A Guide to Master Data Management

Over the recent months, I have been engaged with the boutique Master Data Management (MDM) consultancy parsionate on some visionary projects in the multi-domain MDM area.

The overall approach at parsionate is that MDM is much more than just an IT issue. It is a strategic necessity. This is based on an experience that I share, which is that if you treat MDM as an isolated technological problem, you will inevitably fail!

Multi-domain MDM

MDM implementations today are increasingly becoming enterprise wide and are thus also multi-domain meaning that they cover business partner domains as customer and supplier/vendor, the product domain and a longer tail of other core business entities that matters within the given business model.

The primary goal of a multi-domain MDM implementation is to unify and consolidate heterogeneous data silos across the enterprise. The more source systems are integrated, the more information will be available and the more your enterprise will benefit from a 360° view of its data.

To achieve the desired goal for your multi-domain MDM program, you need to have a clear vision and a long-term strategic roadmap that shows how the various MDM implementation stages fit together with other related initiatives within your enterprise and where the Return of Investment (ROI) is expected and how it can be measured.

Seven Phases of Forming the Roadmap

In the approach at parsionate there are seven phases to go through when forming the roadmap and launching an MDM program.

Phase 1: Identify business needs

Before embarking on an MDM program, consider what data is most important to your business and where you can create the most value by putting this data under the MDM umbrella.

The main rationale is that through MDM organizations can control the way their most important data is viewed and shared in a more efficient, traceable way and thus improve their performance and productivity. An effective MDM implementation helps you streamline your workflows. It breaks down data silos that prevent data from being reused and shared across the enterprise

Phase 2: Set up a data committee

Establishing a data committee (with any equivalent name for a data focussed body) is perhaps the most frequently mentioned aspect of an MDM strategy. This team would usually consist of many different stakeholders who are in a position to enforce the roadmap and the underlaying activities.

This body must be able to convey the importance of data across the enterprise at all organizational levels. The main concern is to make data a top priority in the enterprise.

Phase 3: Set up a data governance program before defining the MDM roadmap

Many organizations shy away from data governance programs that consulting firms suggest because they seem too complex and costly.

The bitter truth is though that if you fail to implement data governance or embed it strategically into the way the organization works, you will inevitably have to address it at a later stage – in a more time-consuming and costly process.

Phase 4: Set clear goals to ingrain the MDM vision in the organization’s culture

It is difficult to promote an MDM program without a clear vision and objectives. At the executive level, the program could be misunderstood as a technological issue. Sometimes decision-makers struggle to understand the value an MDM program will generate. In this case, they will either not fund it or, if it does go ahead, consider it a failure because the result does not meet their expectations.

It is crucial to involve all relevant stakeholders in the roadmap development process at an early stage and engage with them to understand their expectations and requirements. Only then can you ensure that the core elements of a successful MDM program are aligned with the needs of your entire organization.

Phase 5: Choose a step-by-step approach and rapidly implement sub-projects

The most effective way to implement an MDM program is to start with a few key sources and provide meaningful information in a very short time. It is always difficult to implement everything at once (multiple sources, several integration points, and com­plex data) with a „big bang“ or build a data hub without a specific goal in mind.

If a pilot project quickly realizes a series of short-term bene­fits, users and business leaders will appreciate the value of the MDM program. As soon as it becomes clear that the initial project is successful, you can promote a wider roadmap that shows how the next steps will be carried out in line with the strategic goals of your organization. With this iterative approach, the long-term benefits will become clearer.

Phase 6: The mammoth task: Adopt a data-driven mindset

Building a data-driven corporate culture may be considered the supreme challenge of any MDM program. Data culture goes far beyond a simple corporate strategy that uses data for business processes. Rather, it is a mindset that encourages employees to appreciate the tremendous added value of high-quality data.

Many organizations believe they can simply buy new tools to drive digital transformation. That is an illusion. 

Phase 7: Integrate technology

This illusion does however not mean that that MDM technology is not important.

Hardly any other IT system affects as many different departments and data domains in a company and is, implemented the right way, as powerful as an MDM solution. The extreme importance of this type of software within the entire corporate IT infrastructure means that you need to select a system very carefully and strategically.

The Full Guide If you want to read the full guide to MDM mapping out the high road to a successful implementation you can download it from this parsionate site: Master Data Management

What’s in a Data Governance Framework?

When you are going to implement data governance one key prerequisite is to work with a framework that outlines the key components of the implementation and ongoing program.

There are many frameworks available. A few are public while most are legacy frameworks provided by consultancy companies.

Anyway, the seven main components that you will (or should) see in a data governance framework are these:

  • Vision and mission: Formalizing a statement of the desired outcome, the business objectives to be reached and the scope covered.
  • Organization: Outlaying how the implementation and the continuing core team is to be organized, their mandate and job descriptions as well as outlaying the forums needed for business engagement.
  • Roles and responsibilities: Assigning the wider roles involved across the business often set in a RACI matrix with responsible, accountable, to be consulted and to be informed roles for data domains and the critical data elements within.
  • Business Glossary: Creation and maintenance of a list of business terms and their definitions that must be used to ensure the same vocabulary are used enterprise-wide when operating with and analyzing data.
  • Data Policies and Data Standards: Documentation of the overarching data policies enterprise-wide and for each data domain and the standards for the critical data elements within.
  • Data Quality Measurement: Identification of the key data quality indicators that support general key performance indicators in the business and the desired goals for these.
  • Data Innovation Roadmap: Forecasting the future need of new data elements and relationships to be managed to support key business drivers as for example digitalization and globalization.

