While keynoting a recent customer relationship management (CRM) conference, I noticed a huge advertisement in the Wall Street Journal for a CRM software vendor. It was four full pages of information about the company's "revolutionary" CRM software. The size of the ad reflected the size of the burgeoning CRM market, and the investment required to buy that much space in the Journal reflected the amount of money flowing into this market segment.

The ad maintained that this wonderful software could bring together all the data about customers from the distant reaches of the organization. It didn't matter if the data resided in huge ERP implementations, legacy systems, flat files, desktop spreadsheets or third-party data vendors, this magical software could bring it together into a unified, meaningful whole. The ad copy stated clearly that you didn't need a data warehouse or a data mart to accomplish this goal. Just plug in this software and within a week your disparate, non-cleansed, non-integrated, non-architected customer information would be joined together into a seamless resource of powerful, clean, unified, integrated customer information, ready to unleash its power on your organization. What a compelling message and how strikingly false and misleading.

Unfortunately for the breathless marketers and sales reps, there is no free lunch in CRM. If you are to obtain the goal of an integrated resource of customer information that contains detailed information about every contact point and event with an organization's customers, you must first build an integrated, architected information resource. There is no magic, there is no silver bullet, and there is certainly no vendor who has a turnkey solution that will seamlessly integrate into the typical organization's heterogeneous global customer information systems. The reason there are no easy, turnkey solutions to this challenge is that there are no easy ways to resolve the following:

  1. Semantic discordance. There are very few organizations that have agreed on an enterprise-wide definition of customer, to say nothing of product, sale, channel, cost, etc. Many organizations cannot even reach semantic consensus on such basic terms across departments. Without reaching semantic consensus, documenting all aspects of agreement and disagreement, documenting all the various flavors of customer, product, sale, etc., you cannot integrate information from disparate sources.
  2. Data integration. Different operational systems and third-party data providers format data differently. One system may carry dates in a month/day/year format, while another system may carry dates in a day/month/year format. Different operational systems use different codes for the same data point. One may use a 1 to represent a male, while another uses an M. Without a painstaking examination and analysis of each source data point, you cannot integrate data from the multitude of customer contact point operational systems.
  3. Information quality. If someone could invent a quick, slick, turnkey, five- minute way to address information quality challenges, they would retire very wealthy individuals in a very short time. As Larry English, the world leader in information quality issues and challenges, points out, information quality is tough, demanding work that is ultimately extremely rewarding. Again, it is only through detailed examination of the source data systems and processes that information quality issues are revealed and addressed. You cannot paper over these challenges with wizard-driven dialog boxes.
  4. Meta data. It is impossible to sustain an integrated information system without meta data. This meta data must contain information documenting the source and meaning of all data. It also must contain information about the processes required to achieve semantic integration, data integration and information quality. You cannot replace true systemic meta data with a few comment fields in a front-end analytical system.
  5. Architecture. All of these points add up to an architected system. It doesn't matter if you are taking a bottom-up approach of incremental architected data marts, a top-down approach of an incremental data warehouse or a federated architecture approach of integration of multiple data warehouse, data mart and analytical application systems, you must achieve an architected environment in order to have a system that can effectively and efficiently share information about all the different ways that an enterprise touches and relates to its customers. Without architecture, you have no foundation upon which to pass and exchange data and meta data. You cannot replace real architecture with marketing hype, glossy brochures and expensive advertising.

The CRM "hub" of integrated customer information is the holy grail of the CRM market space. There are dozens of vendors racing toward this goal, with more than a few already claiming to have achieved it. There is lots of hype, but no reality yet, if ever. There will be lots of zeros for the winner; but in the meantime keep in mind that in order to deliver CRM, you must first deliver an architected data warehouse infrastructure.

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