January 8th, 2024 | 7 min read
Which data types should be managed by a Master Data Management solution?
Tradition dictates that there certain types of data that should be handled by a Master Data Management (MDM) platform, and others that shouldn’t. However, the sheer diversity and complexity of data mean that these notions are being challenged, as organizations realize that in order to achieve a comprehensive and actionable view of any domain, they need the bigger picture.
Before we examine the types of data that an MDM solution is suitable for, we should first define the different data types:
Metadata: Metadata is often described as "data about data." It provides information about or documentation of other data managed within an application or environment. For example, metadata can describe various attributes of data, such as the author, date created, date modified, file size, and file format. Metadata is used across various fields, from digital asset management in libraries and museums (cataloguing items) to IT (managing data within systems) and web development (SEO metadata like meta tags that describe webpage content).
Reference Data: Reference data is a type of data that is used to categorize or classify other data within a system, database, or application. It typically consists of relatively static sets of standard data, often used for classification and cross-referencing purposes. Examples include geographical information such as country codes and state or region names, currency codes, product categories and industry codes (i.e. SIC).
Master Data: Master data represents the core entities around which business activities are organized. It is the consistent and uniform set of identifiers and extended attributes that describe the core entities of the enterprise and is often used in different systems and processes. Master data typically includes key business entities such as customers, products, employees, suppliers, and accounts. These entities are central to the operations and transactions of a business.
Unstructured Data: Unstructured data refers to information that does not have a predefined data model or is not organized in a predefined manner. It comes in various formats, including text documents, emails, social media posts, videos, images, audio files, and web pages. It can also include data from business applications like customer feedback forms and product reviews.
Transactional Data: Transactional data is generated when a transaction takes place. This could be a sale, a purchase, a booking, a delivery, a payment, or any other kind of business event. Each transaction typically includes details about the event, such as the time, date, participants, and the nature of the transaction. This type of data is highly detailed and specific to individual transactions, and by its very nature is time-sensitive and often generates continuously.
Redefining the boundaries of MDM
Rather than asking “Which data types should be processed by a Master Data Management solution?”, a better question might be “Which data needs to be cleaned, catalogued, normalized, enriched, and governed?”. Or, in the context of a specific use case like building a single view of an entity such as a customer or product “Which data do I need to give me the most accurate and valuable view?”
Where traditional MDM solutions are often limited to the four general master data domains (customers, products, locations, and “other”), there are many occasions where it would make sense to include other data types to build a more useful picture. If we take the example of the customer domain, in addition to having a single view that includes up-to-date contact and billing information, wouldn’t it also be beneficial to understand their purchase history, web interactions, and support requests?
To do this, MDM solutions should be adept at managing unstructured data, such as emails, documents, and social media content, as well as semi-structured data like emails and meeting files. This capability is crucial for deriving comprehensive insights and ensuring that all relevant data is governed and utilized effectively.
The Role of Advanced Technologies
The integration of Big Data technologies, schema-less databases, and horizontally scaled data stores has enabled augmented MDM solutions like CluedIn to handle diverse data types more efficiently. These advancements are critical for MDM systems to remain relevant and effective, and to keep up with the shape-shifting nature of data.
Transactional Data and High-Velocity Considerations
Another common feature of advanced MDM solutions like CluedIn is the ability to scale almost endlessly in order to accommodate various types of data such as transactional data. However, careful consideration needs to be given to whether an MDM platform is the right place for this type of data. The higher the volume of data, the more processing will be required and the higher the cost in terms of both time and money. This cost needs to be balanced against how critical the data is to supporting the needs of the business.
In summary, a modern Master Data Management solution should be versatile enough to handle a wide array of data types, similar to the capabilities demonstrated by platforms like CluedIn. The days of restricting yourself to only putting the “classic” MDM domains into the system are over, rather it should be a case of assessing which data can benefit most from the value MDM can bring. If your current MDM system falls short in this regard, it may not be fully equipped to meet the complex data management needs of today's businesses. The future of MDM lies in its ability to adapt, integrate, and manage multiple forms of data, ensuring comprehensive, accurate, and actionable insights for business decision-making.