FREE DM Review Site Registration!
Sign-up today and access DM Review on the Web!

Your FREE registration entitles you to:
FREE email newsletters

FREE access to all DM Review content

FREE access to web seminars, resource portals, our white paper library and more!

   

Customer Data Integration: Avoiding the Pitfalls of "All or Nothing"

A global manufacturing company faced a dilemma. After a series of mergers and acquisitions - and through some impressive organic growth from its existing customer base - the company had what appeared to be an enviable problem: too much information about their customers.

With each new acquisition they not only gained additional customers, but each of the newly acquired business units came with their own customer relationship management (CRM) or enterprise resource planning (ERP) systems. Some new divisions had multiple views of each customer base, with marketing databases, sales force automation and various other systems that supported their customer-facing endeavors.

In actuality, the problem for this particular company wasn't too much information, but rather too many sources. As a result, the company was left with a confused, disjointed view of its customer base which complicated efforts to provide them with new and better services. To help clarify its view of their customers, the company evaluated customer data integration (CDI). CDI provides the technological framework to integrate all customer information into one operational repository. This approach - an outgrowth of yet another acronym, master data management or master data management (MDM) - would help the company eliminate those silos of customer data and build one single view of the customer.

The CDI process sounds simple, but a quick analysis of the data revealed some significant problems. Each ERP and CRM data source had different methods of defining the customer. Was the customer a distributor or a consumer? Or both? More complex questions began to come up, such as how to account for a customer that appeared in a CRM system of two or more business divisions?

Once they dug a little deeper, the problems became more complicated. The data sources each had different standards applied to data elements. For example, the state of California could be spelled out or abbreviated as Calif., Cali., CA, Ca. and Cal. While such a small bit of data may seem inconsequential, imagine the complications it could cause while attempting to integrate data from different systems.

Rather than proceed with a CDI effort, the manufacturer realized that it had some work to do. Not only was their data inconsistent across sources and applications, but there was no corporate agreement about the fundamental aspects of the customer data.

This story is not unusual. Companies worldwide are finding that the hype of CDI is far outpacing the successful implementations. The rationale for CDI is sound. Who wouldn't want a single view of the customer? But the difference between the promise and the reality doesn't appear to be shrinking.

CDI: Where to Begin?

It is critical to understand that achieving CDI is an evolutionary process; CDI or any other MDM venture should not be an "all or nothing" project. A company that has created a disconnected network filled with poor-quality, disjointed data cannot expect to progress to CDI quickly since, at the core of any CDI project is the consistency, accuracy and reliability of the customer information. Without good information, the final CDI repository will be an expensive but useless amalgam of data.

The danger of a well-meaning but overreaching CDI implementation is written in the history of IT projects. The lessons of large-scale ERP and CRM implementations (where a vast majority of implementations failed or underperformed) illustrate that projects like CDI or MDM are not just a technology issue. An IT failure is rarely the fault of the technology alone. Instead, the typical cause of failure is the lack of support across all phases of the enterprise.

Beyond the Technology of CDI

Too often, companies view CDI as a technology problem with a technology solution. After all, there are CDI technologies on the market that are advertised as solutions to CDI problems. However, the technology alone isn't enough. The execution of any new large-scale IT implementation requires a combination of people and processes to ensure success.

The good news is that the evolution of CDI coincides with the emergence of data governance techniques. Many companies are now implementing enterprise-wide data governance programs which attempt to codify and enforce best practices for data management across the organization.

The tenets of data governance fit extremely well with the stated goals of CDI. In fact, starting a CDI program without a data governance program would be like trying to implement new government without any laws or administrative function. Data governance is the cultural and procedural backbone behind CDI or any MDM effort. Without it, the technology will not have the desired effect - and the program will likely underperform or fail.

Although the goal of data governance is clear - the quality of information must improve to support core business initiatives - there is no definitive roadmap for starting these projects. However, most organizations have data governance programs that encompass two main elements:

  • People: Effective enterprise data governance requires executive sponsorship. Organizations must establish a data governance council composed of both business and IT staffs. These councils are responsible for defining a customer in the new CDI system - and how to mitigate disputes between information sources across corporate boundaries. Guided by an executive, these groups institute policies that control the quality and consistency of information within the CDI repository.
  • Policies: A data governance program must create and enforce what is considered "acceptable" data through the use of business policies that guide the collection and management of data.

After the council creates policies, a group of employees - often known as data stewards - turn these policies into established practices. Data stewards blend elements of both business and IT to build more effective and useful CDI systems. They understand the technical requirements of different applications. But, more importantly, they also recognize the business value and characteristics of data.

Data stewards assume the primary responsibility for managing the accuracy and reliability of a corporation's data. Fostering cooperation between IT and business users, data stewards can resolve disagreements between the two factions and find common ground for a CDI initiative. Data stewards are responsible for:

  • Establishing a "partnership" between business and IT,
  • Analyzing and documenting existing data sources that populate a CDI system,
  • Creating and documenting the business rules of a corporation during the data integration phase of a CDI project,
  • Creating standards and procedures for data governance, and
  • Monitoring data quality over time and resolving problems.

On to the Technology

Once the people and policies are in place, it is time to bring in the technology to support CDI implementations. Often, the move to CDI requires integrated technology to help the company analyze, improve and control the quality of the information that will populate a CDI repository.

To build useful and reliable information, companies need to take the following five-step approach:

  • Data profiling. Inspect data for errors, inconsistencies, redundancies and incomplete information. For companies that have grown exponentially through mergers, acquisitions or increased sales efforts, this phase helps scope the duration and size of the CDI effort.
  • Data quality. After profiling the data, the strengths and weaknesses of this information become clear. The data quality phase provides the opportunity to build rules that correct, standardize and normalize information.
  • Data integration and consolidation. A key element of any CDI effort is the ability to match, merge or link data from a variety of disparate sources. At this phase, a harmonized view of the customer begins to appear.
  • Data enrichment. Once data is consolidated, records can be enhanced through the utilization of internal and external data sources. This could mean adding a unique customer identifier and a priority code to customers based on a variety of factors (products owned, total customer value, etc.). 
  • Data monitoring. After the data has gone through the previous phases, it's critical to check and control data integrity over time. Every month, approximately two percent of all customer data becomes outdated. Regular monitoring helps keep good data from going bad.

After these steps are complete there will be a solid pool of customer information which will serve as the foundation of the final CDI repository. More importantly, the created rules can improve the quality of information as it enters CDI - and they can be applied in real-time as new records are added to any application over time. The CDI repository then starts with clean, useful data and real-time business rules ensure the consistency of that information over time.

For companies investigating CDI, the task may seem overwhelming. However, by approaching CDI as a series of challenges that encompass a mixture of people, policies and technologies, you can take a more rational, evolutionary approach.


Daniel Teachey is the director of corporate communications at DataFlux, a wholly owned subsidiary of SAS, that enables companies to analyze, improve and control their data through an integrated technology platform. Teachey can be reached at daniel.teachey@dataflux.com.

For more information on related topics, visit the following channels:



Industry Vendors