10 Best Practices to Support Your MDM Strategy

The old saying “easier said than done” applies when we discuss best practice in master data management (MDM). Best practice, by definition, is the practice that brings the best results. Therefore, MDM best practices are the ones that give you the greatest possible benefits. See? We told you that part was easy.  

Achieving best practice MDM requires a lot more work. You need the right tools, the right consultants, and the right plan for your organization. Your consultants should be able to collaborate with you and offer significant support in making sure the plan is right for your business. This includes the foundation of your MDM, the data cleansing and governance processes, and sustainment of the data. If your consultants can’t help you through these stages, then it’s probably time to look for new consultants.  

Your organization isn’t exactly like any other, and the processes that constitute “best practice” will probably be somewhat different in detail from the best practices at another organization. However, over the years we’ve assembled the following list of best practices that every business should follow. These are general rules that will require some fine-tuning to truly make them “best practice,” but your MDM strategy will suffer if it’s not grounded in these principles.  

Best Practice MDM Begins From the Ground-up

Best practices in master data start with defining the overall goals and the vision for your organization’s MDM plans. Think of it like building a house. The first step isn’t putting the walls up or the roof on. It’s not even digging and pouring the foundations. Building a house starts by drawing up precise plans that layout the exact dimensions of the building, the patch of land it will occupy, the materials used in construction and so on. Construction can only begin once you have all of that in place. Digging a foundation before you’ve decided on the building’s footprint will cost you more time and money in the end than having a solid plan.  

Building a house this way is obviously foolish, but that’s exactly the approach many organizations take when it comes to master data. They put a lot of effort into data gathering and cleansing, build out elaborate data governance processes, and purchase solutions without firm goals and vision in place.  

Your MDM strategy must start with a clear vision and goals to be effective. You’re flying blind if you don’t have these, no matter how much money your organization is willing to throw at the problem.  

You must define your vision and goals before you start on the foundations of your MDM strategy. We really can’t overestimate how important it is to have these in place. This is a true MDM best practice, regardless of the organization.

1. Define the Overall Goals and Vision of Your MDM Strategy

First and foremost, which business problems are you trying to solve? Be wary of statements that are too general, like “a lack of consistent and accurate master data.” This is the challenge every MDM project is trying to overcome!  

Be as specific as possible. If your goal is to reduce inventory, then the strategy you draft will look very different from one intended to assign unique IDs for medical devices. You need to drill down to define the roots of why you need MDM. This will help you define the overall vision for the strategy. Once you have a vision and goals, you can define the mission, milestones, and the business value for your MDM project. Doing so provides you with the following benefits:  

  • Justifies the budget
  • Motivates the organization
  • Provides focus
  • Measures progress

2. Acquire Executive Sponsorship With a Focus on ROI

No matter the solution you use, MDM must be viewed as more than just software. Master data is complex and touches many parts of the business. Gathering and cleansing the data are only the first steps. The real key to extracting value from your master data lies in sustaining that data over time. This long-term commitment must be supported by your organization’s culture, or it will eventually fail. That’s why MDM strategy must have executive sponsorship to navigate culture change and organizational politics.  

More to the point, you will need to engage with leaders from many departments to ensure your MDM strategy is successful. Having an executive sponsor encourages cross-departmental alignment.

You usually point to the return on investment to secure executive support. Your business case must identify the challenges that will be overcome when the MDM project succeeds, and the contributions this will make to overall profitability.  

Executives want to see a positive ROI for any undertaking but they’re hardly the only ones. You’ll find it easier to get department leaders on your side if you can show how much they’ll save or how many news sales they’ll gain because of your MDM strategy. After all, as the old saying goes, money talks.  

Departmental leaders should also be involved in laying out the data governance rules, so it’s best to get them involved in the MDM process as early as possible.  