Other common components in and around a data governance framework are the funding/business case, data management maturity assessment, escalation procedures and other processes.

What else have you seen or should be seen in a data governance framework?   

How the Covid-19 Outbreak Can Change Data Management

From sitting at home these are my thoughts about how data management can be changed due to the current outbreak of the Covid-19 (Corona) virus and the longer-term behaviour impact after the pandemic hopefully will be over.

Ecommerce Will Grow Faster

Both households and organizations are buying more online and this trend is increasing due to the urge of keeping a distance between humans. The data management discipline that underpins well executed ecommerce is Product Information Management (PIM). We will see more organizations implementing PIM solutions and we must see more effective and less time-consuming ways of implementing PIM solutions.

Data Governance Should Mature Faster

The data governance discipline has until now been quite immature and data governance activities have been characterized by an endless row of offline meetings. As data governance is an imperative in PIM and any other data management quest, we must shape data governance frameworks that are more ready to use, and we must have online learning resources available for both professionals and participating knowledge workers with various roles.

Data Sharing Could Develop Faster

People, organizations and countries initially act in a selfish manner during a crisis, but we must realize that collaboration including data sharing is the only way forward. Hopefully we will see more widespread data sharing enterprise wide as this will ease remote working. Also, we could see increasing interenterprise (business ecosystem wide) data sharing which in particular will ease PIM implementations through automated Product Data Syndication (PDS).

Covid Data Management

The People Behind the MDM / PIM Tools

Over at the sister site, The Disruptive MDM / PIM List, there are some blog posts that are interviews with some of the people behind some of the most successful Master Data Management (MDM) and Product Information Management (PIM) tools.

People behind MDM tools

CEO & Founder Upen Varanasi of Riversand Technologies provided some insights about Riversand’s vision of the future and how the bold decisions he had made several years ago led to the company’s own transformational journey and a new MDM solution. Read more in the post Cloud multi-domain MDM as the foundation for Digital Transformation.

In a recent interview FX Nicolas, VP of Products at Semarchy, tells about his MDM journey and explains how the Semarchy Intelligent Data Hub™:

  • Extends the scope of data available via the data hub beyond core master data
  • Takes an end-to-end approach for the data management initiative
  • Transparently opens the initiative to the whole enterprise

Read the full interview here.

I hope to be able to present more people behind successful solutions on The Disruptive MDM / PIM List Blog.

Human Errors and Data Quality

Every time there is a survey about what causes poor data quality the most ticked answer is human error. This is also the case in the Profisee 2019 State of Data Management Report where 58% of the respondents said that human error is among the most prevalent causes of poor data quality within their organization.

This topic was also examined some years ago in the post called The Internet of Things and the Fat-Finger Syndrome.

Errare humanum estEven the Romans knew this as Seneca the Younger said that “errare humanum est” which translates to “to err is human”. He also added “but to persist in error is diabolical”.

So, how can we not persist in having human errors in data then? Here are three main approaches:

  • Better humans: There is a whip called Data Governance. In a data governance regime you define data policies and data standards. You build an organizational structure with a data governance council (or any better name), have data stewards and data custodians (or any better title). You set up a business glossary. And then you carry on with a data governance framework.
  • Machines: Robotic Processing Automation (RPA) has, besides operational efficiency, the advantage of that machines, unlike humans, do not make mistakes when they are tired and bored.
  • Data Sharing: Human errors typically occur when typing in data. However, most data are already typed in somewhere. Instead of retyping data, and thereby potentially introduce your misspelling or other mistake, you can connect to data that is already digitalized and validated. This is especially doable for master data as examined in the article about Master Data Share.

6 Decades of the LEGO® Brick and the 2nd Decade of MDM

28th January 2018 marks the 60th anniversary of the iconic LEGO® brick.

As I was raised close to the LEGO headquarter in Billund, Denmark, I also remember having a considerable amount of LEGO® bricks to play with as a child back in the 60’s in the first decade of the current LEGO® brick design. At that time the brick was a brick, where you had to combine a few sizes and colours of bricks into resembling a usable thing from the real world. Since then the range of shapes and colours of the pieces from the Lego factory have grown considerably.

MDM BlocksMaster Data Management (MDM) went into the 2nd decade some years ago as reported in the post Happy 10 Years Birthday MDM Solutions. MDM has some basic building blocks, as proposed by former Gartner analyst John Radcliffe  back in 00’s and touched in the post The Need for a MDM Vision.

These blocks indeed look like the original LEGO® bricks.

