Eighty-twenting MDM

Master Data is a set of core information attributes that describe the core entities of an organization, organized in data domains, these domains usually include customers, vendors, products, or assets.

This means that master data is at the heart of every business process within an organisation and because it’s used in multiple systems and processes, bad master data will have huge effects in the business processes.

The negative impacts of unmanaged master data can easily be identified across almost multiple levels of an organization’s activities, impairing the decision-making processes, hence impacting performance, customer, and product profitability, but also with impacts at an operational level, reducing productivity and efficiency, but also leading to compliance risks.

A master data management (MDM) solution creates a single master record for all the critical business data across internal and external sources and applications, providing a comprehensive view of all the data.

Master data management should cover the process of collecting the data that is for each of these domains and provide it to all relevant systems and stakeholders.

The purpose of managing this data is to assure a consistent definition of these business entities and data about them across the organization’s multiple systems, establishing a standard definition for business-critical data that represents a single source of truth.

Getting started

Implementing a master data management solution is an ambitious goal and often, the results are far from the expected in multiple levels. These can be expensive initiatives, time and resource consuming and can span through long time frames, they can become somehow intrusive and disruptive, creating the natural resistance to change within the organization, creating a very challenging ecosystem to work on.

Although information on how to develop a Master Data Management solution is quite abundant, and it quite easy to find a few frameworks that can easily be adapted to the tool of choice, it is still a challenge if it’s not correctly approached and planned.

• The priority must be to establish the business vision. What’s the role of Master Data in the overall business strategy, and consequently in data strategy?
• Whatever the driver behind the initiative it is essential that clear, ambitious objectives are set from the beginning - objectives that can be clearly related to business objectives and evaluated by the business value they generate.
• Creating a roadmap is an essential tool. Mapping all the initiatives needed to complete the Master Data Management objectives, identifying which data domains to address, which data to include in those domains, which systems and processes will be involved, identifying the existing gaps between the current situation and the future situation, and most important the existing gap between business and the existing IT ecosystem.
• Select small, targeted initiatives, where the impact and value of master data can be clearly identified, with business stakeholders that can passionately and effectively articulate the impacts of master data in their business processes and that will be eager to defend the project.
• Apply an agile development mindset to all this process, start with a minimum viable solution and iterate, allow those visible results are presented in short time lapses.
• When these initiatives are successful and deliver the intended benefits, business leaders will be encouraged to push to achieve more, not only focusing on what works well, but also on letting go of what doesn’t work.

The 80/20 factor

Even when approached considering what I’ve mentioned before, there are still additional challenges when trying to integrate data from multiple, disparate sources, often incompatible, with the same data being produced and handled under different rules, with different formats across the organization’s different systems.

This will often lead to increasingly complex project, that will consume time, resources and most likely lead to an underperforming solution.

Before this happens it's important to consider the scope of data to be integrated, what is in fact critical considering the objectives and necessary to generate maximum value from the solution.

As I’ve described in an previous article (https://www.dhirubhai.net/pulse/eighty-twenting-data-jose-almeida/) the application of an 80-20 rule can also, in this situation, be a very useful tool. ?

Again, what is proposed here is to question what data should be included in the Master Data Management scope, what should be included to maximize the result, minimizing the resources and time needed to achieve those objectives.

Asking questions like:

• Which 20% of domain attributes are mostly used across multiple areas/systems?
• Which 20% of domain attributes are more critical for compliance purposes?
• Which 20% of attributes are more critical within a domain?
• Which 20% of attributes are more frequently used in the business processes?
• Which 20% of the data sources have the highest quality?

Doing this exercise can lead to a change in perspective from the initial situation, and lead to the decision to exclude or include attributes, to include or exclude data sources, to what data is in fact master data, to what data can be managed locally within its original systems, etc.

Tweaking the scope can have a huge impact, not only on the solution, but also on the adoption, on the costs and bottom line on the success of a project.