Deciphering MDM

This article is co-authored by

Yatin Jaisingh (Lead Consultant, Thoughtworks)

Kshitij Bhatia (Sr. Product Manager, VMware)

Introduction:

With the increasing use of complex vocabulary and acronyms in the industry and especially on social media and professional networking platforms, it has become increasingly difficult to understand the real meaning of what is actually written. We live in a world of complex jargons where an HR recruiter is referred to as a head-hunter or a talent acquisition specialist and a Brand marketing manager is referred to as a Brand Evangelist. Woaaaaah, aren’t these too fancy! ☺

In this article, we provide you with an insight into one of the most sought topics in the industry today — Master Data Management, in a way that becomes easy to consume, no fancy jargons and has a longer retention span in your mind.

Let’s get started:

Imagine any Organization of your choice. Each organization is becoming data-heavy with information about employees and their Payroll, companies’ products and services, customers, policies, physical assets like laptops, servers, accessories, vendor information, contact information, loyalty programs, etc., and many more.

Now, think from where does all this data come into existence. If you just consider one Data about Customer, it has several touch points across the data supply sources like Sales Team, Vendors, Online Portals, Marketing Touchpoints, etc.

But in what form does this customer data come? Some may come through online/physical forms, or sometimes as provided by customers and as understood by the sales/marketing teams. So, the complexity and reliability of the data is always a big question mark and solving this challenge is not easy.

For instance, You might have hundreds of stores selling thousands of products from different brands via multiple channels to millions of customers. This results in so much data to be managed at a stretch. This also results in making sure the data is set up right. This also results in making sure that the data changes are tracked and managed effectively.

Data is a resource just like other resources and we need to effectively manage data. Managing resources would mean right data, right time, right place, right volume, and right quality.

• As per the research by D&B (Dun & Bradstreet) data degrades 27% per year and 61% in 2 years and in the third year, your data will not be usable at all.
• As per the research of HBR (Harvard Business Review), the bad data cost 3 trillion dollars per year in the United States alone, just imagine its global impact.

We know, this seems very complex and the first thought that would come to mind is “Wouldn’t this be better if all the data for an Organization resides in one centralized place and all the businesses use the same data across their processes?

Trust us, this is not what happens on the ground. As an end customer, I interact with my profile data, products data, etc and all of these can change as well. All of these disparate data are managed by different departments in different systems.

Key Questions that business keeps looking

• Who are my most profitable customers?
• How many Customers do we serve?
• What have these customers bought from us in the past?
• What are the various customer touch points that help me find prospects?
• What is the buying potential of these customers?
• Can I serve these customers better based on their purchase history?
• Can I recommend other products to these customers?
• Which Products are best sold?

This seems to be a BIG problem but why does this really happen?? 🤷‍♂️

Let us highlight some of the key reasons:

• Data Transparency

First challenge that organizations have is that different departments maintain data in complete isolation from others and a lot of times they are also unaware of whether the data is existing in the ecosystem and available for consumption. Business needs the ability to acquire, manage & share accurate and updated information with them. With the volume, variety and fragmentation of data increasing at a massive rate, the necessity is to maintain quality, consistency and clarity across the data supply chain. And, most importantly making this data available to all the Business users for consumption.

• Trusted Data

Data influences our day-to-day decision making and we assume the information is accurate. The price we see on the web, the product that we order, the information that is being displayed. The expectation is simple — The information should be accurate.

For Organization to meet this expectation from the market, it’s not easy. There are systems, process that touch data, modify it, and sometimes decay it. To provide this accuracy, businesses need trusted data.

• Data Silos:

To circumvent the above problems, businesses started creating their own set of governed data which they trust and manage. So now in an organization we have multiple versions of data all being trusted and governed. Each dept again will suffer from the same problem but in a different way — Data silos, same data but multiple versions of trust.

Let’s put all of this in the context of Horror Stories to help you understand better. And remember, this is one of the many that keeps happening in the Organizations.

“A patient Ms. X mets with an accident. A person walking nearby in order to help the patient dials 911. A rescue team arrives and in order to identify the patient checks her wallet and finds a Hospital ID card. The team finds it to be a better choice if they take the victim to the same hospital assuming they would have some past records and medical history of the patient. They arrive at the XYZ hospital with above assumption and admit the patient.

