FREE DM Review Site Registration!
Sign-up today and access DM Review on the Web!

Your FREE registration entitles you to:
FREE email newsletters

FREE access to all DM Review content

FREE access to web seminars, resource portals, our white paper library and more!

   

Publisher reserves the right to serve qualified requesters only.

Making the Case for CDI

It is an article of faith among data warehouse developers that data mandates process. There are consulting firms - including my own - that form entire practices around data management, and emerging concepts like the data supply chain liken the life cycle of data to the manufacturing process, comparing the various development stages data undergoes, the tools employed and the fact that there's a de facto assembly line for data in companies, much like for tangible goods.

But explain all this to an attendee at a CRM or an ERP conference and you may get a blank stare or, worse, a dismissive roll of the eyes. You're one of those data zealots, and they'd much prefer to simply get back to developing their customer dashboard, thank you very much. After all, that's what the users really want.

CRM practitioners, for instance, are gradually learning that data isn't an afterthought. As operational CRM systems enter the mainstream, executives are turning their attention to analytical CRM. With the encouragement of CRM vendors, many of whom are watching revenues for their core products wane, managers are exploring how to leverage the newly generated information from their sales force automation and campaign management systems, among others, to reach new levels of customer understanding and increase customer loyalty.

But as data warehouse practitioners have known for a long time, developing analytical systems is a vastly different undertaking than building operational ones and the distinctions can mean the difference between successful business intelligence and a scrapped project.

Everything Old is New Again

Enter the new era of customer data integration. In many ways CDI is a natural outgrowth of several established disciplines, largely IT-based, but with far-reaching business impact. They are:

  • Data Integration
  • Data Warehousing
  • Service-Oriented Architecture (SOA)
  • ETL (Extraction, Transformation and Loading)
  • Data Quality
  • Business Intelligence

Here's a formal definition of CDI:

Customer Data Integration (CDI) is the collection of processes, controls, automation and skills necessary to standardize and integrate customer data originating from different sources to support a variety of business initiatives.

A core tenet of CDI is the creation of what's usually referred to as a "360-degree view," but in CDI circles has come to be known as a "customer data hub." Said hub is really an integrated customer database that incorporates automation of data quality, correction and correlation prior to physically storing the customer data in the hub's database. The customer data hub can also be known as the "master customer reference database," or even simply the "one true view."

Before you dismiss CDI as just a vendor-driven ploy to repackage and sell more software, consider this:

  • Your sales executives can't intelligently assign territories because no one knows the difference between a "parent company" and a "billing entity."
  • Your company's marketing department claims CRM success but still spends almost a hundred thousand dollars a year on duplicate postage.
  • Your sales partners are cannibalizing each other's prospecting activities and their (your) customers are confused.
  • Your company just acquired a competitor, and someone needs to reconcile customers across accounts and addresses.
  • Your call center representatives are assigning redundant trouble tickets because they're not sure who the right customer is.
  • You just received a letter thanking you for your purchase from a company you've never done business with.
  • No one is using the (multimillion dollar) CRM system anymore. They just don't trust the data.

These are all symptoms of a larger problem of customer data - usually originating from multiple places - that has not been well integrated. The bad news is that most CRM vendors didn't see it coming or, worse, over-hyped their technologies to include claims of a "single version of the truth" where there wasn't one.

The good news is that CDI is filling a market need. Many companies have realized that their initial customer-focused business initiatives were only band-aids covering a much larger wound. CDI promises a sustainable remedy, and not a moment too soon.

Data Integration: Still a Big Deal

You'd think they would have solved it by now. After all, it's not like data integration is new to companies. Since the advent of data warehousing in the 1980s, and even before then, IT departments have rolled up their sleeves to reconcile data from multiple sources in an effort to provide business people with meaningful information from their production systems that could be strategically useful.

