CDI Comes in All Shapes and Sizes

HOLLYWOOD, FLA. -- There is no "one size fits all" when it comes to customer data integration (CDI), and requirements differ depending upon the type of industry, customer, or organizational structure your business operates in or caters to, according to a presentation at the Gartner MDM Summit here today. "Simply linking your ERP and CRM systems via CDI might work for certain companies, but for others it could involve an entire realm of enterprise applications," said John Radcliffe, research vice president at Gartner, during his presentation, "CDI Is Not One Size Fits All and It Will Keep Evolving." "It's a case-by-case procedure," he added. When defining their company's CDI strategy and goals, and when determining which systems to integrate with a CDI hub, businesses should consider both the verticals they belong to and the business environments they operate in, Radcliffe said. He cited a U.S.-based retail client with over 80 million customers that needed to consolidate customer records from its Web site, order management system, and credit card/transaction database to effectively gain a single view of each of its customers. "If you're in manufacturing, you could be considering an entirely different solution, such as supply chain management," he said. "In hospitality, you're talking the reservation system." Companies also need to consider the type of strategy, or architecture, that their business should take when deploying CDI, Radcliffe said, before identifying three such strategies:

• Registry Style: This CDI hub is simply acting as a "data traffic cop," directing data from application-to-application or from application-to-data-source. This is generally the least complicated CDI implementation, and is a good starting point.
• Transaction Style: In this case, the CDI hub is pulling data from a predefined "golden record" and pushing that data out to enterprise applications at the time that end users request it.
• Hybrid Style: A combination of both the Registry style and Transaction style. This is generally the most complicated, and also the most rewarding, but requires the most work and resource investment on the part of the organization.

To determine which style to pursue, Radcliffe told the audience, you "must review your organization's capabilities in leveraging a single view of the customer, and create a shared business vision that is driven by the business case, and only underpinned by IT." In fact, Radcliffe stressed, ensuring that technology plays the proper -- and limited -- role is essential: "CDI, just like MDM and data quality, is not a technology issue," he warned. "It is a business problem, and everybody from C-level executives down must be made aware of it." To that end, Radcliffe also emphasized the importance of data quality as another critical factor in CDI success. "If you don't have the data quality practices in place, all you're doing with CDI is efficiently moving garbage from one enterprise application to the next," he said. With sales of CDI software and systems becoming more commonplace, and with a number of success stories and best practices emerging, Radcliffe also discussd the vendor market for CDI hubs. The market itself is divided among a series of camps:

• the application suite and platform providers, such as Oracle and SAP;
• specialty providers, vendors such as Siperian, Initiate Systems, and Purisma;
• application platform vendors such as IBM; and
• data quality and BI vendors such as Business Objects and SAS Institute's DataFlux.

While all have their advantages and disadvantages, the key takeaway is that none of these CDI providers tackles the entire master data management (MDM) arena, though they're all moving in this direction. "None of them offer what I like to call 'the Swiss Army knife' MDM solution. They're either focused entirely on CDI or offer CDI and parts of MDM," Radcliffe said, referring to CDI's overarching big brother."But they're all moving toward MDM because the market is pushing that way. While most businesses start their MDM initiatives by tackling CDI, they're realizing they're eventually going to have to go after MDM, so vendors are responding accordingly." Radcliffe also said that, overall, the CDI market is still a maturing one, and "not yet at the plateau of opportunity." Addressing attendees' vendor-selection needs, Radcliffe suggested that customers should "choose a best-of-breed provider if the situation fits, but weigh [that] against the future consolidation that always takes place in a maturing market." One consequence of a still-maturing market, he added, is that it grows at an explosive rate, leaving service providers and vendors alike with a shortage of skilled labor. "This happens to every market that suddenly becomes hot, so be careful you're not working with somebody that's receiving their on-the-job training, if you know what I mean," Radcliffe warned. "Be sure to carefully evaluate all the players and service providers you're working with, because CDI can be expensive," he added, citing the $1 million average price tag of CDI software -- and the fact that implementation and service-provider costs often run double or triple that amount. "Most importantly, do a sound job of identifying your business needs, the industry you operate in, and your customer data," he told the crowd. "Those are going to be the factors that drive a successful CDI implementation."

Related articles: Feature: Dirty Little Data Secrets CRM's real truth requires enterprises to clean up -- reorganize -- customer information with data integration solutions, Web-services technology that integrates data applications. Feature: The 2007 Market Awards: Data Quality Customer data is still king, but data quality is expanding to include other forms, such as financial and product data, along with master data management (MDM) solutions. DataFlux Accelerates the Path to MDM and CDI Version 8 of the company's flagship product introduces new, prebuilt Accelerators that assist customers in overcoming common data challenges associated with making the transition to MDM and CDI. MDM for Companies on the Go Purisma's new Data Hub product focuses on improving data accuracy while reducing implementation times, which one analyst says increases these types of solutions' value. Viewpoint: The Next Generation of CDI CDI must evolve to deliver customer information wherever and whenever it is needed. Fanning the CDI Flames New data management solutions and best practices will spur enterprises to switch from simple data warehouses to reactive CDI data marts. IBM Tops Gartner's 2006 CDI Hub Magic Quadrant Big Blue becomes the first vendor to land in the leaders circle of a CDI hub quadrant, but must improve its ability to execute. Informatica Joins Leaders in Gartner's 2007 Data Quality Magic Quadrant Others in the top category are DataFlux, Business Objects, Harte-Hanks Trillium Software, and IBM; expect to see increased M&A activity as smaller, international players make their presence felt.

Related Articles

If MDM's So Great, Why Aren't Companies Mastering It?

Two recent reports throw a bit of cold water on master data management, despite significant benefits and business process gains.

DataFlux Has the Edge in Gartner's Magic Quadrant on Data Quality Tools

Magic Quadrant for Data Quality Tools '08: Business Objects, IBM, Informatica, and Harte-Hanks' Trillium Software join the SAS Institute unit in the top box.