Process management is a requirement for sustaining an environment that consistently produces quality information. Process management goes beyond process control; that is, it consistently produces products whose variation is within defined specifications.

Process management requires a focus on the customers of the products produced by the process. It requires the definition of product specifications that meet end customer requirements, not just the specifications that may or may not meet customer needs.

Process as used here refers not just to single activity processes (input-process-output), but also to the entire value chain from the customer request (for a product) to the customer benefit (customer satisfaction with the product).

The automobile assembly line is a process or value chain. It is the entire assembly line as a whole that must be managed, not just each standalone process or activity. Individual activities may be controlled to meet its specification; but if the product produced does not fit with the components of the next processes or activities, process failure occurs and scrap and rework is required.

Information quality practitioners often think of data or information quality as the focus of their attention without thinking of the customers of that data or of the process activities (both automated and manual) that produce (create and maintain) it. Defective or nonquality data is the symptom of a defective process. To simply fix the defective data is only "information scrap and rework" – not true information quality management. Correcting data only to propagate a downstream database or data warehouse without correcting the data at the source where it is still being used is suboptimized and defective "information scrap and rework." This type of error correction is:

  • Suboptimized because it leaves defective data at the source databases, allowing processes that use the source data to fail.
  • Defective because it introduces a new information quality problem – inconsistency of the data in the target database or data warehouse and the source. (The redundant data cannot be reconciled, aggregations of data cannot be drilled down to its originating source and incremental updates to the target can cause subsequent contamination by the still remaining defective data at the source.)

Defective, nonquality data is a symptom of a "defective" process. There are many precipitating and root causes of defective data, such as:

  • Misentering data at the source.
  • Not having appropriately defined update processes that capture updates to volatile information subject to information quality decay.
  • Environmental causes such as lack of training, inappropriate performance measures or no accountability in management for information quality.

From a process management standpoint, other causes include:

  • Lack of robust information product specifications.
  • Defining information product specifications from the perspective of only the immediate (business-area) requirements, missing potential requirements from all categories of downstream knowledge-workers who require the data.

To manage information quality, one must manage the processes across the information value chain from origination information producer to its most remote information customers. Information process management includes:

  • Identifying an information group, such as customer, product, order or insurance claim.
  • Identifying and defining the information value/cost chain. 1
  • Defining the information product specifications based on all information stakeholders (producers and customers).
  • Defining the data movement specifications (when data must be moved from one data store to another).
  • Managing the processes that create, update and store the data by the information producers.
  • Managing the processes that extract, transform and load or propagate the data.
  • Managing the processes that retrieve, format and present the data to the knowledge-workers.
  • Providing feedback across the value chain for continuous improvement.

Information process management requires:
Process identification: What is our enterprise mission, what objectives must we accomplish, what strategies shall we take and then what processes must we perform?

Customer identification: Who are the information customers? These are the knowledge-workers – both internal and external, such as end customers and regulatory reporting authorities whose requirements must be understood and met.

Supplier identification: Who are the information producers? These are the people who are the natural create point of knowledge about objects and events because they perform activities in which data becomes known. They must have training and processes designed by which to error-proof the collection, gathering or creating of the data. These individuals must also know who their information customers are and their requirements so they understand the importance of the information they create, but may not require themselves to perform their jobs. Information producers require the resources necessary to perform their jobs properly to produce quality information.

Customer product requirement specification: Define the information requirements (i.e., what data must/should we know to accomplish our mission and meet our objectives). Then, define the information quality requirement characteristics (completeness, accuracy, timeliness, etc.) that meet the most rigorous expectations based on risk and cost of process failure. This information requirement is the same team approach that Deming describes in Quality Point 4, when he says purchasing should be a team effort, consisting of "product engineer and representatives of manufacturing, purchasing, sales or whatever other departments will be involved with the product."2 Defining data requires input from all information stakeholders, not just the immediate beneficiaries of an application being developed.

Process definition: Before you can put a process into control, let alone optimize it, the process must be defined so that it is repeatable. To define a process the first time, Masaaki Imai describes the "standardize, do, check, act" cycle as a requirement before processes can be effectively improved.3

Process improvement: To continually meet the ever-increasing quality expectations of knowledge-workers and end customers, a plan-do-check-act process is required to improve processes to the point of optimization.

Process optimization: Processes can be in control, yet not optimized. An organization can have superbly managed business-area activities, yet fail because the business areas activities are suboptimizing the business areas' own work by not working as a team across the value chain.4 To optimize a process, one must evaluate the activities that are cost-adding and eliminate or minimize them. Creating redundant databases or files is cost-adding if data can be shared from a single or replicated database. The interfaces and redundant databases are cost-adding only. This is "muda" (waste) of processing and should be eliminated so that value – not cost – is added "at every step in which a work piece or a piece of information is worked on."5

Management accountability: Managers of activities must be held accountable for the quality of the information created by the information producers who report to them. Managers' accountability is not just to its up-line management, but also horizontally to its information customer peer managers who depend on the information produced by their business area. This is called managerial information stewardship.6

Information quality is the outcome of managed processes. Defective data is the symptom of broken processes. Unnecessary redundant data and transforming interfaces reflect suboptimized processes that contain waste and are sources where errors may be introduced.

The goal of information quality is to eliminate waste of information scrap and rework, unnecessary processing, and to increase business effectiveness by increasing customer satisfaction (both in the organization's products and services, and in the information it provides to internal and external information customers).

By managing the processes across the information value chains, one eliminates unnecessary costs, increases customer satisfaction and, with it, customer lifetime value.

What do you think? Let me know at


1. English, Larry. Improving Data Warehouse and Business Information Quality. New York: John Wiley & Sons, 1999. (See pp. 160-163 for a discussion of IQ Assessment, Step 3 "Identify the Information Value and Cost Chain.")
2. Walton, Mary. The Deming Management Method New York: Perigee Books, 1986. p. 64.
3. Imai, Masaaki. Gemba Kaizen: A Commonsense, Low Cost Approach to Management. New York: McGraw-Hill, 1997. p. 5.
4. Deming, W. E. Out of the Crisis. Cambridge: Massachusetts Institute of Technology Center for Advanced Engineering Study, 1986. p. 63.
5. Imai, M. p. 79.
6. English, L. pp. 406-407.

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