Get in Touch
With Our Experts
+49 711 / 75886-600
Please Type A Message
Continue
Let us know where we can contact you
Please click to start verification
Back
Send message
Thank You For Your Message!
We will contact you as soon as possible.
Close Window

The Program Manager’s Guide to the MDM Galaxy

03 April 2019
I’m convinced that, for most companies, the biggest challenges of digitisation do not lie in the front-end or the user experience. The problem lies in the existing data; that is why Master Data Management (MDM) is such a big issue. Many companies are already beginning to think beyond individual data domains and to consider the issue as a whole. They understand that it is a matter of implementing Master Data Management for all relevant entities. Only a comprehensive Master Data Management, combined with a sustainable investment in adequate data quality and data governance projects, will provide a long-term and strategic platform for the digitisation process. In many companies, existing PIM and CRM projects are merged into or replaced by MDM projects.

Other companies remain more or less inactive because they don't really know how and where to start. I have received many comments about my article "Why is everyone suddenly talking about PIM?". Time and again I was asked how you can initialise and run MDM programs successfully. As a consultant to European enterprise customers, I am often asked by customers what is the best way to start a master data management project. "What should I consider and how should I proceed? Please be very pragmatic and give me some best practices to follow. After all, you and your colleagues are the experts...” This is why, in this article, I would like to explain the key principles of our approach.

The biggest mistakes are made at the beginning. Or, to put it another way: laying the (right) foundations is the key to success.

Prerequisite 1: Qualified employees
May I introduce…?  The Data Ninjas

Who are the “data people” in today's businesses? There used to be a clear line between IT and business. There were people in different departments who defined requirements and IT people who implemented these requirements in IT projects. The problem, however, was that the two sides often failed to understand each other properly and, moreover, the customer was often forgotten in the process.

In recent years, new roles have emerged in order to build a bridge between the various departments as well as to external partners such as suppliers and customers. Because of the interdisciplinary nature of their work, "data stewards", "business analysts", "data architects" and "data owners" often cannot be clearly assigned to specific departments. Gartner analysts predict that these roles are likely to increase as the number of data projects continues to grow. “Data stewards" are, for example, employees whose main task is to monitor the quality of information and compliance with information processes within the company and to optimise the workflows. According to Gartner, 32% of companies today have "data stewards". This figure will rise to 71% by 2020.

Businesses invest in data projects and employees who consider data to be a real asset. In most cases, these “Data Ninjas” cannot be clearly assigned to a particular department. They want and need to address data issues across processes and organisations.

Prerequisite 2: Qualified Data

Companies struggle every day with missing, incomplete and incorrect data. Employees often spend more time preparing the data for a particular task than they do evaluating the data itself. It's not as if companies wouldn't collect any data. In fact, they gather mounds and mounds of data. The point is that this data is often inadequate for the task at hand.

For instance, take a simple product. Some data fields that you need for a product are obvious. A product will have a specific number and other identification codes as well as units of measure and, of course, a category.

However, if we think of the product in a larger context, we will have to add many more fields to our list.

Our example data model needs to be expanded considerably if you take into account the demands of the market, the customer or the manufacturing process: What information do the sales and marketing colleagues need to sell the product? Which data is generated during the sales process that may be valuable for later evaluations?

We gain a new perspective when we consider our product from the point of view of the systems involved: master data is not stored in one single system but in numerous different systems. Is this information consistent across systems and processes? Who is responsible for communicating changes and updates to the data models to other colleagues

In our example, we only looked at one single product. If we now consider this product in the context of other master data domains, we will get a more accurate picture of the dependencies.  If marketing wants to analyse which customer has bought which product through which sales channel, they already need to consider three different entities. In many companies, this data would be difficult to come by and, if so, without any guarantee as to its consistency, accuracy and integrity. That's precisely what MDM is about. MDM programmes are often born out of the need for high-quality and analysable data - a prerequisite for several internal and external applications.