3. Build a Strong Foundation for MDM

Continuing with our construction analogy, our vision and goals provide us with our blueprints in the form of our mission and defined business values. Implementing our chosen master data solution is essentially like putting on the walls and roof. That means we must lay the foundations first or the walls will soon crumble and fall. The next five items in our list define these foundational components.

4. Establish Strong Data Governance Standards

In short, governance defines who “owns” the data and the business rules that are followed with it. In other words, how is your data governed? Strong data governance requires equally strong master data control polices and organizational accountability. Your governance strategy should also lay out which metrics will be used to determine the success of your MDM strategy.

Agreeing on the organizational governance model upfront makes any issues that come up easier to resolve. It will also provide clear direction for analysts when they’re configuring new validation and taxonomy solutions.  

5. Map Out Your Data Processes

These define the procedures that must followed in the creation of master data as a single source of truth. We advise you to put the focus on maximum automation, while building audit functions and traceability into the process. Your data processes should support data quality initiatives such as classification and cleansing, as well as procedures to capture metrics regarding the quality of the data.  

6. Monitor Governance With Metrics

A strong foundation for MDM must include objective metrics that define the performance of your governance mechanisms. The precise nature of these metrics will vary from organization to organization, but we’ve summarized some of the most common below:  

  • Audit efficiency
  • Non-compliance
  • Audit effort
  • Defect prevention ratio

7. Adopt the Right Architecture and Standards

You must define how the data is organized, illustrating the connections between entities and business objects (architecture) as well as determining a complete taxonomy (standards).  

One of the advantages of Master Data-as-a-Service (MDaaS) from Prometheus Group is that the taxonomy was rigorously developed over many years to cover all situations and all industries. It leverages governed standards such as ISO 14224 and others, while the solution ensures that any new records are created following your organizational standards. Our team of solution experts and data scientists are constantly updating and revising the taxonomy to ensure it stays in sync with your needs.  

8. Select the Best Tools for the Job

Whatever tool you select must be suited for the job you’re giving it. Forcing a tool designed for one purpose to perform another doesn’t always fail immediately. Anyone who has ever used a crescent wrench to hammer in nail knows that. It works, but not as well as a hammer.  

The tool you select must be able to deliver on the business goals you have defined and the processes you have set out. It should be scalable, to account for future growth and changes, as well as offer process automation and integrated governance.  

9. Start Small, but With an Eye on the Future

With the foundation in place, you can finally start on the walls and roofs. Leaving the analogy aside for the moment, this means it’s finally time to start really working with your data.  

Gathering, deduplicating, and cleansing your data is a major undertaking. It’s easiest to start small and build up from there. Ideally, you should target one of your datasets that is relatively small and stable.  

Starting small like this has several advantages. First, the work will go faster than it would if you were applying the process to all your organization’s master data. This means you will be able to show a quicker return and demonstrate good results as soon as possible. Doing so makes it much easier to earn funding for the rest of your strategy.  

Second, figuring out where your weak points are as soon as possible will provide much better results in the long run. Working with a smaller dataset will expose these weak points before they can impact the rest of the project, allowing your team and consultants to make whatever changes are necessary to ensure they do not arise again.  

10. Treat MDM as a Strategy, Not a Project

To be successful, you must treat MDM as a strategy rather than a project. A project has a defined end date. Once the goal is achieved, the project can be considered complete. MDM shouldn’t be viewed this way.  

Clean, accurate master data can be your organization’s single source of truth. That’s not something you should want to give up. However, unless significant focus is put on sustaining the data, then your organization will default to giving up that single source of truth. Without sufficient sustainment of the data, before too long your master data will be in the same state as it was when you began. You will get benefits from it during the time your master data is dependable, but it’s much better to be able to rely on it for the future as well.  

Our MDaaS solution ensures that your data is constantly synced and updated, thus handling the sustainment piece for you, so your master data continues to provide the business value and ROI you originally envisioned.  

Click here for more information on how MDaaS can help you ensure master data quality from capture through to sustainment.