Upen Varanasi
Tuesday, February 3, 2009
Everyone must have heard of Web 2.0 by now. Various software gurus across the globe have been touting Web 2.0 for the past 3-4 years. Wikipedia has this to say of Web 2.0: “The term “Web 2.0” describes the changing trends in the use of World Wide Web technology and web design that aim to enhance creativity, communications, secure information sharing, collaboration and functionality of the web”. One can make the argument that a similar term can be applied while talking about the changing trends in enterprise software. Some of the key new factors shaping the behavior of enterprise software are:

* Global Scope: Enterprises span national boundaries and are becoming more global in scope. This is introducing concepts of globalization and localization to enterprise application architectures.

* Communication and Connectedness: Applications can no longer operate in “Silos”. Better communication between applications is needed to handle the desire of business users to have more powerful tools that enable quick and effective decision making. Applications need to communicate with one another in a way that goes well beyond the concepts of Enterprise Application Integration (EAI). True applications integration can only be achieved on the basis of new modular and services based architecture.

* Better Trading Partner Collaboration: There is tremendous need for better communication across enterprise boundaries: vendors and customers. The “Wal-Marts” of the world have redefined the definitions of collaboration across their supply chains.

The above mentioned trends are forcing enterprise applications to be more flexible, scalable and part of a collaborative application framework. We could apply the term “Enterprise 2.0” to describe the next generation of enterprise applications.

Enterprise 2.0

Two key concepts that form the underpinnings of Enterprise 2.0 are Service Oriented Architecture (SOA) and Master Data Management (MDM). To truly understand the changes being wrought by these concepts, let’s take a look at Enterprise 1.0 or the earlier generation software and compare it to the latest trends.
Enterprise software applications in the 90’s were built on a client-server or simple web-based architectures. The applications tended to be built using Object-Oriented techniques. Enterprises would typically have hundreds of applications with each catering to a specific problem area such as eCommerce, CRM, ERP, Supply Chain etc. This led to the creation of application “silos”. Application vendors either solved a specific problem only or in the case of the mega-vendors, huge monolithic application stacks got created that claimed to solve all the problems within an enterprise. Needless to say, these application stacks are “Jack of all domains, Master of None”. There was a gradual realization that enterprises need “Best-of-Breed” solutions that can talk to one another effectively to create a robust application network within the enterprise. Business users would get the benefit of superior technology without compromising on communication.

Enterprise applications needed to talk. Over time it became obvious that “Service” was a better unit than the earlier concept of “Object” while trying to make applications talk to one another. This led to the emergence of SOA. All the major vendors began rewriting/re-architecting parts of their application to make them more “services” friendly. But one problem remained. For the services to effectively communicate with one another there had to be a common definition of master entities or master data. MDM is a new and emerging field that deals exclusively with the management of master data within an enterprise.


MDM comprises a set of processes and tools which consistently define and manage the non-transactional data entities of an organization (also called reference data) such as Product, Customer, Employee, GL, Assets etc. MDM systems act as a dictionary for effective communication between enterprise applications. This is similar to the role of a dictionary in the realm of human communications. MDM systems handle the entity definitions as well the “CRUD” (Create, Update, Delete) processes associated with them. These systems essentially become the “System of Truth” for master data. The master data is kept in “synch” across various enterprise applications in a process known as “Internal Synchronization”. Some of the key benefits of implementing MDM systems are:

* Effective SOA implementation: Effective SOA implementation needs the implementation of one or more MDM systems.

*Consistent and Higher Quality of Master Data: MDM systems ensure that there is consistent master data across enterprise applications. By centralizing master data management, enterprises benefit from superior quality of master data. This has a direct impact on sales (increase), invoice errors (decrease) and customer support (better handling).

*Increased Agility: Enterprise can quickly and effectively execute on key business processes such as New Product Introductions, Product and Pricing Updates, New Customer Introduction etc. This can lead to significant competitive advantage with respect to their competitors.

Multi-Channel Retailer

Let take an example of a “multi-channel” retailer handling product master data to illustrate the usage of an MDM system. “Multi-channel” retailers utilize more than one channel (Store, Online and Print) to reach their customers. For a retailer, product master data is a key component fueling their eBusiness initiatives. Multiple systems and processes handle product data within a retailer’s internal environment. Retailers on-board the supplier/manufacturer data, enrich it with their merchandising and promotional information and disseminate this information across various enterprise systems such as SCM, ERP, CRM, eCommerce Engines, Data Warehouses etc. Retailers need to ensure that the information is consistent across these systems. Retailers also have to deal with localization / customization of master data for a particular geography, customer segment or a channel. The master data across their various selling channels such as Store, Online and Print needs to be consistent. An MDM system would centralize all the above mentioned processes and act as the single “source of truth” of master data across the enterprise. The MDM system would enable internal and external synchronization of master data. A few of the key synchronization touch points are:

* ERP: Synchronize part number, descriptions and price information

* SCM: Synchronize inventory levels and vendor information

* eCommerce: Synchronize marketing, technical and commercial information

*Print Publishing: Synchronize information with desktop publishing systems to enable printed catalog creation

*Data Warehouse: Synchronize master data for reporting purposes

* Merchandising:Synchronize promotions, up-sells, cross-sells etc.

* External synchronization: synchronize master data with industry specific data pools.
The image below describes the systems and processes involved in managing product master data within a retail environment.


SOA and MDM are becoming cornerstones of new improved enterprise software architecture. Enterprises looking to implement SOA strategies need to implement some form of a MDM solution first. Before implementing an MDM solution, it is recommended to understand the logical master data model within an enterprise first. This should drive the adoption of the MDM solution and identify the key entities (product, customer or vendor) that need a master data management approach.

The author is Co-Founder and CEO of Riversand Technologies

Share on LinkedIn

Previous Magazine Editions