Through the 2nd decade of MDM and in coming decades we will probably see a lot of specialised blocks in many shapes describing and covering the people, process and technology parts of MDM. Let us hope that they will all stick well together as the LEGO® bricks have done for the past 60 years.

PS: Some if the sticking together is described in the post How MDM, PIM and DAM Stick Together.

A Pack of Wolves, Master Data and Reference Data

Pack of WolvesDuring the last couple years social media have been floating with an image and a silly explanation about how a pack of wolves are organized on the go. Some claims are that the three in the front should be the old and sick who sets the pace so everyone are able to stay in the pack and the leader is the one at the back.

This leadership learning lesson, that I have seen liked and shared by many intelligent people, is made up and does not at all correspond to what scientists know about a pack of wolves.

This is like when you look at master data (wolves) without the right reference data and commonly understood metadata. In order to make your interpretation trustworthy you have to know: ¿Who is the alpha male (if that designation exists), who is the alpha female (if that designation exists) and who is old and sick (and what does that mean)?

PS: For those of you who like me are interested in Tour de France, I think this is like the peloton. In front are the riders breaking the wind (snow), who will eventually fall to the back of the standings, and at the back you see Chris Froome having yet a mechanical problem when the going gets tough and thereby making sure that the entire pack stays together.

What Will you Complicate in the Year of the Rooster?

rooster-6Today is the first day in the new year. The year of the rooster according to the Lunar Calendar observed in East Asia. One of the characteristics of the year of the rooster is that in this year, people will tend to complicate things.

People usually likes to keep things simple. The KISS principle – Keep It Simple, Stupid – has many fans. But not me. Not that I do not like to keep things simple. I do. But only as simple as it should be as Einstein probably said. Sometimes KISS is the shortcut to getting it all wrong.

When working with data quality I have come across the three below examples of striking the right balance in making things a bit complicated and not too simple:

Deduplication

One of the most frequent data quality issues around is duplicates in party master data. Customer, supplier, patient, citizen, member and many other roles of legal entities and natural persons, where the real world entity are described more than once with different values in our databases.

In solving this challenge, we can use methods as match codes and edit distance to detect duplicates. However, these methods, often called deterministic, are far too simple to really automate the remedy. We can also use advanced probabilistic methods. These methods are better, but have the downside that the matching done is hard to explain, repeat and reuse in other contexts.

My best experience is to use something in between these approaches. Not too simple and not too overcomplicated.

Address verification

You can make a good algorithm to perform verification of postal and visit addresses in a database for addresses coming from one country. However, if you try the same algorithm on addresses from another country, it often fails miserably.

Making an algorithm for addresses from all over the world will be very complicated. I have not seen one yet, that works.

My best experience is to accept the complication of having almost as many algorithms as there are countries on this planet.

Product classification

Classifications of products controls a lot of the data quality dimensions related to product master data. The most prominent example is completeness of product information. Whether you have complete product information is dependent on the classification of the product. Some attributes will be mandatory for one product but make no sense at all to another product by a different classification.

If your product classification is too simple, your completeness measurement will not be realistic. A too granular or other way complicated classification system is very hard to maintain and will probably seem as an overkill for many purposes of product master data management.

My best experience is that you have to maintain several classification systems and have a linking between them, both inside your organization and between your trading partners.

Happy New Lunar Year

IT is not the opposite of the business, but a part of it

Yin and yangDuring my professional work and not at least when following the data management talk on social media I often stumble upon sayings as:

  • IT should not drive a CRM / MDM / PIM /  XXX project. The business should do that.
  • IT should not be responsible for data quality. The business should be that.

I disagree with that. Not that the business should not do and be those things. But because IT should be a part of the business.

I have personally always disliked the concept of dividing a company into IT and the business. It is a concept practically only used by the IT (and IT focused consulting) side. In my eyes, IT is part of the business just as much as marketing, sales, accounting and all the other departmental units.

With the raise of digitalization the distinction between IT and the business becomes absolutely ridiculous – not to say dangerous.

We need business minded IT people and IT savvy business people to drive digitilization and take responsibility of data quality.

Used abbreviations:

  • IT = Information Technology
  • CRM = Customer Relationship Management
  • MDM = Master Data Management
  • PIM = Product Information Management

It is not all about People or Processes or Technology

People Processes TechnologyWhen following the articles, blog posts and other inspirational stuff in the data management realm you frequently stumble upon sayings about a unique angle towards what it is all about, like:

  • It is all about people, meaning that if you can change and control the attitude of people involved in data management everything will be just fine. The problem is that people have been around for thousands of years and we have not nailed that one yet – and probably will not do that isolated in the data management realm. But sure, a lot of consultancy fees will go down that drain still.
  • It is all about processes. Yes it is. The only problem is that processes are dependent on people and technology.
  • It is all about technology. Well, no one actually says so. However, relying on that sentiment – and that shit does happen, is a frequent reason why data management initiatives goes wrong.

The trick is to find a balance between a priceworthy people focused approach, a heartfelt process way of going forward and a solid methodology to exploit technology in the good cause of better data management all aligned with achieving business benefits.

How hard can it be?

Bookmark and Share