MDM system implementation project. To whom and when? What for and how?

Image rawpixel.com Freepik

Valery Nemtsov, an expert on integration issues of the company "Constanta" told  about how to implement the project of MDM-system implementation and get as few bumps as possible. Consider businesses of different sizes, because for a small factory and agricultural holding MDM - it is not the same.

From my own experience:

I will help you to determine whether you need such a project or not;

I will tell you about the optimal technology of the project implementation.

First of all, I want to draw your attention to the fact that MDM-systems and projects for centralization of information systems are not a tribute to fashion, but the tools for quality changes in the work of employees in the IT landscape when the company is ripe for such changes. Otherwise you get a complication of the processes of working with the Master Data and do not feel the benefits of the project. Therefore, when starting a project, you should first define what results you want to achieve and whether you have chosen the right tool for this.

WHO? WHAT IS THE ROLE OF AN MDM-SYSTEM AT DIFFERENT STAGES OF COMPANY LIFE?

The master data management function is found in one way or another at different points in the development of any enterprise:

1. Mono-enterprise:

• One business line.

• One production site

• Up to 3 information systems

• Deliveries to 1-3 regions

When we are talking about a mono-enterprise, most often the role of an ERP system is taken by an ERP system that solves the tasks of customer service, production accounting, and bookkeeping. This is convenient from the point of view of the simplicity of organizing Master Data management in this case: the requisite structure is described in one place, access is given to a certain group of people to work with Master Data, the rest just consume the result, and the number of integration flows is minimized!

2. A group of companies that actively develops within 1-3 regions

• Several business lines

• Several production sites

• 3+ information systems

• Deliveries to 1-3 regions.

When we speak of a company that is actively developing, but is not yet a federal player in the market: it acquires other production sites, opens affiliated legal entities, which must operate in a unified information space, complicates the IT landscape in order to improve tools for automating separate business components: for example, ERP remains the basis for accounting, while customer service and production management are separate bases within the IT landscape. Here the prerequisites for the restructuring of processes within the company and the allocation of separate tools for working with Master Data are already beginning to take shape. Depending on further development plans, either 1C:MDM or 1C:ERPUH are chosen here. We will talk about the pros and cons of choosing systems for MDM in a separate article.

3 Holdin

• Several business segments

• Several production sites

• 10+ information systems

Widest geography of deliveries

When a business grows into a holding company, and both the task of data consolidation and the task of connecting new business units to the existing IT landscape are evident. Here, the task of managing the Master Data is most often already solved by one of the options from the second stage and there is already a simple scaling of existing processes.

WHEN IS IT RELEVANT TO DO AN MDM PROJECT?

• Many bases in the IT landscape (3+)

• A lot of errors in the Master Data

• Need for data consolidation for reporting

When there are all the prerequisites within a company to establish a master data management (MDM) function as a separate system, the need for an MDM system and a project to launch it arises. To accomplish this successfully, several task groups are identified, the successful resolution of which will yield a high-quality outcome:

Conceptual tasks

(performed by the IT director and architects):

1. Selecting a solution for the MDM system

2. Select an environment for data exchange

As part of the solution to these tasks, the top-level inputs to the upcoming project are usually formed. The selected solutions should fully satisfy the concept of the enterprise IT landscape development and will determine the further technical and economic framework of the project.

Data boundary tasks

(performed by architects and analysts):

1. define a list of data domains to be centralized

2. define the segments within each data domain and the rules to work with them

3. Design the data flows in the IT landscape, taking into account the emergence of the MDM system

4. Decide on the boundaries of data normalization in the startup process

5. Correctly select "baseline" data sources within the landscape for initial occupancy

6. Prepare restrictive measures for direct modification of centralized data in subscriber systems

This is the "very core" of the project, which defines the inner workings: which data to centralize, and which not; how to connect the new system to the IT landscape in a way to reduce the risks of connecting the existing information systems and transactional data exchange between them, what data you will take as input, how we will process and how we will work with them in recipient systems. Typically, these tasks are solved in the formation of requirements for the MDM-project and lay the groundwork for the future system, but even if these tasks are worked out competently, it will not be enough without the next group:

Organizational and technical tasks

(performed by architects, analysts, and the business customer):

1. Build a process for dealing with Master Data requests (and not only within the IT system, but also organizationally)

2. Choose a launch plan (gradual connection of recipients or a one-time "big bang")

3. Form a team of experts who will further engage in the maintenance of centralized data

It is the solution of the last group of tasks that is the most important element to change the processes of working with Master Data within the entire IT landscape, provides further order of the system and facilitates the adoption of new rules by business if all the solutions are found organically for your company.

WHAT DO WE GET OUT OF IT?

In terms of IT, the output is a solution to the conceptual issues within the landscape:

1. Centralized storage of Master Data

2. Control of data integrity and consistency

Unity of processes of work with Master Data

From the business point of view we obtain:

1. Reducing the cost of obtaining quality Master Data when connecting a new business unit

2. Reducing errors in accounting systems which can cause financial losses

3. Ability to more easily obtain consolidated reporting in unified terms