Why Data Modeling is vital in Master Data Management?
Master Data is the key piece of all your business operations. Customers, Products, Employees, and Factories run through all your processes, tying together all the sales and purchases and marketing efforts your organization is working on. This critical role of Master Data has, naturally, garnered a lot of attention from data professionals for many years already, giving birth to the area of Master Data Management, or MDM for short. Even though individual implementations of MDM differ greatly in terms of systems and processes utilized, there are some things that you should always consider. The very first step of any successful MDM initiative is the one where Ellie can help you.
The usual problem: what is my Master Data, really?
It’s easy to say what are your core Master Data entities: it’s usually Customers and Products! All kinds of Customer MDM solutions and Product Information Management systems exist to give you tools for managing that part of your Master Data.
The problem, however, is way more complex than you might think. Often it appears that there is no shared consensus of what a Customer or a Product actually is!
An example from the Customer area might be something like this. There are Customers that your Finance department deals with; they pay you money for their purchases. There are, also, Customers who receive your deliveries. Furthermore, there are Customers that are the subjects of your Sales department’s efforts. Perhaps you have Customers who are the end-users of your products, whom your Technical Support deals with, or there are Customers who have signed up for a free subscription of a product but who are not really paying anything.
Everyone involved in any of these situations in your organization calls their counterparts Customers. So, then, what does your Customer Master Data contain? All of them? Maybe some part of them? If only a part, then why not the others – after all, aren’t they Customers too?
To model is to understand
Armed with Ellie and good Data Modeling techniques, you can start approaching and documenting what is actually going on with your Master Data. What does the business reality look like? What are the core entities that you have data about, and how are they related to each other?
Modeling not only helps the modeler understand things better, but it often also reveals misunderstandings and confusion among the actual business – and then helps to solve all that! For example, you might realize that there are various subtypes of Customers: some are Organizational Customers and some are Persons, and they might be managed completely differently. Or perhaps you come to understand that your Customers have different roles in which they act towards your various processes: some act as Payers, some act as Users, and yet others might be both at the same time!
Building on the shared understanding
A well-made conceptual model explaining these different types and roles of Customers and how they relate to your Contracts and Products and Deliveries is extremely valuable for your entire organization. It will not only help you make decisions about Master Data Management (“we will put those and those in our MDM system, but not those!”), but also it acts as basic documentation of your core business data to any future projects. Maybe next year you need to start setting up a new ERP: at that point, you better know what kinds of roles your Customers can have and how your system needs to be able to handle them.
Consider also the very important but not-so-exciting topic of requirements analysis: if a corporate VP comes up and requests a report with the top 200 Customers ranked by their size, how do you interpret that? And perhaps even more importantly, what did the VP actually mean? What Customers are we actually talking about? If you have a nice-looking model with all these terms defined, you can simply point a finger at it and say “those!”
For us at Ellie, this is all about understanding what the data means and sharing that understanding with everyone. This is also the first step your MDM initiatives should always take: ensure shared understanding.