The hospital accessed the past records of the patient and started with the treatment. However, the results were not as expected.

If we further dig deep on WHY, the root cause found was that the complete patient profile data was not fully updated and some of the information was also residing in silos on individual systems. Example — The patient had a past history of a drug allergy but was maintained in separate system which were not connected.

An incomplete patient profile may have damaging consequences, financial as well as reputational. The key issue here was the fragmented, inconsistent and unreliable information on the patient that may have resulted in an incomplete patient profile and further healthcare risks.”😢

So what do we do now? Do we have a solution? Yes, let us take you there now!

Master Data Management:

The first stumbling block you’ll face with MDM is when your peers ask you to explain what this three-letter acronym is. Let us list down different industry definitions for you.

“Master data management (MDM) is a technology-enabled discipline in which business and IT work together to ensure the uniformity, accuracy, stewardship, semantic consistency, and accountability of the enterprise’s official shared master data assets. Master data is the consistent and uniform set of identifiers and extended attributes that describes the core entities of the enterprise including customers, prospects, citizens, suppliers, sites, hierarchies, and a chart of accounts.” Definition of Master Data Management (MDM) — IT Glossary | Gartner

Does this make sense? Don’t worry, you are not alone. So let us make it easy for you.

Let us first explain to you what Master Data is.

Master Data, a term commonly used in very complex phrases by many, in simple language is the core data that is critical to any business.

Master Data or more popularly known as the Golden record or the Single Source of Truth (SSOT) is characterized by the data that is:

• De-duplicated- Non-redundant
• Centralized- Single Source of Truth for multiple departments, Line of businesses
• Consistent- Changes rarely. If changes do happen, propagated to consumers
• Trustworthy- Reliable, updated data.
• Enriched- Enhanced with relevant context

One of the best examples of explaining master data could be the Customer data which may comprise the name, email, phone and address (Address Line, City, State, Country, Zip etc.). When any organization has a large number of customers, this data may become highly complex. Any missed updates or errors may result in false insights about the customer (Sales potential, purchase history, relevancy etc.) and may lead to ineffective decision making.

For instance, for the same customer, “XYZ Technologies Limited”, the information that exists in 3 different transactional systems is different, and you end up assuming these 3 as different customers.

Master Data solves this problem, and ties everything together by systems and processes that can identify these 3 as same entity, consolidate, deduplicate and also enrich with additional information either from other internal systems or by using the 3rd part data enrichment service providers like Duns & Bradstreet (D&B). And, it is not a unidirectional flow of data, rather it’s a two way street, where the Master Data can also be consumed by these transactional systems.

In the “XYZ Technologies Limited” example, we just spoke about one category of Master Data which is “Customer”, but Master Data can be divided into 3 broad categories and involves multiple attribution of Master Data. The figure below will help you understand better.

Now you would have understood it right, MDM or the Master Data Management, is the practice of consolidating data from different sources and creating a unified data that is trustworthy, clean, deduplicated, updated, enriched and can be consumed by all the business functions across the organisation for effective and reliable decision making.

Master Data Management is creating one version or a golden record for each of the data entities, be it a person, organisation or a product. This involves creating one Single Source of Truth (SSOT) for the said entity that has been de-duplicated, cleansed and enriched. This version forms a reliable source of data that is managed and shared with the business users for reliable decision making and accurate reporting.

Forward Looking:

There was a time when different business functions like Sales, marketing, order management personnel would depend on the information stored in silos or personal storage devices or computers. Often the applications deployed across the organization for various executive decision making, data visualizations, and reporting were also driven by the fragmented information available in different sources leading to conflicting outcomes. But, as organizations are moving towards more data-driven decision-making, investments in developing Master Data at the core of the business are gaining popularity.

Now that you understand what master data and MDM is, In the next blog we would discuss some key questions about it and decipher it further for you. If you have some other questions, feel free to let us know and we would be glad to answer those for you. Also, we don’t shy away from feedback and would love to hear back from all of you.

*If you enjoyed reading my story, please do hit the Follow button on my profile!