Transforming Business Through Data Quality Improvement

1. Create Constancy of Purpose for Data Quality Improvement

Deming’s first point describes the beginning point of a quality environment: “create constancy of purpose for the improvement of product and service.” Constancy of purpose means dedication to
quality as a way of doing business. The purpose of this constancy is to become competitive, stay in business, and provide jobs.

Business and I/S management must balance two sets of problems: those of today and those of tomorrow. Business failure occurs when management focuses too much attention on today’s immediate needs,
such as quarterly profits, at the expense of adequately solving tomorrow’s problem, such as discovering and satisfying customers’ emerging requirements. The pressure of solving today’s
problems-meeting project target dates, converting applications and databases to support the Year 2000, “fixing” application bugs to get operations back up, cleaning data from legacy databases for
the data warehouse -must be balanced with creating tomorrow’s solutions. What good is it that applications are “Year 2000 enabled” if the enterprise goes out of business because it was not able
to develop applications that drive critical new processes? What good is clean data in the data warehouse if it is the wrong data? What good is measuring data quality if it only leads to
finger-pointing, blame and deeper entrenchment into political freedoms of proprietary databases? Deming says no company without a plan for the future will survive. No I/S organization without a
plan that positions the enterprise to meet tomorrow’s knowledge requirements and the ability to rapidly adapt to new ways of performing business processes can-or should be allowed to-survive.

One aspect of constancy of purpose is innovation. Innovation, however, is not innovation for innovation sake. New I/S products and services must not simply be new applications or new technology
platforms. Creating a web site because it is the technology du jour is not innovation.

What is its purpose? How does it move one to accomplish the mission. I/S must improve how its business partners achieve strategic business objectives, and at the same time improve its
end-customers’ lives. The ramification is that data producers may be asked to capture attributes about business events that are needed by downstream knowledge workers. These attributes may not be
needed within the data producers department, but are required to effectively save the end customer.

Constancy of purpose for data quality means that data resource management (DRM) must continually ask, “how are its information products going to capture and deliver knowledge resources that enable the business to achieve its mission, and how will it improve end-customer satisfaction?” It also means application development must continually ask, “how is this application going to support reengineered processes that enable the business to achieve its mission, and how will it improve end-customer satisfaction?” Quality information systems products will be reusable and will require low maintenance. Quality databases will be minimally redundant (except where planned AND managed) and will be reusable by subsequent applications and business areas by simply adding any newly required data and not requiring structural database changes. Quality applications will be non-redundant (except where replacing planned obsolescent applications) and will add value?

To achieve quality innovation, Deming says, top management must have a “declared unshakable commitment to quality and productivity”. Permanent data quality improvement occurs only when senior I/S
and business management recognize two facts. One, the amount of time and money spent fixing today’s problems as the result of non-quality data is unacceptable The second fact is that this wasted
time results from creating short term non-integrated applications and not spending enough time building a stable and flexible information infrastructure that solves tomorrow’s problems.

Information systems productivity is not “how fast can we develop and implement an application. Information systems “quality and productivity”-or “qualitivity”-is how it can deliver stable (does not require a lot of enhancement requests) applications as a result of reusable quality components such as shared data. This requires planning and development of reusable information infrastructure components that enable the enterprise to accomplish its mission and continually satisfy its end customers”.

A critical aspect of constancy of purpose of data quality is that the “obligation to the customer never ceases”. I/S products and services must always be planned, designed, built and implemented
with the downstream knowledge worker and end customer in mind. The only success measure is how well do they satisfy the end customer and meet their needs and expectations. This does NOT mean how
well does it satisfy the immediate users of that application. It DOES mean how well does the product of this application (data) satisfy the downstream knowledge workers needs to satisfy the end
customer.

The ramifications of Deming’s point 1 for data quality are two-fold:

  • Does the enterprise vision and mission include the concepts of information as a strategic business resource that adds value to its products and services for customer satisfaction?
  • Does I/S have a vision and mission that includes developing products and services with success criteria being measured not just by the immediate users of the services, but by downstream
    knowledge workers and end customers?

