New Book Review: "Customer Data Integration"
New book review for Customer Data Integration: Reaching a Single Version of the Truth (SAS Institute Inc), by Jil Dyche and Evan Levy, Wiley, 2006, reposted here:
After discussing customer relationship management (CRM), master data management (MDM), and CDI, the authors discuss the challenges of data integration and the far-reaching effects of inaccurate, bad, and missing data before presenting their recommended CDI development "framework", which is essentially a CDI-focused project methodology. In subsequent chapters, the authors discuss a number of other related topics, including data governance, data management, and data stewardship, followed by presentations on different types of CDI, making the case for CDI, and how to move forward with CDI without making common mistakes. The authors also provide a glossary of terms following the concluding chapter, but some of the terse definitions will likely leave the reader wanting more information and therefore searching the body of the text for better explanations. In some cases these explanations can be found, but the executives and managers that are the target of this text need to be discerning as they separate the wheat from the chaff, as the critical content of this book is better suited for a whitepaper less than one-fourth the size.
The authors state in their introduction that the goals of this book are to (1) define customer data integration (CDI) and to (2) describe "how" CDI works, but shortly afterward indicate that this book actually concerns the "whys". In recalling my reading a few years ago of "Rethink: A Business Manifesto for Cutting Costs and Boosting Innovation" by Ric Merrifield (see my review), I could not help draw a comparison with the goal of that author to focus on the "whats" of business processes, only to succumb to a dominating discussion of the "hows" in which these goals are accomplished. However, I think the authors of this book actually provide a mix of the "whats" and the "whys", with some degree of the "hows", even though the "hows" are kept at a very high level and would be typically more complex from a technical standpoint. From the perspective of the authors, addressing the "whys" is important because CDI justification is made more accessible to executives and managers.
As a consultant, what initially drew me to this book is the style and simplicity of some of the line drawings that were provided in chapter previews, and after a read I still think this is one of its primary strengths, with Chapter 7 ("Making Customer Data Work") especially well done as it discusses the CDI "hub", which acts as an intermediary to retrieve customer data from source systems on behalf of client applications. Unfortunately, Chapter 3 ("Challenges of Data Integration") resorts to diagrams that are reminiscent of those seen in enterprise service bus (ESB) whitepapers of the past decade in which chaotic data infrastructures are magically transformed into orderly spaces without really explaining how they work (keeping in mind that describing the "hows" is the second goal of the authors). Apart from the effective diagrams that are provided, the "Manager Do's and Don'ts" sidebars which conclude each chapter are probably the best part of this book, and are recommended reading for those seeking the pursuit of data integration efforts either for the first time or after one or more failed attempts.
From Chapter 1 ("Executives Flying Blind"): "Don't think that reengineering business processes will fix the problem. Even the tightest sales process or call center script is only as good as the data available to the people putting it into practice". From Chapter 2 ("Master Data Management and Customer Data Integration Defined"): "Don't ignore difficult people or departments when doing your CDI research. The need for an authoritative source of customer data comes from all corners of the company. People or departments that feel slighted might try their own initiative and become the saboteurs". From Chapter 6 ("Who Owns the Data Anyway? Data Governance, Data Management, and Data Stewardship"): "Don't succumb to 'warm body syndrome'. Just because someone has been with the company a long time ('She knows how our business runs') or has worked on another project that has involved customer data doesn't mean that she is qualified to participate in a data governance council. Be sure to vet data stewards as knowledgeable about data and ready to deliver. Data stewards ideally come from inside the business. However, they need to be willing to practice formal and rigorous data management tasks, and be accountable for doing them. Having a deep understanding of specialized data because of regular usage or historical exposure is the best job qualification."
While this book is from SAS, one if its strengths is that it is vendor neutral in a similar fashion as "Enterprise Master Data Management: An SOA Approach to Managing Core Information" from IBM (see my review). If you do not have time to read the entire book, the following sections are recommended reading: "Data – Always the Bridesmaid" and "Five Mainstay Challenges of Data Integration" in Chapter 3 ("Challenges of Data Integration"), "Preparing for Data Quality: Some Advice from the Trenches" in Chapter 4 ("'Our Data Sucks!': The (Not So Little) Secret About Bad Data"), "The Truth About Managing Data as an Asset" in Chapter 6 ("Who Owns the Data Anyway? Data Governance, Data Management, and Data Stewardship"), and "Keeping the Saboteurs at Bay" and "Common Arguments against CDI" in Chapter 8 ("Making the Case for Customer Data Integration").