan of system test running, environmental data and personnel preparation of the test running, checking the process of system testing running, system improvement and perfection, sorting out project related documents, the generation of the latest on-line version and preparation for the formal operation.
The on-line running of the master data system involves the system switch in relevant business systems. System switch is not a pure issue of information technology, attentions also need to be paid on issues of project management. The ways of system switch include direct switching, parallel switching and subsection switching.

System testing is a very long stage in information system development cycle. It is the final gate that guarantees system quality and reliability, and it is the final review on system analysis, system design and system implementation.

Testing steps include unit testing, integration testing, confirmation testing, system testing, performance testing and interface testing.

As the IT infrastructure and data entry standard of an enterprise, all relevant business systems have related interfaces with the master data management system. Therefore, interface testing is the top priority of the tesing work in the implementation process of the master data management system.

The purpose of program design is to use computer programming language to realize every detail of system functions. The implementation of the master data management system is generally adopts the form of 'mature suite+professional implementation team+appropriate personalized custom developments'. Thus, the program design of the master data management system mainly conducts customization developments on the basis of software products.

In the process of the program design of the master data management system, it should mainly concern about the maintainability, reliability, understandability and efficiency of the program.

The methods of program design mainly adopt the methods of fast prototype and object-oriented design.

Most of the master data projects adopt the route of 'mature suite+professional implementation team+appropriate personalized custom developments'.

In order to enable the normal operation of the master data system according to the expected goals, and all functions are properly and fully utilized, the necessary trainings to users are a non-negligible work before the system goes on-line, which includes training plans and training contents.

Personnel training plan is to make corresponding training plans combined with the project's overall plan based on comprehensive and objective considerations on training needs, which includes training time, training site, training instructor, training object, training mode and training content.

Training content determine what kind of person should accept what kind of training. Trainings on master data management system architecture, system overview, key terms, menu and functions, common problems should be conducted on system operation personnels. For managers, trainings of master data architecture, cost-benefit analysis, support to the development of enterprises, the main report and interface, the main requirements should be conducted on them. Trainings on administrators include the master data management system architecture, system architecture, system document, typical user problems, supplier's support, problem solving strategies and training skills on operators and managers.

The main task of the data preparation stage is to complete the collection and cleaning of all existing master data in the enterprise according to the established data standards to form unified and standard master data, and import it into the master data system. Data preparation is the top priority in the implementation stage of the master data management system, and it is the necessary guarantee for enterprises to realize data governance goals and the successful launching of systems. Meanwhile, data preparation is also a time-consuming and complex work, which requires the joint participation of all business departments of the enterprise. The main content of this stage includes data preparation, plan formulation, data collection, data cleaning and importing. Data preparation plan is the guideline and work plan of data preparation. It is an important prerequisite for data preparation to deepen understandings on enterprise business, master data system and master data standards. The data preparation plan includes data scope, data preparation principles, data preparation responsible person and content allocation, data preparation personnel requirements, data collection format and the time plan of data preparation.

The data collection process shall be conducted in strict accordance with the data format in the data preparation plan, which not only meets the data standard requirements, but also takes the data structure of the original system into account, thus ensuring the integrity of the collected data, and lays a foundation for subsequent data cleaning. The purpose of data cleaning is to check the collected data according to the established data standards, and check the integrity, consistency and uniqueness of the collected data, and eliminate or correct the data with problems. After the completion of data collecting and cleaning, standard and high quality data are formed. The data importing stage is to import these data into the system efficiently and smoothly.

Information system implementation is a complex project. Information system software is only the physical carrier of management logic. Its essence is a change of enterprise management process and management means. Therefore, the implementation process of information system is actually the integration process of organization, personnel, process and system. After the completion of master data architecture design and implementation planning, the master data system construction will be transferred to the implementation stage. In the implementation stage, the physical model in the architecture design should be transformed into the master data management system in actual operation. It has very strict requirements on system expansibility, flexibility, reliability, maintainability and system quality. At this stage, work is complex and workload is large, for which multi-department cooperation is required. Only through reasonable arrangement and coordination can the system be implemented smoothly.

The main tasks implemented by the master data management system:
The main tasks implemented by the master data management system is the realization of the physical model proposed in the architecture design stage. It implements an actually operating master data management system according to the implementation plan, completes data transformation and system integration and improves the technical supports for the master data system so as to achieve the periodic goals of the master data system planning and improve the master data management capability of enterprises. The main tasks in the implementation stage include: data preparation, personnel training, program design, system testing, system operation and launching, system evaluation and project management.