How Mature is Your Data Management Environment?

In the past, it was fairly straightforward for a company to maintain a competitive advantage: build a better product, beat the competition on price, or create an irresistible buzz in the
marketplace. Today, a competitive advantage is not so cut and dried. If a company wants to succeed long-term, it must leverage information to enter new markets, maximize customer value, make
results-oriented decisions, and keep costs down. For many of today’s successful players, particularly the ones that sustain leadership atop their sectors, success is tied to the efficiency
with which they operate and the quality and pace of their business decisions.

What fuels them? How do they set the wheels in motion to make better business decisions—decisions measured more by the promise of reward than by the threat of risk? The answer lies in data
quality, which starts at a basic understanding of the evolution of data.

This article examines the Enterprise Data Management Maturity Model, which helps companies identify and quantify their data maturity, and assess the risks of undervalued data management practices.
The maturity model also helps organizations understand the benefits and costs of moving to the next stage.

Introduction

Gaining a competitive advantage in the market used to be a straightforward, common-sense process. You could provide industry-leading products and services, setting the standard in your market, or
you could excel at marketing and sales functions and create an irresistible “buzz” within your prospect base.

Times have changed drastically. Today, gaining a competitive edge is extremely difficult. Companies must build new systems, implement new strategies, or identify new markets in order to compete or
survive. Historically, what has been ignored is proper data management that supports your ability to make reasonable, results-oriented decisions. Companies misunderstand how effective data
management yields a competitive advantage.

Although many organizations know data is an important corporate asset, data is quite different from other corporate assets. Unlike tangible corporate assets that have a structured value and
depreciation schedule, it is difficult for many companies to place a definitive value on data. As a result, the perceived lack of tangible value makes justifying data management efforts tricky.

Forward-thinking companies understand one key idea: the cost of ineffective data management is much higher than the cost of successfully managing data. Put simply, organizations depend on data.
Regardless of industry, revenue size, or competitive environment, every company relies on its data to produce information that guides effective decisions. The quality of the results from any
analysis is only as good as the quality of the inputs (the data) that feed that analysis.

Data management establishes and deploys the roles, responsibilities, policies, and procedures pertaining to the acquisition, maintenance, dissemination, and disposition of data. To succeed, a
data-management program requires a partnership between the business and technology groups. The business areas are responsible for establishing the business rules that govern the data and are
ultimately responsible for verifying the data quality. The information technology (IT) group is responsible for establishing and managing the overall environment—architecture, technical
facilities, systems, and databases that gather and house data throughout the enterprise.

Given the broad focus of data management, an effective program relies on a combination of people, processes, and technology. We’ll explore the major issues in building better data and how to
utilize these three elements to achieve more effective data management policies. More specifically, this article defines an organization’s data management maturity based on the processes
practiced and the value an organization places on data management.

Most organizations approach data management in the same way they have for decades. Data, as an entity, is a backoffice problem that is the domain of the IT staff. Business users have a limited role
in traditional data-management routines, although their job function may be the most affected by inconsistent, inaccurate, or unreliable data.

Another complication in most data management structures is a narrow focus into the depth and breadth of data-related problems. Often, people see only the data in front of them. There is little
coordination across corporate or geographical boundaries. This effect—plus the separate intellectual silos for business and IT users in data management tasks—leads to confusion,
disputes, and a narrow view of events. Solid data management can help you achieve a more complete picture and facilitate crossboundary communications on data management.

The Enterprise Data Management Maturity Model
The Capability Maturity Model for Software (also known as the CMM and as SW-CMM), published by the Software Engineering Institute (SEI) and Carnegie Mellon University, is a well-established model
that defines software development maturity of organizations based on procedures and processes. However, it does not address the maturity of an organization with respect to how data is managed.

A new maturity model—the Enterprise Data Management Maturity Model—helps companies identify and quantify their respective levels of data maturity. By assessing an organization’s
data management maturity, you can understand the risks associated with undervalued data management practices. The maturity model also helps organizations understand the benefits and costs
associated with a move to the next stage.

Organizations must recognize the advantages of refining and strengthening their data management processes. Those that plan their evolution in a systematic fashion gain over those that are forced to
change by events that are outside their control. By implementing change on your terms (within a reasonable timeframe), you can accurately set goals for data maturity.

Understanding the maturity model can help you control your data management environment. Knowing your current stage and why you are there will help you understand how and when to move to the next
stage.

Assessing the current level is only the start, of course. Organizations also need to determine what stage is appropriate for them and establish actions and priorities for improvement.

Figure 1 shows the Enterprise Data Management Maturity Model. Consisting of four stages of data management maturity in a continuum, movement from one level to the next is fluid and takes time.
Often, different parts of an organization may be at different stages of the maturity model. Ultimately, an organization (or any of its parts) may not choose to move to a higher maturity level if
the costs outweigh the benefits.

Figure 1. The enterprise data management maturity model

As Figure 1 illustrates, the potential rewards escalate as a company moves from level to level, and each stage of the model requires contributions and investments. We will examine aspects that
define each stage of development, including:

People: Who is involved and what contributions must they make?

