Nearly all organizations planning to implement a customer relationship management program will need a data warehouse or decision support system.

But companies should not expect that a data warehouse (DW) and decision support system (DSS) will immediately permit them to forecast the future or assist in developing a corporate strategy. Using a DW/DSS for customer relationship management (CRM) is a learning process. The learning developed in one stage forms the basis for more advanced analysis and applications. See Figure 1 for an illustration of the evolution of CRM using DSS/DW.

This discussion of the evolution of DW/DSS use concentrates on the characteristics of use at each developmental stage and the questions asked by people using the DW at different stages of the process. The stages are also helpful in positioning technology, applications, data and the integration or use of information resources.

DW/DSS Type One: Reporting

The first type of DSS is characterized by a large quantity of predefined queries. Technologists set up queries after receiving requests from business users, or the queries are determined by management to communicate what needs to be known after a period or a process is completed. These types of reporting systems usually provide a complete set of charts/graphs/cubes about a specific area of the business which answer the most frequently asked questions about the company, its market or customers. Section A of Figure 1 shows the distribution of use types and business value.

Figure 1: Evolution of CRM Using DSS/DW

Reporting is the major function of a DSS in the early stages. The underlying queries are known, and the data is mostly summarized and presented quickly. This stage of a DSS answers the strategic question: "What happened?" or what may be called "hindsight viewing." Some typical beginning data warehouse examples are:

What are the total revenues, sales, expenses, volumes or products produced?
Where did most of the sales, revenues, deliveries or services occur?
What are the comparisons to or differences from past period(s)?
What are our most/least productive resources (money, products, transportation, people)?

The initiation of new levels of information and reporting has its benefits, since it:

  • Provides easier access to previously inaccessible data.
  • Focuses on elements of information that are known by the management requestors but may not have been previously delivered or accessible.
  • Generates initial awareness of actions and problems.
  • Defines extensions to standard reporting systems.
  • Elevates the needs for more data and the transformation of data into information.
  • Opens the eyes of several key managers on what they possibly could learn if they invested more in the data warehouse.

Some management attitudes may inhibit future growth and maturity in the initial stage or early uses of a DSS and DW. One such attitude is over-expectation and the desire to see all of the data on all customers in one system at one time. Users must realize that the initial implementation is only a fraction of the actual potential volume, quality, credibility, creative usage and return on investment possible from the DW.
Additionally, some people may think that the wonderful new graphical or creative display of information is the data warehouse itself or is the truthful exposition of a customer relationship. This is rarely the reality, and the management team must understand what is being achieved and what must be achieved in the future.

The initial focus should be on building the framework, the infrastructure for getting initial results and reports, allowing for expansion of information and applications in future stages. Just learning real transformation and database normalization techniques is difficult enough for the information technology and management teams.

Solving such business and data issues as "What is a customer?" "What is a product?" or "What is a channel to our customers?" can keep an experienced management team busy for months. Adjudicating the characteristics among the various business divisions, applications and databases can shorten this process.

Figure 2 shows the distribution of the reporting types of DW/DSS usage. These include mostly reporting, some ad hoc queries and possibly some analysis applications.

Figure 2: Reporting Applications and Questions

When using the DW for reporting in customer relationship management, the focus is on defining the characteristics and habits of your customers. Some of the initial questions are:

Who are our customers? (age, income, gender, group)
Where do they live? (geography, economics, styles, etc.)
What did they buy in the past? (historical views)
How did they buy it? (financial transaction information)
Which ones are the most profitable? (known margins, etc.)
What is the cost to support them via their chosen channel?
Which groups of customers buy similar products?
What is the average customer revenue? Our expenses?
What is the annualized customer churn rate?

Reporting applications provide some answers found in many businesses' databases, but data warehousing provides new views and an ability to use combined, cross-organizational, detailed data to understand the past.

Stage one represents the core essentials – and the most immediate data needs – of the user's requirements. In some implementations, this is restricted to summarized data, because the providers or requestors (business management) have only a partial understanding of the real value of detailed data. They have yet to discover the full potential of the data in their repository.

Some organizations with only limited knowledge of the actual potential of data warehousing limit their investments in the early stage to prove the initial value of the DW or to deliver on the promise of better data.

Some companies predefine their queries into their corporate data warehouses and do not permit their business users to ask ad hoc queries of historical customer data. Organizations that control their data resources limit the potential of creating new business opportunities and relationships. Reasons for limiting the use of data usually fall into two categories: the IT department wants to maintain good security, quality, performance or costs; or a business department wants to own the data, manage it and limit its use by others.

Creative and leading enterprises have a strategy for getting information to users or making it available. However, a lack of detailed experiences in the user community may mean that the DW/DSS system still focuses on the familiar or predetermined.

