Addicted to Data, But in Quality Denial

ISO 9001:2000 requires the collection and analysis of a lot of data. Organizations rely on this information to make important decisions. However, a new survey by Forrester Research shows most enterprises have a big problem with information quality (IQ). Forrester principal analyst Lou Agosta finds that while many firms have improved their IQ, problems persist. They are “addicted to data”, he writes, “but in quality denial”.

Indeed, company-wide approaches to IQ seem to be the exception rather than the rule. Fully one-fifth of the firms responding to the survey had no consistent IQ strategy, and those that did, tended to have scattered approaches. Almost a third of the respondents dealt with IQ inconsistently, on a localized basis.

Consequently, even though companies are gathering huge quantities of data from business processes and transactions, much of it is obsolete, duplicated, or just plain wrong, subverting its usefulness as actionable information. Moreover, inconsistent and inaccurate information creates business uncertainty, leaving a company swimming in a tide of data, but unable to confidently act.

According to Agosta, it is not enough for companies to sift through data, looking for defects on a case-by-case basis; that’s “firefighting, not information quality improvement”. What firms need to do is to get serious with IQ improvement, and approach it strategically, as an information product quality control issue, in terms of a service-level agreement, or as a commitment to a system design for information quality.

Most importantly, companies have to be willing to take a close look at themselves and establish IQ processes before deploying technology to solve the problem. Enterprises have to create an information quality “safe harbor” that will allow employees to expose IQ problems without retribution, and then take steps to correct them. Agosta observes that a major reason why IQ issues remain unresolved is that employees fear that management will “shoot the messenger”.

Companies need to implement a clear and consistent IQ policy and quantifiable processes across the enterprise and identify and empower an IQ evangelist to lead and coordinate the effort at the highest level. Agosta concludes that the problem, though substantial, can be solved through the application of IQ best practices.

The above portion of this article was based on an article by Matthew Friedman in Enterprise Applications Pipelinehttp://www.enterpriseappspipeline.com>.

The Guidance document on Terminology used in ISO 9001:2000 and ISO 9004:2000 defines “data” as facts, especially numerical facts, collected together for reference or information. In our July 2001 newsletter, we further defined data, information, knowledge, and wisdom.

Data: Factual material such as measurements and statistics, frequently quantifiable.
Information: Data endowed with meaning and purpose, which may differ according to individual interpretation.
Knowledge: Understanding of a science, art, or technique, which is gained from study, experience, or association.
Wisdom: Use of knowledge, with ability to discern inner qualities and relationships and make sound judgments and decisions.

So, what are some of the “dimensions” of data quality?

Accessible: Data is available and retrievable.
Complete: Data is not missing and of sufficient depth and breadth for our work.
Concise: Data is compactly represented.
Consistent: Data is presented in a consistent format.
Friendly: Data is easy to manipulate and apply to different tasks.
Correct: Data is accurate and reliable.
Defined: Data definitions are clear.
Interpretable: Data is in appropriate languages, symbols, and units.
Objective: Data is unbiased and impartial.
Relevant: Data is applicable and helpful
Secure: Data is protected and access is restricted.
Timely: Data is sufficiently up-to-date for our work.
Understandable: Data is easily comprehended.
Value-Added: Data is beneficial and provides advantages from its use.

Information quality is a hot topic. Perform an Internet search on “information quality” and find conferences, courses, papers, and practices on the subject.