Prerequisite 3: Established standards

At parsionate, we focus on mid-sized and large businesses that are used to working with leading global research and advisory firms such as Forrester and Gartner. That's why we decided early on not to reinvent the wheel and to lull our clients to sleep with countless new terms and methods - as many other consultants do. Our process model in MDM consulting is based on established standards, which we have analysed and refined. We cooperate with Gartner because we like their expertise and logical approach. Years ago, Gartner colleagues developed a model called "The Seven Building Blocks of MDM".

Our projects are based on this model, which is why I would like to outline it briefly.

1. Vision

Before developing an MDM strategy with a customer, we need a clear common vision. First, we talk to the executive management. What are its business goals? Are there any high-priority business goals that require specific data? This first phase, in particular, will provide the strategic backbone for our MDM programmes. Thus, for example, a forthcoming internationalisation of the sales organisation can be used as a starting point to harmonise customer data. In other, more “sensitive” industries, regulatory requirements may prevail.
In addition to the actual business requirements, the vision also provides us with a comprehensive overview of the stakeholders and their expectations. It is also important to understand what is the role of data within a company. Do people working in this company already rely on data to make decisions? How do IT and the other business departments work together? How much experience do the individual departments have?
All results will be translated into a manifesto, that should be as simple as possible and for which we seek the support of top management.

2. Strategy

For an MDM programme, we need to identify clear objectives, i.e. business objectives that the company is trying to accomplish, from which we can derive data and IT goals. How can the MDM programme help to achieve these business objectives? What is the impact of each phase on the company’s strategy?
After defining the objectives, we will define the scope of each phase. Which applications, processes, teams and quality standards will be affected? By the way, you will need to re-evaluate these items regularly. It is imperative to define when and how often a specific team or person will verify whether these strategic goals have been achieved or not.

Our proposal for an MDM programme will be based on these objectives and the outlined scope. For each phase of the programme, we will define, together with the management, the workflows, milestones, results, resources and costs.

3. Metrics

Measurability is essential! For each stage of our MDM programme, we need to identify metrics to assess the impact of the measures on business performance. KPIs are defined for all business processes. For example: Whereas in the “vision” we may have stated that we want to be able to create product variants more flexibly and in the “strategy” we may have defined that a new, cross-company product creation process should be set up for this purpose, at this stage we need to determine suitable metrics to measure the effectiveness of this new process. This could be, for example, the time needed for a product variant to be created from the initial idea to the actual availability in the online shop. We measure the actual processing time and set our targets.
There are different types of metrics. Typical examples are: performance (duration, effort, time), financial aspects (costs, risks) and master data itself (quality, availability, completeness, accuracy).

By the way, with some creativity, it is even possible to measure "soft" factors such as the system's usability, for example through regular user surveys. Assessments should be carried out regularly. At this stage, it is important to define who will be responsible for reviewing and documenting the KPIs in the subsequent projects and also later on.

4. Information Governance

A major goal of any MDM programme is to make data available to employees and other people involved in the process. Information governance is a holistic approach to managing information. Information governance aims to make information available to those who actually need it in order to process it. At the same time, it streamlines management, reduces storage costs, and ensures compliance.

To make MDM governance a reality, we start by establishing a company-wide interdisciplinary task force which consists of representatives of the main data streams in a company. For this task force, we need a reporting system and well-defined decision-making processes. Who is responsible for ensuring compliance with the standards and laws? Whose task is it to facilitate the dialogue between the different business areas?

Compliance is an increasingly important issue. In addition to general standards and regulations that all companies must comply with, such as the GDPR, many industry standards and regulations must also be adhered to. Thus, organisations will need to clearly define who is accountable and responsible for ensuring that they meet their compliance obligations. Not only during the project but also afterwards, as standards continuously evolve.

Failure to comply with information standards is not the only risk; the provision of data in itself poses risks as well. Therefore, we need to carry out a risk assessment. Who is allowed to access, see, forward and use which information? How critical are which violations? Who is responsible for ensuring that there are no violations? How is the data protection officer involved in the MDM programme?