Once business users learn ways of interrogating the warehouse's detailed data and accessing its multiple and complex combinations, the magnitude of the return on investment changes to high growth and high profitability. This usually emerges in subsequent stages.

DSS/DW Type Two: Analyzing

Once we have learned what happened in the initial use of a DW/DSS, we move on to the more complex, ad hoc queries of the second type of DSS: analyzing. This stage focuses on the question of why did it happen. (See Figure 1, section B.)

This is an organizational process of understanding the factors that brought about the results "discovered" earlier.

This is an important transformation in understanding the value of the data warehouse. As shown in Figure 3, the use of customer information now accelerates the ability to segment and analyze customers and their actions. In addition, the types of questions become much more sophisticated, and the abilities of the information environment are better understood throughout the organization. "You mean I can ask any question of that new DW system?" is a typical comment. Some typical data warehouse ad hoc business queries that arise in this phase are:

Figure 3: Analyzing Applications and Questions

Why did our team not meet or exceed its forecast or goals?
Why were volumes so low or deliveries later than expected?
What caused the most positive results or highest margins?
Where do we actually achieve our best ROI?
Why are inventories or resources not moving well?

This type of DW/DSS encompasses data mining through models and detailed mathematical correlation. It also has the ability to drill down into a database for minute details and come away with deductive conclusions based on the data. Business users discover trends and patterns not readily apparent from the straightforward reporting of the earlier stages. Awareness of system capabilities and allowing people to ask a wider range of questions drive a new desire to utilize the information infrastructure and begin to change people's thinking about what the systems are there for.

Stages three and four CRM analysis focuses on understanding customers.

Why is average customer revenue down?
Why is annualized customer churn so high?
Why did the campaign not meet plan?
Why are sales of a product below plan?
Why did they buy it from you?

The reporting applications now need to interact with one another to reveal a new reality. The types of applications use more sophisticated tools, but the hallmark of the change is the use of analysis methods and models to answer the questions surrounding "Why did it happen?" This move beyond reporting is significant, because detailed, historical data is now exploited to understand much more about past behaviors and other characteristics formerly unknown to management.

This ability to understand the past is the key to understanding the future, which is the hallmark of stage three.

DSS/DW Type Three: Predicting the Future

Forming knowledgeable, high-percentage predictions is a specialized skill that truly separates leading companies from the rest of the pack. Those who can anticipate trends and capitalize on them before they become common knowledge have an obvious edge in the marketplace.

A comprehensive data warehouse, having an "analytical modeling" capability, where the queries ask "What will happen?" provides immense abilities to achieve a prophetic facility. (See Figure 1, section C.)

The more mature stages provide the pathway to highest profitability and high ROI. Some of the questions that phase three (with its corresponding predictive applications) can help answer are:

Which customers are at risk of leaving? (Customer retention application)
What products or services will the customer buy? (Market segmentation)
What is the best way to reach a customer? (Channel optimization)
How will a new product sell? (Demand forecasting)

The applications are now very sophisticated and utilize advanced techniques of decision support, parallel query functions, massive detailed historical data, cross-functional information pertaining to customers, finite knowledge of behaviors, scoring of credit, payment ability, behaviors, propensities, pure prediction and complex strategic decisions.

Once a modicum of maturity in information management has been achieved, the enterprise becomes more mature, networked, positioned, integrated, knowledge- based and highly flexible.

The enterprise can now get answers to questions such as "What will happen in the future?" This is the type of DSS or data warehouse that is characterized by the applications and questions in Figures 4 and 5.

Figure 4: Stages of Development and Prediction Capabilities

Figure 5: Predictive Applications and Prediction Questions

Type three of DSS with CRM – thinking like your customer includes questions such as:

Which customers are more likely to leave?
Which customers are more likely to buy?
Are they likely to defect?
What's the impact on profit when I change price?
What's the best channel to reach a specific customer?

Types of DW/DSS – A Summary

The most powerful use of the data warehouse and CRM applications is in the mode of active data warehousing where the institution of real-time, event- driven, predictive capabilities are focused on action-oriented communication to customers from the marketing or services organizations of a firm. This advanced stage of a DW brings forth a highly profitable result and forges the use of "relationship technologies" in ways not even perceived in the first five stages.

Predefined queries and reporting are certainly the foundation of a corporate data warehousing and CRM system. But the analytical and predictive applications are the bellwethers of competitiveness. These kinds of queries soon become indispensable elements in any company's ability to utilize and learn from information.

A successful enterprise uses the data warehouse and CRM to achieve a triple view: knowing the past, analyzing the present and predicting the future (with a high rate of accuracy, while measuring and refining the processes and models).

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access