As organizations seek to become digital to the core, the impact of data trends and the opportunities they could offer cannot be ignored. That is why this blog focuses on exploring the potential of combining two of the major technology domains Master Data Management and big data analytics.
Out of the several use cases discovered, some have been presented in the blog. However, before the probable relationship is discussed, it is necessary to understand these two domains individually and their areas of applications.
MDM and big data analytics
In simple terms, master data is the information generated and consumed by business processes across companies’ value chains and business functions. Typically, companies establish a governance framework around their master data that is supported by a technology-based maintenance process to gather and store the master data in a uniform and consistent way and to share it with all relevant stakeholders. Organizations have so far reaped the value of enterprise data and master data management (MDM) solutions in several ways e.g. through increased operational efficiencies, enhanced management decision making, and meeting stringent regulatory requirements. While some organizations have fully implemented an MDM strategy and most others are harnessing the value, the market demands higher performance MDM solutions. These solutions include managing multiple MDM domains, MDM on cloud solutions, MDM mobile applications and the potential synergies between MDM and big data analytics.
In contrast to the master data, big data is a large set of structured and unstructured data generated through different internal and external sources such as social media, e-commerce, transactional data, sensor logs etc. So far, organizations have leveraged big data analytics in several ways e.g., by modifying e-commerce sites in real-time, decreasing machine maintenance costs, personalizing consumer offerings, developing interactive visualizations, and building smart city projects.
Integration yes or no?
However, the question that is most commonly asked is whether there is any correlation or a bidirectional relationship between both domains? There have been conflicting opinions and skepticism over the integration of MDM with big data analytics. While some argue that MDM would not be relevant in the future, as big data analytics can solve the challenges that are solved by MDM, others emphasize that big data analytics might not be of any relevance to MDM. Based on our deep consulting expertise in Enterprise Information Management (EIM), we argue that an effective overlap exists between both domains, each enhancing the performance of the other (see figure 1). Organizations can reap the full potential of big data analytics when it is used in complement with MDM and vice versa.
There are many use cases for the beneficial interplay of MDM and big data analytics. Two prominent examples are:
Data Completeness: Big data analytics combines data from a variety of sources such as the Internet of Things; combining this data with internal master data can extend and enhance the 360° view of the customer. Visualizing this data also improves decision making.
Structure and trustworthiness: MDM provides structure to the unstructured big data and an approach to govern the data. MDM can be a starting point and a structural basis for big data such as customer sentiments, intent, product perceptions etc. Only MDM delivers the trusted data required to reap the business value of big data. A potential use case is fraud detection, both inside and outside the company.
In a nutshell, if companies want to remain competitive, the opportunities of combining big data analytics and MDM cannot be ignored. Almost every area of the company can benefit from using data better: taking more effective strategic decisions, making processes more efficient or discovering new business models. Organizations have invested significant time and money in coping with data trends. However, past credibility has been damaged when expensive analytics investments didn’t perform as expected. Data trends are sometimes hyped and before making any investments organizations should assess the current state processes, technology, governance, and improvement opportunities.