MDM is not a technology but rather a discipline that needs to be meticulously introduced in an organization. MDM is a requirement for all organizations, especially in manufacturing where millions of dollars are spent in marketing and efforts to provide a better customer experience. Lately, MDM has evolved as a necessity for businesses with a direct focus on customers.

What Is MDM?

To understand MDM, it is essential to comprehend  company data in its entirety, especially all the sources that need data or that provide valuable data. Understanding the data and logically grouping that data in respective modules can be termed as an MDM function. In standard terms, MDM is the business capability that delivers a single, trusted and complete version of critical data assets to applications and end users to support efficient operational business processes and strategic decision-making.

Many places you would find MDM referred to as set of guidelines while others refer to it as a tool. In the latter case, it is assumed that individuals have basic knowledge about MDM. In many presentations, the MDM definition is combined with the software.  Assuming that software will be MDM for your company  is one of the common mistakes that many organizations tend to make while evaluating an MDM tool during the course of MDM initiative.

MDM of What?

We already discussed what MDM is, but the important question is what do you want from MDM? Most commonly, this is a need to manage customer contacts or products. Managing customer contacts is termed customer MDM and managing products is called product MDM.Here’s what to consider before and during the implementation of both.

Most companies want to implement MDM as something that will improve everything in the organization, from customer experience to marketing efforts and more. Faulty implementation of MDM can cause financial losses and a frustration of resources. In a general sense, three reasons for a poor implementation are selecting the incorrect product, bad MDM architecture, and a lack of stated business requirements to set the course for the initiative.

In an effort to avoid some of these issues right from the beginning, establish a well-defined problem statement. Everyone involved should know this statement, which covers the MDM requirement and states the end goal. Many times, organizations start MDM implementations based on past experience and challenges but they do so without defining a clear-cut goal.  In big organizations, it is very common to have such a problem due to pressure from senior management. In such cases, senior management believes that they have enough justification and use cases to go ahead with MDM implementation. The implementation team proceeds in order to achieve their performance goal. In both cases, involved individuals fail to create a problem statement, and the implementation fails to attract business attention down the road, particularly if implementation is perceived as something that doesn’t meet their requirements.

Once you have the problem statement clearly defined, the next step in your initiative should be to assess the business needs. This will be a full set of requirements (as determined by the business) in order for them to run their daily activities smoothly. Daily activities can be anything, including customer interaction to marketing campaigns. In this phase, it is very important to get requirements documents and obtain necessary approval from the business. This is an essential step, and a considerable amount of time should be spent in here. I highly recommend that the organization should not launch an MDM implementation if business engagement is not properly obtained.

MDM Tool Selection

There are numerous MDM products available and plenty of information can be found online. However, before selecting a tool, it is highly recommended to conduct a proof of concept for various MDM tools to understand the features offered. Vendors will generally conduct a POC free of cost, and the POC should be done with multiple MDM tools so that a good comparison can be made of various MDM features. Here are some important elements to watch for in doing the POC:

  • Infrastructure: How much infrastructure dedication is required for the tool once it’s deployed? Note: it is likely that infrastructure will double the cost of the MDM implementation.
  • Ease of deployment: Consider if you have the resources to handle the deployment requirements. If tool is difficult to deploy then lot of resources may be spent in doing setup. Sometimes it takes multiple iterations to do the setup properly, resulting in a high end cost.
  • Modification: Once deployed, how easy will it be to modify the existing configuration in the deployed tool? Determine if such modifications will require additional involvement by the vendor.
  • Data stewardship functionality: Stewardship is an important feature of a tool (and of the initiative as a whole) and a data steward, acting as a representative of the business, would utilize this feature to make corrections. If this feature is not available or is limited, making corrections to business processes may be severely impacted.
  • Customization: Though not vital, it would be a bonus if the tool permitted external things to be added. Some tools offer ad hoc coding features to support customization.
  • Software complexity: One of the added advantages for an organization  comes from MDM software that is simple to understand.

More Prerequisite Steps

You need to identify appropriate data sources. Often, organizations decide to push all sources for rationalization, and this can cause a serious problem of data availability. It is important that only relevant sources should be selected for rationalization.

Additional elements of the MDM implementation include:

  • Data profiling
  • Data cleansing
  • Identification of downstream applications and consumers
  • Data governance
  • Decide on method of implementation: Single entity versus  multi-entity MDM, registry style or transaction style (or combined).

(See Figure 1, left, for a high-level view of processes that are essential for a successful MDM implementation.)

Finalize Your Solution

Companies often opt to involve consultants to assist in the MDM implementation, particularly at organizations that lack internal expertise and experience. A consultant can work with the vendor to develop a solution design for the MDM implementation.    

Post-Implementation Tasks

Once MDM deployment is complete, there are a few things that should be monitored continuously:

  1. Performance: MDM can reap maximum benefits when rationalized data is available at the right time. In most cases, the business demands real-time availability of rationalized data, especially in the case of sales.
  2. Growth of data: At some point, organizations should have an archival data strategy to keep up performance and keep down costs.
  3. Compression ratio: Compression of similar records coming from different sources should be above the threshold specified at the beginning. In order to determine compression ratio, proper data profiling is recommended.

If an MDM strategy is correctly implemented, an organization can realize returns across its business units, particularly with customer information. Although in need of a definite path and lots of work, it is a myth that MDM is difficult to implement and requires lots of expertise.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access