Trouble is, the combination of new operational systems and the increasing velocity of data entering the company has meant that data sources are multiplying faster than most companies' ability to harness and control that data. An interesting study by Lyman and Varian at U.C. Berkeley in 2003 found that 5 exabytes of data were produced in 2002. That's the equivalent of 800 gigabytes of data for every person in the world! Many companies I work with don't have 800 gigabytes of customer data (yet, anyway).

The point here is that those of us in the data warehouse and BI space have understood this phenomenon for a while now, but many in the CRM market are just now realizing that they've underestimated "the data problem." Users won't use the new CRM tool because they don't trust the data. Data latency in the CRM system renders meaningful analysis unlikely or impractical. The standard crop of canned reports accompanying the CRM package aren't enough, and anyway they aren't accurate. The lack of rigor around data management renders many development organizations inert, looking for an entry point or a pretext for connecting and controlling their data in a more effective way.

The new crop of CDI products from companies like Purisma and Siperian bundle the various components of data integration. Likewise, established players like Siebel and DataFlux are leveraging existing functions and architectures to play more fully in the CDI space. In either case, data integration services, as shown in Figure 1, are now a set of discrete tasks that fit into a larger CDI implementation framework and are often complex in their own right. Most CDI tools automate these tasks in a seamless and repeatable way.

Figure 1: Data Integration as a Discrete Set of Steps

As important as automating data integration steps becomes, adopting CDI can force a different conversation: one of organizational ownership. CDI can thrust the issue of formal data management roles and responsibilities into the spotlight. Including data cleansing, matching and transformation functionality all under a single umbrella of effort, for instance, enforces a process that may heretofore not exist. Moreover, it also fosters a dialog about who in the enterprise owns these disparate tasks for the long term.

Indeed CDI can take even a best-practice company to new capability levels by offering more straightforward choices when it comes to assigning discrete roles and responsibilities around data management - something that even evolved IT departments have yet to tackle. Having CDI means designating a customer data steward whose role is well defined and who is accountable across organizations for meaningful and usable customer data. It also means assigning business subject matter experts who understand the business rules around customer data is no longer a luxury but a mandate.

In their April 2005 Harvard Business Review article, "The Quest for Customer Focus," authors Ranjay Gulati and James B. Oldroyd state, "Gathering, standardizing and organizing customer information that comes from all across the organization requires companies to establish a coordination infrastructure."

As I've advocated elsewhere, assigning discrete data management roles begins to establish a set of behaviors that reinforce a consciousness around data as less of a project enabler, and more as a corporate asset, used by a cross-section of the company, its value quantified, its awareness reaching the executive boardroom.

The resulting efficiencies and economies of scale can translate into an ROI story that's hard to ignore. By quantifying the cost and time savings realized through automated integration, by sun-setting systems and technologies dedicated to processing and integrating data for specific purposes, and by eliminating duplicate work efforts - often representing hundreds of man-years of effort - an IT department can not only financially justify CDI, but institutionalize its accompanying processes across projects and departments.

Lots of companies have hung the data management shingle, claiming to have dedicated data stewards and formalized processes. But in the dark of night, many IT managers have to admit that these data stewards also have "day jobs," and the processes are performed on an as-needed basis. There really is no enterprise commitment to data.

Thus, as important as CDI is from a data integration standpoint, its real value may be in automating and enforcing key processes and job responsibilities that will enable a company to fund and sustain the management of both customer and non-customer data on behalf of the enterprise. It's been borne out again and again: investing in data management is often less risky than the opportunity cost of not managing data.


Jill Dyche is a partner and co-founder of Baseline Consulting (www.baseline-consulting.com), a data integration and business analytics delivery firm. Her first book, e-Data (Addison Wesley, 2000) introduced managers to the concept of enterprise data integration. Her second book, The CRM Handbook (Addison Wesley, 2002) is the CRM best-seller. You can reach her at jilldyche@baseline-consulting.com.

For more information on related topics, visit the following channels:



Industry Vendors