Without quality data, customer relationship management (CRM) will fail time and again. That's because much customer data is dirty, and it must be cleaned before organisations can consider a CRM exercise.
CRM has had its share of hype, failure and revival. Vendors, users, customers and analysts have all wagged their fingers at various points of failure. But the overarching fact is that CRM failed to focus on the customer with the initial deployment of point solutions.
Some vendors blamed a lack of integration that failed to unite sales, service and marketing, while others isolated a lack of integration between front-end and back-end systems. They were correct. Integrating those business components and processes is important. But, no matter how tightly or seamlessly they may be integrated, if the data they contain is faulty, the CRM project will meet with just a measure of success.
Defining CDI
An accurate definition of customer data integration (CDI), from Whatis.com, states that it's "the process of consolidating and managing customer information from all available sources, including contact details, customer valuation data and information gathered through interactions such as direct marketing. Properly conducted, CDI ensures all relevant departments in the company have constant access to the most current and complete view of customer information available. As such, CDI is an essential element of customer relationship management."
CDI is succeeding where CRM failed because it sustains good data quality processes and methodologies. It also unites distributed customer systems into a single version of the truth that can then be cleaned.
The market is alive to this opportunity. Gartner reckons that the market for CDI solutions will exceed $1 billion by 2008. That's a leap up from $680 million last year. In parallel, analyst IDC predicts the master data management (MDM) market will grow to exceed $10 billion by 2009.
But for those predictions to become reality, organisations must understand that using data quality in point efforts, such as for single applications, will result in limited success. CDI must become an integral component of MDM strategies.
Who needs CDI?
Some will argue that operational data stores perform CDI's function and they would be almost correct.
Mervyn Mooi is director at Knowledge Integration Dynamics.
Some will argue that operational data stores perform CDI's function and they would be almost correct. However, one significant difference is that operational data stores, unlike CDI solutions, are meant to be queried, not updated in real-time.
Another possible argument against CDI is that organisations that have conducted enterprise information integration; extraction, transformation and load; and data warehouse projects have a solid foundation for their data management and don't need CDI. But again a key difference arises from employing those tools to conduct CDI-like operations and true CDI.
Warehouses only gather information into a single resource. They pay no heed to the quality of the data. Enterprise information integration simply joins the dots and ensures that information system disparity becomes a thing of the past, again without heed to data quality. ETL comes closest but does not contextualise customer data and present it through a unified resource in real-time.
CDI contextualises data and ascertains how it is related to specific customers. Critically, a single CDI hub in the organisation uses this capability to standardise customer information.
However, with enterprise information integration, warehouses and extract, transform and load tools in place, organisations have a solid foundation on which to base their CDI initiatives.
From an architectural standpoint, organisations with those components need only find a means of inserting the CDI hub between current data quality systems and enterprise applications so that it can work with them.
Returning to the statement that CDI must become an integral component of MDM strategies: CDI manages customer data, which is a component of the total enterprise data. Critically, it draws data from existing systems, contextualises it and ensures it is good quality and can then send it back to repopulate those enterprise systems. It keeps enterprise systems informed and accurate.
* Mervyn Mooi is director at Knowledge Integration Dynamics.
Share