This is the seventeenth in a series of discussions of quality guru W. Edwards Deming's fourteen points of quality and their ramifications for data quality. Here I describe Deming's quality point 12, "Remove Barriers to Pride of Workmanship," a requirement for enabling sustainable data quality improvement A major pitfall of the Industrial Age was to take pride out of workmanship. During the agricultural age, a cottage-industry craftsman made his products from start to finish. This is "my work." But the principle of specialization of labor forced workers to see only part of the work product, removing the feeling of accomplishment. When Volvo reengineered its manufacturing process, rather than have each person perform a small set of tedious tasks, such as bolting the left hand doors to the car, a team of workers would follow an automobile down the assembly line and as a team perform all of the assembly. Three things happened. Morale went up ­ "I built this car!" Productivity went up due to less absenteeism and higher enjoyment of work. And quality went up ­ the pride of workmanship was back. When a person identifies with their work product in its whole, there is a sense of ownership and a desire to do quality work. The result is that both quality and productivity go up.

The essence of Deming's point 12, "Remove barriers that rob people of pride of workmanship," is simple. Allow people to feel good about their work, and they will do good work. Period. Barriers and other obstacles in organization structure, lack of training and lack of feedback rob the worker of "his birthright, the right to be proud of his work, the right to do a good job." 1

Information age technology gives knowledge workers a way to overcome the separation and disjointed work environment caused by specialization of labor. By collaborating in their work through shared databases, information producers and knowledge workers do not just do their isolated job, they share in the entire process and the finished product.

But while information technology enables fewer people to do more work, management must not view people as expendable commodities. People are, and will always be, the most important asset in an organization. And while money is important, what really motivates employees "to perform ­ and to perform at higher levels ­ is the thoughtful, personal kind of recognition that signifies true appreciation for a job well done," Bob Nelson asserts with confirmation by numerous studies.2

Information producers, by and large, want to do a good job. Give them an opportunity to "work with pride, and the three percent that apparently don't care will erode itself by peer pressure." 3

Guidelines for removing data production barriers:

  • Train information producers how to capture data accurately.
  • Let information producers know who uses the data they create and the processes that require it and, therefore, the importance of their information products. Have knowledge workers swap jobs with the information producers who create the information they use. Likewise, have the information producers go to the knowledge worker areas to observe what is done with the data. Have the information producers perform the exception processes if data is missing or inaccurate.
  • Remove quotas or incentives based on speed or volume of data created. "Money and time spent for training will be ineffective unless inhibitors to good work are removed."4
  • Prototype human-factor design of applications with information producers so they provide feedback as to its intuitiveness, ergonomics and ease of use. Minimize unnecessary keystrokes and hand and eye movements.
  • Implement business rules that prevent inadvertent errors. Minimize data entry of data that can be selected from a list of values or that the application can access and provide.
  • Do not implement business rules that force errors. For example, if an edit requires a birth date when the information producer has no way to get it, they will have to make one up. Rather, allow for a value of "unknown."
  • Eliminate redundant data creation and private, proprietary databases. If data cannot be physically shared from one database, provide for managed replication of the data and control its synchronization based on the knowledge workers' real requirements.
  • Identify where information producers must do workarounds to capture required data because of poor database design and unacceptable costs or time to modify the database and application. Eliminate unmanaged redundant databases and applications to free up information systems professionals to perform value-adding work on sharable databases and reusable applications.
  • Encourage information producers to identify where problems exist in the application design, data design or in data creation and update procedures. Accompany this with fast response and feedback to their suggestions.
  • Involve information producers in application design and data definition.
  • Develop clear and complete data definition including complete lists of valid domain values and business rules.
  • Have management perform the data production work so they know the data production job and the working conditions. When salaried workers took over production during a strike, the managers found horrible working conditions that included machines that were out of order and badly in need of maintenance. After the strike, they instituted a system that enabled employees to easily report problems with machines, materials and other problems along with the mechanisms for responding to the reports.5
  • Assure that the information producers are the natural point of data production for the data being created or maintained. If not, move the data production process to the best place in the enterprise. The right "place" is the first business process that comes to know the data.
  • Develop processes that automatically capture data electronically rather than have it entered by hand. For example, use bar code readers, optical scanners and chips that electronically sense and capture data. This frees the information producers to have time to capture data that cannot be automatically captured.
  • Provide feedback frequently to information producers, not just when data is bad but also when the data enables significant business success.
  • Recognize and reward information producers for improving information quality that eliminates the costs of poor quality data downstream.

What do you think? Send your comments to Larry.English@ or through his Web site at

1 Deming. Out of the Crisis. P. 77.
2 Nelson, Bob. 1001 Ways to Reward Employees. New York. Workman Publishing. P. XV.
3 Deming. Out of the Crisis. P. 85.
4 Deming. Out of the Crisis. P. 53.
5 Deming. Out of the Crisis. P. 85.

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