CRM professionals constantly tell me that poor customer data integration and management is one of the biggest barriers to getting value from their CRM initiatives. In a recent Forrester survey of over 1,000 North American and European decision-makers, 44% said that data management was a “critical” or “important” priority for 2008. A constant diet of poor quality data is a sure way starve your CRM applications and dim your customer intelligence.

In recent research note, my colleague at Forrester, Ray Wang, points out the right approach to customer data management is elusive. Over the past two decades, CRM professionals have explored a variety of approaches to managing customer master data across the enterprise.

Initial efforts focused on the development of data warehouses that would deliver "real-time" access to customer data. However, data warehouses failed on the "real-time" access, and end users compensated by deploying myriad purpose-built data marts.

Another theory was that CRM applications themselves would provide companies with a 360-degree view of the customer. Instead, multiple instances, disparate enterprise ERP systems, and poor data integration left enterprises with at best a 90-degree view and at worst "yet another view."

Others believed that business intelligence (BI) applications would provide the focal point to customer intelligence across multiple data sources. BI efforts succeeded in showing how poor customer data quality really was — leaving users scrambling to fight a losing battle to keep customer data clean and updated.

In light of these apparent dead-ends, what should you do to improve your organization’s customer intelligence?

•Implement quick wins focused on data quality. The cost and complexity of customer data management is leading organizations to reduce the scope and breadth of their data strategies in the face of economic uncertainties. Narrow your focus and invest in targeted data quality improvement to drive quick wins. Common data quality initiatives include deduplicating customer records within a CRM system or a marketing data warehouse, or implementing real-time address verification services to ensure the collection of high-quality postal address data at the point of capture.

•Pursue rapid implementation methodologies. Some solution vendors are able to deliver full-blown customer data management implementations in less than nine months using templates and tool kits. Work with vendors who have these capabilities or with partners that can lower the risk and complexity of implementation. Although not historically typical, some of the quickest implementations have been completed in less than nine weeks.

•Map the information supply chain to key data elements. The eight steps that comprise the information supply chain (i.e., creation, classification, maintenance, transformation, security, delivery, assessment, and retirement) provide a great starting point to analyze and track data across your information architecture. Conduct a gap analysis that optimizes existing and future states. Keep in mind the data governance implications of future-state policies.

•Keep best-of-breed solutions top of mind. Given the wide range of business requirements and low tolerance for implementation risk, existing customer relationship management applications may not provide the right solution. While traditional CRM solutions should definitely be considered in shortlists, in many cases, best-of-breed vendors can succeed in addressing heterogeneous data requirements and support for multiple sources of data domains across multiple systems with ease and better ROI.