Guidelines for creating constancy of purpose for the improvement of data product quality and service:

  • Create a vision and a mission statement if your I/S or DRM organization does not have one
  • Assure that the mission statement is directly driven by (not just linked to) the enterprise vision and mission
  • Assure that the I/S and DRM objectives have measures that illustrate how they contribute to the long-term enterprise objectives
  • Assure that the mission statement focuses on your customers (the downstream business beneficiaries and not just the immediate beneficiaries of I/S products and services
  • Assure the mission statement focuses on the enterprise customers (the ultimate beneficiaries of I/S products and services
  • Develop plans that are both strategic and long range, and that can be delivered quickly, incrementally and are reusable
  • Measure the reuse of I/S and DRM components as a measure of quality against the development of redundant and non-value adding interfaces
  • Assure that everyone in I/S and DRM subscribes to both the business and the I/S and DRM vision statements
  • Once you have done this, make constancy real by reading and contemplating your vision and mission statements daily for the first five minutes of the day, until your mission becomes the
    subconscious driver of your daily activities
  • Keep asking yourself, “is what I am doing or about to do, moving me closer to achieving our vision and mission?” Try this daily for a month, and see if it does not help you achieve a
    constancy of purpose for data quality.

2. Adopt the New Philosophy of Quality Data
Quality point 2 extends point 1 “Create constancy of purpose for the improvement of data quality”: “Adopt the new philosophy of quality data.” Deming admonishes that quality must become the new
religion. This is not religion of quality for quality sake, but a passion for continually delighting the customer. This quality “religion” demands behavior changes. The new economic era has
re-defined the rules for successfully competing in the global marketplace. Quality is no longer optional. Consumers have new expectations based on the availability of reliable products and
services. The new economics demand new standards-not new data naming conventions, but new standards of data quality.

“Reliable service reduces costs,” Deming says. Mistakes, rework and delays are what raise costs. Philip Crosby in Quality is Free reconfirms this. The “unquality things,” such as doing
things over, around, or instead of because of non-quality are the things that increase costs.

Quality point 2 really means a transformation of management, according to Deming. Management must dismantle the organizational structures that have created barriers to quality, and caused
inefficiency in performing business processes.

Data quality point 2 likewise means data quality is no longer optional. Every data warehouse project reconfirms this. The requirement for data quality is driven by the fact that business can no
longer afford the luxury of the costs and problems caused by poor quality data. Every hour the business spends hunting for missing data, correcting inaccurate data, working around data problems,
scrambling to assemble information across dis-integrated databases, resolving data-related customer complaints, etc., is an hour of cost only, passed on in higher prices to the customer.
That hour is not available to add value.

Two information product and service facts are also clear:

1. Reliable data management reduces information systems costs

Unfortunately, conventional wisdom disagrees. Data management is often perceived as adding costs to an application. To be sure, a data management function that only adds costs to
information systems should be eliminated and replaced with an effective value-adding data management program. Data management is not simply developing application-specific data
models and defining application data. Quality data management defines and models data that is reusable throughout the enterprise. Quality data management is measured by how much of its defined data
and databases are reused and shared. By eliminating the need for redundant applications creating data redundantly, redundant databases, and unnecessary transforming interfaces, quality data
management reduces costs of applications development and maintenance, as well as the costs to fix problems caused by inconsistent redundant data.

2. Reliable data reduces business costs

Quality data likewise reduces business costs by eliminating the costs of scrap and rework caused by inaccurate or missing data. But more significantly, quality data eliminates missed and
lost business opportunity due to poor customer caused by non-quality data. When customers receive poor service because of problems, they may not complain to you. They simply go elsewhere, and take
with them their customer lifetime value. If all companies provide the same level of (non)quality, they may simply trade unhappy customers. Even this has costs. It costs 4-5 times as much to gain a
new customer as to retain a happy one. But, when someone raises the quality bar, the rules change for everybody.