For many companies, information governance also involves assessing the impact of the project on their investment plans, because it is at this stage at the latest that the costs (including the cost of risks) and the positive effects of higher-quality data should be determined.

5. People

Since MDM is not just about introducing new software, this phase of the project deals with the organisation and the individual roles. With the same team as in the governance phase, we develop a role matrix ("RASCI") to chart who is responsible for which aspect and who can and must take which decision. We also define who will be responsible for the overall MDM programme management and who will be expected to convey the importance of MDM and data to the different departments. This team will coordinate, communicate, align, manage and control all projects within the MDM programme.

Other roles include programme managers for individual data domains ("Domain Managers"), data quality managers ("Information Stewards") and the members of the individual project teams.

6. Process

In this phase, the information life cycle within the company will be defined, i.e. for each data domain, a team will specify all processes involved - from creation to storage/archiving or deletion. In most MDM programmes, businesses can no longer ignore the need for fundamental changes in the way they work. For example, the question of where product data is created is extremely important for many different processes in a company. If, for example, a retailer decides that, in the future, his data management system will be based on the product data provided by his suppliers, the processes by which product data is to be provided by the suppliers must be completely redesigned. Not only the suppliers but also the employees will have to adapt to a totally different workflow. This will require people to be trained and new processes to be integrated into the organisation.

Another example related to customer data: The GDPR states that customers have the right to request the deletion or removal of personal data. In order to meet this requirement, you need to know where information about customers is stored, who can process this information or has access to it, which systems are involved, how you can delete or block access to this data in these systems, etc. This requires clear responsibilities and well-documented processes.

Furthermore, you should specify the required data quality across all information processes. For each step, you should define mandatory / optional / target requirements. It is also necessary to determine who will be responsible for assessing and verifying these measures.

7. Infrastructure

Now - and only now! - will we start implementing the systems. The good news is: We have already defined and designed most of the matters that are often forgotten or overlooked in IT projects - matters that can have dire consequences if they are not taken into account.

The following are major issues in MDM projects: data models, hierarchies (e.g. classification standards, product group structures, inheritance, product-article relations), data quality rules, user interfaces (for employees, but also for external partners such as suppliers and customers), interfaces to peripheral systems, workflows and, of course, the technical infrastructure (system architecture, security, hosting, etc.).

Furthermore, it is necessary to ensure smooth operation and maintenance of the systems. Not only from a technical point of view, in terms of application management, but also organisationally (who is responsible for updates, process improvements, further development, etc.?).

We found that training courses that are conducted in parallel with the project are a good way of involving employees at an early stage.

As you can see, with Gartner's "Seven Building Blocks of MDM" we have an excellent basis for addressing all relevant topics in a data project - from the idea and the dialogue with top management to IT topics and systems.

It's the big moment of the Data Ninjas - to put your strategy into practice and achieve results.

Back in the days of yore, when life was less complicated, IT projects were carried out by the IT department, and business departments were only involved in the requirements phase. You cannot manage data projects like that anymore. I can safely say that because my colleagues and I have seen enough failed digitisation initiatives in recent years. Nowadays, businesses are looking for MDM programmes that focus on data as a value and assess their benefits from the customers’ point of view. Will the end customer benefit from it? Can we offer him any added value that would simplify his life? How does this added value set us apart from our competitors?

If we see data projects as the key to digitisation, then Data Ninjas are the analysts, communicators, business people, IT specialists and customer experts. They do not think in terms of departments and "IT vs business“ but focus on customer benefit. Our customers have launched some excellent Master Data Management initiatives, and we noticed that there is a new generation of people who are the key to ensuring that these initiatives contribute to the long-term success and growth of these businesses.

This article was originally published on LinkedIn

Ihr Webbrowser ist veraltet

Aktualisieren Sie Ihren Browser damit diese Webseite richtig dargestellt werden kann.

Zur Infoseite browser-update.org