Once you identify your goal, many options exist for turning data into knowledge.
Profiles
Profiling is exactly what it implies: the act of using data to describe or profile a group of customers or prospects. It can be performed on an entire database or distinct sections of the database. The distinct sections are known as segments. Typically they are mutually exclusive, which means no one can be a member of more than one segment.
Segmentation is the act of splitting a database into distinct sections or segments. There are two basic approaches to segmentation: market driven and data driven. Market-driven approaches allow you to use characteristics that you determine to be important drivers of your business. In other words, you pre-select the characteristics that define the segments. This is why defining your objective is so critical. The ultimate plans for using the segments will determine the best method for creating them. On the other hand, data-driven approaches use techniques such as cluster analysis or factor analysis to find homogenous groups. This might be useful if you are working with data about which you have little knowledge.
Response
A response model is usually the first type of targeting model that a company seeks to develop. If no targeting has been done in the past, a response model can provide a huge boost to the efficiency of a marketing campaign by increasing responses and/or reducing mail expense. The goal is to predict who will be responsive to an offer for a product or service. It can be based on past behavior of a similar population or some logical substitute.
A response can be received in several ways, depending on the offer channel. A mail offer can direct the responder to reply by mail, phone, or Internet. When compiling the results, it is important to monitor the response channel and manage duplicates. It is not unusual for a responder to mail a response and then respond by phone or Internet a few days later. There are even situations in which a company may receive more than one mail response from the same person. This is especially common if a prospect receives multiple or follow-up offers for the same product or service that are spaced several weeks apart. It is important to establish some rules for dealing with multiple responses in model development. A phone offer has the benefit of instant results. A response can be measured instantly. But a non-response can be the result of several actions: The prospect said “no,” or the prospect did not answer, or the phone number was incorrect.
Many companies are combining channels in an effort to improve service and save money. The Internet is an excellent channel for providing information and customer service. In the past, a direct mail offer had to contain all the information about the product or service. This mail piece could end up being quite expensive. Now, many companies are using a postcard or an inexpensive mail piece to direct people to a Web site. Once the customer is on the Web site, the company has a variety of available options to market products or services at a fraction of the cost of direct mail.
Risk
Approval or risk models are unique to certain industries that assume the potential for loss when offering a product or service. The most well-known types of risk occur in the banking and insurance industries. Banks assume a financial risk when they grant loans. In general, these risk models attempt to predict the probability that a prospect will default or fail to pay back the borrowed amount. Many types of loans, such as mortgages or car loans, are secured. In this situation, the bank holds the title to the home or automobile for security. The risk is limited to the loan amount minus resale value of the home or car. Unsecured loans are loans for which the bank holds no security. The most common type of unsecured loan is the credit card. While predictive models are used for all types of loans, they are used extensively for credit cards. Some banks prefer to develop their own risk models. Others banks purchase standard or custom risk scores from any of the several companies that specialize in risk score development. For the insurance industry, the risk is that of a customer filing a claim. The basic concept of insurance is to pool risk. Insurance companies have decades of experience in managing risk. Life, auto, health, accident, casualty, and liability are all types of insurance that use risk models to manage prices and reserves. Due to heavy government regulation of pricing in the insurance industry, managing risk is a critical task for insurance companies to maintain profitability.
Many other industries incur risk by offering a product or service with the promise of future payment. This category includes telecommunications companies, energy providers, retailers, and many others. The type of risk is similar to that of the banking industry in that it reflects the probability of a customer defaulting on the payment for a good or service.
The risk of fraud is another area of concern for banks and insurance companies. If a credit card is lost or stolen, banks generally assume liability and absorb the charged amounts as a loss. Fraud detection models are assisting banks in reducing losses by learning the typical spending behavior of their customers. If a customer’s spending habits change drastically, the approval process is halted or monitored until the situation can be evaluated.
Activation
Activation models are models that predict if a prospect will become a full-fledged customer. These models are most applicable in the financial services industry. For example, for a credit card prospect to become an active customer, the prospect must respond, be approved, and use the account. If the customer never uses the account, he or she actually ends up costing the bank more than a non-responder. Most credit card banks offer incentives such as low-rate purchases or balance transfers to motivate new customers to activate. An insurance prospect can be viewed in much the same way. A prospect can respond and be approved, but if he or she does not pay the initial premium, the policy is never activated.
Insights, tools and best practices to empower you to Lead from the Future™

There are two ways to build an activation model. One method is to build a model that predicts response and a second model that predicts activation given response. The final probability of activation from the initial offer is the product of these two models. A second method is to use one-step modeling. This method predicts the probability of activation without separating the different phases.
Cross-Sell and Up-Sell
Cross-sell models are used to predict the probability or value of a current customer buying a different product or service from the same company (cross-sell). Up-sell models predict the probability or value of a customer buying more of the same products r services.
As mentioned earlier, selling to current customers is quickly replacing new customer acquisition as one of the easiest way to increase profits. Testing offer sequences can help determine what and when to make the next offer. This allows companies to carefully manage offers to avoid over-soliciting and possibly alienating their customers.
Attrition
Attrition or churn is a growing problem in many industries. It is characterized by the act of customers switching companies, usually to take advantage of “a better deal.” For years, credit card banks have lured customers from their competitors using low interest rates. Telecommunications companies continue to use strategic marketing tactics to lure customers away from their competitors. And a number of other industries spend a considerable amount of effort trying to retain customers and steal new ones from their competitors.
Over the last few years, the market for new credit card customers has shrunk considerably. This now means that credit card banks are forced to increase their customer base primarily by luring customers from other providers. Their tactic has been to offer low introductory interest rates for anywhere from three months to one year or more on either new purchases and/or balances transferred from another provider. Their hope is that customers will keep their balances with the bank after the interest converts to the normal rate. Many customers, though, are becoming quite adept at keeping their interest rates low by moving balances from one card to another near the time the rate returns to normal. These activities introduce several modeling opportunities. One type of model predicts the act of reducing or ending the use of a product or service after an account has been activated. Attrition is defined as a decrease in the use of a product or service. For credit cards, attrition is the decrease in balances on which interest is being earned. Churn is defined as the closing of one account in conjunction with the opening of another account for the same product or service, usually at a reduced cost to the consumer. This is a major problem in the telecommunications industry.
Lifetime Value
A lifetime value model attempts to predict the overall profitability of a customer (person or business) for a predetermined length of time. Similar to the net present value, it is calculated over a certain number of years and discounted to today’s dollars. The methods for calculating lifetime also vary across products and industries.
As markets shrink and competition increases, companies are looking for opportunities to profit from their existing customer base. As a result, many companies are expanding their product and/or service offerings in an effort to cross-sell or up-sell their existing customers. This approach is creating the need for a model that goes beyond the net present value of a product to one that defines the lifetime value of a customer or a customer lifetime value (LTV) model.