Ramifications of DQ point 2 for information systems management:

  • The purpose of applications is not to automate processes. Applications must capture, maintain, and deliver the knowledge of the enterprise.
  • Data, as a product of information systems, demands new standards of excellence to reduce the costs of non-quality.
  • Point 2 means a transformation of information systems management. In 1980, James Martin asserted that one should not view database as a grandiose project, but rather as a change in the
    management of data processing. Some 17 years later many I/S organizations still do not understand this. The I/S mission is not to deliver automation solutions to business areas. I/S must deliver
    solutions that solve boundary problems between business areas. This is where non-quality data kills the effectiveness and the efficiency of the enterprise. When I/S solves these problems it has
    enables information as a strategic enterprise resource. Without this, I/S cannot transform their enterprise into a realized Information-Age organization.

Guidelines for adopting the new philosophy of quality data product and service:

  • Revisit your information management mission. Is it truly visionary and bold? If not, read last month’s column, “Creating Constancy of Purpose for Data Quality”, and re-think it.
  • DQ point 2 means internalizing data quality as a regular habit-not a fire-fighting reaction when problems occur.
  • Count the number of transforming interface programs (omit managed replication programs and extract programs that retain the same data definition and domain values) and redundant applications
    and databases that capture data about the same or similar data types. The truth horrifies most management: 43 redundant applications and files maintaining a “key data element”; 92 different part
    files, many with different primary keys with no ability to integrate them; 175 customer files-requiring 6 people 4 months to answer the question, “who is our best customer?”; and one organization
    discovered 400 brand files!!! Do not gloss over this. Ignoring this can cripple the enterprise.
  • Now, estimate the costs of developing and maintaining a transforming interface program, create program, and database file over their respective lives. Multiply this times the number of
    redundant occurrences to get the cost of redundancy.
  • Estimate the amount of time I/S spends in resolving problems due to inconsistent data to get the cost of redundancy “scrap and rework” (non-quality of the redundancy).
  • Add these two figures together. This represents only a portion of the costs of non-quality data in the enterprise. If you are happy with the sum as a percentage of I/S budget, please e-mail or
    call me about your best practices.
  • If you are unhappy with the results, develop a plan to make fundamental changes in your I/S organization, including:
  • A strategic data management program that provides leadership and education.
  • Data model development by teams of business subject matter experts, not simply the primary beneficiaries of an immediate application. It can-and must-be developed rapidly by involving the
    right people with the right charter.
  • Data models developed in three levels of abstraction: 1) Enterprise-wide owners view readable by senior business management; 2) Detailed conceptual data models by business resource (subject)
    supporting cross-functional business value chains; and 3) physical databases measured by stability, flexibility, and reuse criteria.
  • Application development around value chains measured by data reuse and non-redundancy.
  • Information policy that articulates the enterprise values for information as a strategic business resource, business accountability for data quality and acceptable behaviors for stewardship
    of information. Senior management issues this policy.
  • An inventory of critical information redundancy and a plan to eliminate unnecessary redundancy and maximize quality shared information.
  • A data quality improvement process whose objective is to increase business effectiveness and decrease non-quality data costs by measuring data quality, identifying defective processes,
    facilitating root-cause analysis, and providing education in data quality principles.

Adopting the new philosophy of data quality does not mean saying one believes in it and creating slogans. It means acting those beliefs and changing behavior to make it happen.

What do you think? Send your comments to LEnglish@infoimpact.com or through his Web site at www.infoimpact.com.

Share

submit to reddit

About Larry English

Larry, President and Principal of INFORMATION IMPACT International Inc., is one of the most highly respected authorities in the world on how to apply information quality management principles to Total Information Quality Management. He has provided consulting and education in more than 40 countries on six continents.Ê

Larry was featured as one of the Ò21 Voices for the 21st CenturyÓ in the American Society for QualityÕs Journal Quality Progress in its January 2000 issue. Heartbeat of America, hosted by William Shatner, awarded him the ÒKeeping America StrongÓ award in December 2008 honoring his work in helping organizations eliminate the high costs of business process failure that enable them to eliminate the high costs of business process failure caused by poor quality information. Larry was honored by the MIT Information Quality Program for a Decade of Outstanding Contributions to the field of Information Quality Management in July 2009. You may contact him by email at Larry.English@infoimpact.com.


Top