The purpose of this white paper is to outline the importance of data quality with reference to single view of customer (SVC). In any organization SVC is the foundation of successful customer relationship management (CRM) across financial institutions, utility companies, telecommunications service providers and retail outfits. The paper defines the breath of the issue and how poor data quality can hamper, delay and even defeat organizations in their attempts to implement CRM and SVC. It describes a typical solution to the problem of poor quality customer data based on Informatica Data Quality enterprise data quality management software. The white paper goes through the steps that should be taken to ensure data quality issues are eradicated. It is the opinion of the author that SVC can only be achieved with reference to the state of the underlying data quality. The best way to achieve the necessary levels of data integrity, conformity and consistency is through the implementation of an end-to-end data quality process.
Introduction
Single View of Customer (SVC) is the lynchpin of effective customer relationship management (CRM). With reliable SVC a CRM system can really deliver on its promise enabling your business to profit from understanding and anticipating the needs of current and potential customers. Yet even in organizations that have invested significantly in CRM initiatives, high quality and reliable SVC often remains an elusive goal. The root cause of this is data quality. Low quality customer identification and description data is pervasive in most organizations and is the major stumbling block in the way of achieving SVC.
Because CRM and SVC are so central to an organization's ability to do business, everyone who uses the system, from sales force and call center personnel to executive-level professionals and marketing teams, must be able rely on the quality of the data it contains. Without confidence in the currency and quality of CRM information these systems fall into disuse and ultimately fail. Successful organizations understand the direct link between data quality and business performance, and that information-intensive applications such as CRM can only achieve results if the data they depend on is reliable, complete and accurate. Analyst firm Gartner says data quality has emerged as the number one cause of CRM project failure.
The longer the problem is left to fester the worse it will get. According to PricewaterhouseCoopers customer data degrades at a rate of two percent a month, or nearly 30 percent per annum. Proactive efforts to identify and overcome data quality issues are vital for SVC and CRM applications to succeed. SVC implementation should start by identifying data quality problems, such as missing, non-standard or inconsistent data, and correcting such problems.
But achieving the high levels of data quality needed for SVC and CRM to succeed is more than a one-off exercise. The only way to ensure that accurate, consistent and timely SVC data is delivered into the future is through an ongoing data quality management program. Data quality levels need to be tracked and assessed against targets on a regular basis. New data should be cleansed and checked against business rules as it goes into the system to ensure the highest possible levels of data quality are maintained.
An end-to-end data quality management process such as that outlined in this white paper will enable any organization to rapidly identify and correct data problems and to prevent the erosion of data quality levels over time.
The four stages necessary to achieve this include Profiling, Standardization, Matching and Consolidation.
- Profiling Investigates the content of key data fields to Identify and quantify data quality problems
- Standardization Extracts, cleanses and standardizes key customer identification and description data
- Matching Matches selected customer identification and description data from each customer record to identify similar or related entities
- Consolidation Generates a final master table of unique customer records and key tables linking master data with source systems and transaction tables
Tools For The Job
On large datasets it would be impossible to carry out all stages of a data quality management process manually. A powerful and flexible tool capable of implementing customizable business rules across any type of data is essential to carry out the task.