Process: What activities must be performed?

Technology: What investments in technology must be made?

Risk and reward: What risks does the organization face at the current stage, and what might it gain from moving forward?

Stage 1: Unaware

At the initial stage of the Enterprise Data Management Maturity Model, confusion reigns. Cooperation about data issues between departments or job functions is rare. Companies at Stage 1 have few
defined rules and policies regarding data management. In fact, the same or similar data may exist in multiple files and databases. A company in this stage may have a billing database with
information on customers; but similar data may also exist in a sales automation and a marketing database.

Complicating the problem is that most organizations are in denial about the quality and usefulness of corporate data. Companies in the “unaware” category have little or no visibility
into data management costs or performance. As a result, data quality varies widely across the enterprise; a financial database may contain high-quality data while a marketing database is filled
with inaccurate address data or redundant records.

Finally, unaware companies have unorganized data management activities. A typical technique for companies in this stage is for data owners or IT staff to correct egregious errors only on an
“as-needed” basis. Except in isolated pockets, there is no understanding of why problems exist, or what impact these problems may have in the future.

Due to the risks of the first stage, competitive pressures often serve as the driver for improving data maturity. To progress, companies need to put measures and processes in place to recognize
problems with data integrity or usability. Often it is enough to merely acknowledge that data-management issues cause organizational problems and to target the source of these problems.
Recognition— plus a commitment to fix data management issues—will help an organization begin to understand data management problems, risks, and returns.


People


Process


Technology


Risk and Reward

  • Success depends on the competence of a few talented individuals
  • Organization relies on personnel who may follow different paths within each effort to reconcile and correct data
  • No management input or buyin on data-integrity problems
  • Executives do not comprehend the extent of data problems
  • Organizations tend to blame IT for data-quality issues
  • No defined data management processes in place; data management is chaotic and project-focused
  • Firefighting mode: address problems as they occur through manually-driven processes
  • Infrequent long-range resolution to problems
  • Redundant data exists throughout the organization, leading to wasted resources across functional units
  • Tools tend to be generalpurpose software (Microsoft Excel, Microsoft Access) and no intensive data management software is in use
  • No data profiling, analysis, or auditing is used to determine data characteristics
  • Data cleansing or standardization may occur in isolated areas or data sources
  • Technologies in place support manual quality improvement methods
  • Risk: Extremely high; data problems can result in lost customers (due to poor understanding of the customer’s value) or improper business procedures; a few scapegoats receive the
    blame, although processes are not in place to properly assign culpability
  • Reward: Low; outside of the success of an individual employee or department, companies reap few benefits from data management

Table 1. Characteristics of an “unaware” company

Stage 2: Reactive

When an organization reaches Stage 2, it understands data-management problems as they occur, and comprehends that data is critical to its success. Data-quality issues are addressed only as major
problems occur or projects derail. At best, the organization hopes to react to problems to mitigate the severity of outcomes.

At this stage, non-integrated point solutions perform different, specific tasks. Organizations experience variable quality and some predictability in data integrity. In addition, successful
individuals receive assignments to the most critical business initiatives to reduce risks and improve results in specific processes. Organizations realize that data management may be of value but
are not willing to provide the time and money to prevent problems. Most American companies are in Stage 1 or Stage 2, as the impact of data quality is just beginning to become a known variable in
the minds of executives and managers.

At Stage 2, solutions are non-integrated, disparate point solutions. The impetus for progressing to Stage 3 is often a strategic vision of certain managers or executives that better data management
can lead to tangible business results. To advance, companies have to integrate processes and technologies to achieve more with their data resources.

Organizations must also begin to document, establish, and enforce data management policies as a core competence of application development. As new systems and data stores come online, employees
must be cognizant of the data management procedures and protocols of the organization. To ensure the policies are in place, some level of compliance testing is necessary. Finally, organizations
must reach a consensus on ownership of the data-management processes and assign responsibility and support.


People


Process


Technology


Risk and Reward

  • Success depends on the skills of a group of technical employees (database administrators, IT staff, etc.)
  • Individuals create useful processes, but no standard procedures exist across groups or locations
  • Long-range solutions are infrequent
  • Little corporate management buy-in to the value of data
  • Stronger data management roles emerge, but the emphasis remains on correcting data quality issues as they occur
  • Most processes are shortrange and focus on recently-discovered problems
  • Within individual groups and departments, tasks and roles are standardized
  • Tactical data management tools are often available (such as solutions for data profiling or data quality)
  • Most data is not integrated, but some individuals or departments attempt integration efforts in isolated environments
  • Some database administration tactics emerge, such as reactive performance monitoring
  • Attempts to consolidate data (such as a data warehouse) require scrap and rework due to data quality issues
  • Risk: High, due to a lack of data integration and overall inaccuracy of data throughout the enterprise; while data is analyzed and corrected sporadically, data failures can still occur
    on a crossfunctional basis
  • Reward: Limited and mostly anecdotal; most ROI arrives via individual processes or individuals; little to no corporate recognition of data management benefits

Table 2. Characteristics of an “reactive” company

Stage 3: Proactive

Reaching the third stage of the maturity model allows companies to avoid risk and reduce uncertainty. At this stage, data management starts to play a critical role within an organization, as data
goes from being an undervalued commodity to an asset that helps organizations make better decisions. As a company in this stage matures, it receives more tangible value from consistent, accurate,
and reliable data.

