Data Warehouse Quality

Every time I put the book back on the shelf, I wind up taking it down a day later to refer to it. Every chapter is jammed full of value. For those of us who are responsible for information quality initiatives, chapter 7, Measuring Nonquality Information Costs, is a must read. If you want to know how defective your data is, then read chapter 6, Information Quality Assessment, to learn how to measure and quantify your information problems. Concerned about the cultural issues in your organization? Chapter 13, Implementing an Information Quality Improvement Environment, will get you started in the right direction. If you are a "knowledge worker," supervisor, CIO, or on the business or IT side, this book is for you.

This is an important and unique work that addresses a big problem: data quality. Why is this a problem? Data warehouses are proliferating at a dizzying rate. Since data warehouses are fed by production databases, many of which are legacy systems, the poor quality of existing data quickly becomes [painfully] apparent. I spent the last half of 2000 bringing data warehouses into production and can attest to this sorry fact. However, the author drives home this point in chapter 1, titled "High Costs of Low-Quality Data" by giving nearly three pages of eye-opening examples from real life. This alone should inspire anyone responsible for data integrity or quality, or who uses data to carefully read this book.The big question is "what is quality"? Specifically, "what is information quality"? Answers to these basic questions are given early in the book, and sets the tone for what follows. The foundation of data quality is carefully built by how the author applies quality principles to information, which segues into a chapter on improving information quality. It quickly becomes obvious that Mr. English is a Deming fan - although I am more in the Juran camp, I like the way that the author places data and information quality into a recognizable framework.Things get interesting in the chapters on assessing data and information quality. The two chapters devoted to this subject are strengthened by the chapter on measuring the costs of non quality. This is a great foundation for a business case for data and information quality improvement, which can be expensive.The rest of the book is a step-by-step approach to getting data quality under control using data reengineering and cleansing; proactive measures for data defect prevention, and how to establish an information quality environment. Although I found every chapter to be both informative and thought provoking, I particularly liked the concept of information stewardship (this goes far in aligning IT and business, and places roles and responsibilities where they belong), and the chapter on implementing a quality improvement environment. This is especially valuable because it clearly outlines the critical success factors and steps needed to get there.Who should read this book? Obviously DBAs, data architects and anyone else responsible for designing and implementing data warehouses. It should also be read by key business process owners because they, after all, own the data (or should) and depend on it as the basis for information. In fact, Mr. English's approach and writing make this book highly accessible to non-technical readers, which is probably the book's most valuable aspect. I personally believe that this book is the best on the subject and strongly recommend it.

The chapters on estimating your organization's cost of having non-quality data make this book a requirement for every Information Resource Management professional's book shelf.Larry English is one of the few authors in the industry to show you how to work the numbers when it comes to data quality. While others drop hints at why quality is good, Mr. English tells us how to collect the data and how to calculate the true costs of sub-standard data management practices.As our profession matures we will be called upon to make more explicit business cases for what we recommend to senior management -- Larry has some great methods for us to use. This will save us months of having to figure it out for ourselves.

The book is fantastic. English obviously has plenty of front line experience. He doesn't simply state the problem and offer suggestions. He empowers the reader to join the data quality movements by giving them the tools necessary to convince the decision makers that data quality is worth the investment. Chapter Seven, "Measuring Nonquality Information Costs," is worth the price of the book alone, because it gives us a solid ROI model to throw at the bean counters. The writing style is extremely accessable. The book is so well organized that it can be read straight through or employed as a reference.

This book takes the reader from understanding and applying principles of Quality Management to a step-by-step guide for implementing a quality improvement program.I liked the organization and writing style of this well-referenced text. I found the diagrams and tables very helpful.In the introduction, the author states that the text is a "concept book, a textbook, a reference book, and a practitioner's guide." At first I didn't believe it. Now that I've finished the text, I believe every word.When I read the "Author's Warranty" (and how many books do you know come with a warranty?), I had an inkling that this would be a good book.I was wrong. This is a GREAT book!