How do you create a career in MDM?
In this interview, Prashanta Chandramohan, creator of the “MDM – Geek’s Point of View” blog shares his insights on creating a successful career in the MDM sector.
Data Quality Pro: How did your career evolve into an MDM focus?
Prashanta: I started my career as an MDM consultant. I worked with one of India’s largest fast moving consumer goods company in setting up a product information management system. When we talk about Indian organization, its all about volume. We consolidated huge amount of product data from around 20 different source systems and provided one single collaborative PIM system. This was a colossal effort and a great learning experience for me.
As I started reading and understanding more about MDM, I got more devoted to this technology. I worked with many customers in Asia for 4 years. Helped them in establishing their master data hub and achieve ‘Single View of Customer’ data. I have spent ample amount of time in training customers and partners on MDM technology and aspects of data quality and data governance which are crucial to success of MDM vision. Customers in Asia can be very demanding and expect good return for every penny spent. It’s also a challenging work environment as you try to fit in products which are primarily designed to target European/North American customers. These projects have been successful and have kept a challenge seeker like me on toes.
In MDM arena, things move very quickly. Technology our customers use today to run MDM and data quality projects is changing rapidly. It’s hard to keep up with this change. But looking at the growth and the boom in this industry, I am happy I am here. I have thoroughly enjoyed this journey so far and would like to keep up with its pace.
Data Quality Pro: What are the core skills that you draw on to consult on MDM projects?
Prashanta: There was a time when we use to act as a resource contributing to specific area of MDM program. Like I have done a developer role several times, designed MDM data model in some, and worked as integration specialist in others. But, today’s MDM consultants have to go beyond their individual roles. Because MDM projects now are not only about setting up master data management system, but also about making sure the data is consolidated, profiled, cleaned and standardized before moving to MDM repository. We also need to apply appropriate data matching rules so that the data duplication is reduced considerably.
Having worked in multiple MDM products, integration technologies, ETL (Extract Transform Load) tools and data standardization, matching algorithms I have gained key skills required to execute a master data management program. I understand customer’s pain points of handling master data and can help guide MDM projects from initial inception till final delivery. Having had a chance to work with customers from different industry verticals like retail, telecommunication, insurance, financial and healthcare, I have gained ability to model MDM system to cater to diverse customer requirements.
I have worked in many implementations; seen common pitfalls which customers face and can give expert advice on planning, solution design and execution of data quality and master data management programs.
Data Quality Pro: In terms of technology, where do you see this aspect of the industry evolving? Where is the scope for greater innovation?
Prashanta: MDM has evolved to a great extent over last 3-4 years as more and more organizations are realizing the value this technology can bring to them. The good news for customers is, there are many vendors in market today to choose from. To stay in market and deliver, every vendor is creating competitive products, solutions and packages, which can help these organizations suffering from data issues. There is lot of innovation as we are seeing continuous improvement in every aspect of master data management.
Recently, there has been lot of talks around social MDM which I personally like. I see it to be the next big thing as it allows organizations to drastically enhance the customer insight. There is enormous data about customers out there which is:
- Near real time
- Puts customers in the driver’s seat and let organizations know exactly what they want
- The insights that can be extracted out of this data can help in more accurate cross sell/up sell
Now the challenge is to take this “big data” consisting of unstructured, fast changing textual information from social network into “real” usable master data. Gone will be the days when organizations use to create weekly or monthly reports. This change needs lot of innovation in the way our products are designed today.
Data Quality Pro: Let’s talk about data governance, I still hear horror stories where companies view MDM solely as a technical project but from your writing I know you don’t hold the same view. What are the key components of a data governance strategy that you would recommend?
Prashanta: This has been discomfort ever since I started this expedition. The fact is, most of the customers don’t think about data quality and data governance till mid way into the project. Majority of them seem to think, buying a well-established, expensive MDM product is going to solve their master data problem.
I would like to use the term ‘MDM Readiness’. Meaning, every organization want to manage their master data efficiently, but they have to be MDM ready to embark this journey. The first step to become MDM ready is to establish a data governance practice. This practice should be headed by a data governance body consisting of all the major stakeholders of the organization who can take and action decision on data. The decisions of this body should boil down to following action items.
- Data profiling
- Data consolidation from sources (Giving priority to most trusted source)
- Transformation of Data (To adhere to quality standards which can be used across the organization)
- Standardization of key master entities. (Address, Name in case of Customer data, Product description in Product data etc.)
- Continued focus on governance
Again, before the information is fed into MDM, data quality improvement has to be given utmost importance.
From my experience, I have not seen this happening more often than not. Governance and data quality are not pre-thought and as we start consolidating data to load to MDM, data quality issues start popping out causing significant delays to projects.
I have seen project plans getting extended 2x to 3x times because the data quality realization happens much later in the MDM cycle. These organizations end up doing a lot of clean-up activity as a result, which could have been avoided if only the standard MDM practice is followed in first place.
Data Quality Pro: In your experience of completing multiple MDM projects, how does MDM affect the culture and attitude of the business towards data?
Prashanta: Well, no one would like to see a drastic change in the way they look at their data. Although, initial MDM phases intimidate many customer folks, there has been a great tendency to understand the value-add MDM does to the organization. The senior executives who are usually the only ones aware of the real value this technology brings in, and strive for more efficient data quality and governance, play crucial role in changing the attitude of data users.
One of the important factors is to gain confidence of the users of the system. We do this during and after the implementation by conducting awareness sessions for these users. These sessions explain advantages MDM brings in to the organization.
As the customers start seeing improved quality of the data they are dealing on a day-to-day basis, they get adapted to system fast, and start treating it as a valuable asset. While most of the time is spent on addressing data management shortcomings prior to the MDM implementation, customers talk about how confident they feel about data integrity and its standard representation after MDM is functional.
Data Quality Pro: Finally, what advice would you give someone who recognises the opportunities in MDM and wish to move more into this discipline?
Prashanta: We all know MDM is not just about technology. It involves business knowledge, understanding stakeholder’s needs and working with a diverse set of teams involving data quality and governance bodies.
So, my advice is to look at the big picture and not just MDM. Learn about data and tools which help you in improving quality of data. Data de-duplication and standardization are two very important aspects that play key role in any MDM implementation. There are many technologies available today for cleanse, standardize, match and de-duplicate data. Read and try your hands on in these tools to know-how. This knowledge helps a lot when you are implementing MDM.
Image credits: cc Stephen Wolfe
Prashanta Chan is a Senior Technical Specialist on Infosphere MDM Server and Initiate Master Data Service at IBM. He hosts a popular blog: MDM – A Geek’s Point of View and has spent the last six years in different parts of the world providing MDM solutions for clients across different sectors like financial, insurance, telecom and retail.
He is currently helping a large North American customer with an MDM initiative.