At Stage 3, a company looks beyond the horizon to understand the impact of data problems on missioncritical information. The requisite technology to support high levels of data inspection and
correction are in place. The organization begins to receive executive- and management-level approval for data management projects. A few early adopters have the building blocks in place to achieve
this level of maturity.

Advancing to the fourth and final stage is as much an evolution of culture as an evolution of people, process, or technology. The culture shift starts to change the staff ’s behavior, while
new and better processes and technologies give workers a better framework for data improvement. The advances made in the previous stages provide a solid foundation for data management. To evolve to
Stage 4, implement these advances continuously and consistently, primarily by documenting and replicating best practices throughout the enterprise to reach the pinnacle of the Enterprise Data
Management Maturity Model.


People


Process


Technology


Risk and Reward

  • Management understands and appreciates the role of data management in corporate initiatives
  • Data management initiatives receive the personnel and resources necessary to create high-quality data
  • All or most areas of the organization are involved with data management processes
  • Executive-level decision makers begin to view data as a strategic asset
  • Corporate data is more standardized, consistent, and measurable; preventive measures are in place to ensure high levels of data quality
  • Data metrics are sometimes measured against industry standards to provide insight into areas needing improvement
  • Data management goals shift from problem correction to problem prevention
  • Data management technology providers become strategic partners with the organization and help define best practices while implementing the technology
  • A corporate data management group emerges to maintain corporate data definitions, synonyms, business rules, and business value for data elements
  • Ongoing data audits and data monitoring help the company maintain data integrity over time
  • Risks: Medium to low; risks are reduced by providing better information to increase the reliability of decisions
  • Reward: Medium to high; data quality improves, often in certain functional areas and then in broader realms as more employees join the early adopters

Table 3. Characteristics of a “proactive” company

Stage 4: Predictive

At Stage 4, organizations achieve almost complete certainty of results. Data quality is an integral part of all business processes, and is ingrained throughout the enterprise. Processes are
entirely (or almost entirely) automated. To keep data within accepted limits, data management processes are implemented in real time and continuously validated. Since historical issues of data
quality are known and understood, data-defect prevention is the primary focus of Stage 4 organizations. There are cross-organizational approaches to data quality, helping companies address data
problems that overlap business silos. Finally, an important distinction of organizations in this stage is that data management becomes a business process and not a technological tool.

At the final stage of the maturity model, a major culture shift has occurred within the entire organization. Instead of ignoring the implications of data management (or treating data quality as a
series of tactical projects), a comprehensive, enterprisewide program elevates the process of managing business-critical data. With backing from executive management and buy-in from all business
functions, the program can flourish, creating more consistent, accurate, and reliable information to support the entire organization.


People


Process


Technology


Risk and Reward

  • Full management buy-in for data management processes and standards
  • Data quality improvement has executive-level sponsorship with direct CEO support
  • A data management group operates across the organization and has the support of data quality stewards, application developers, and database administrators
  • Entire organization is committed to “zero defect” policies for data collection and management
  • Procedures help the organization achieve the highest levels of data integrity
  • Processes in place ensure data remains consistent, accurate, and reliable over time through regular monitoring of data quality
  • New initiatives begin only after careful consideration of how the initiatives will impact the existing data management infrastructure
  • Data-management tools are standardized across the organization
  • All areas of the organization utilize the standard metadata and rules definitions created and maintained by the data management group
  • Results of data quality audits are continuously inspected, and any variations are resolved immediately
  • Data models capture the business meaning and technical details of all corporate data elements
  • Risk: Low; data is uniform and tightly controlled, allowing the organization to maintain high-quality information about its customers, prospects, inventory, and products
  • Rewards: High; solid, company-wide datamanagement practices can lead to a better understanding of an organization’s current business landscape, allowing management to have full
    confidence in any data-based decisions

Table 4. Characteristics of an “predictive” company

Summary

Achieving the highest level of data management is evolutionary. A company that has not concentrated on the quality of its data cannot expect to progress to the latter stages immediately, because
any improvement in data management involves a number of factors. To improve, you have to change the entire culture of your organization, from personnel to technology to management strategies.
Companies that have achieved significant benefits from improved data know that data maturity is not just a technological approach to understanding and correcting data. It’s also about
implementing sound processes to collect and manage information. The Enterprise Data Management Maturity Model recognizes that an examination of people, processes, and technology identifies ways to
improve data integrity.

Understanding the Enterprise Data Management Maturity Model is the first step to take if your organization is looking to improve its overall data quality. The model helps you understand where your
organization stands in its development and determine what, if any, measures to take to advance to the next stage. From there, you can learn techniques that can maximize the value of data and start
treating data as an important strategic asset that can make your business more competitive.

Share

submit to reddit
Top