essay

Optimal_Database_Marketing_Strategy_Development_an..._----_3_-_Defining_Customer_Data_Requirements.pdf

Home >Business & Finance homework help >Marketing homework help >essay

Defining Customer Data Requirements

3 Keri Lee had mixed emotions when she received the news that her appli- cation for a new position at DSI was approved. Although she enjoyed her job selling software to customers in the b-to-b market, she was seeking a change. The constant travel required in the sales position was reducing the time Keri wanted to spend with family and friends. She also wanted to be part of a team that was more directly involved in developing marketing strategies and programs. Her new position, as associate product manager in DSI’s consumer’s division, would present this challenge. The consumer division was new, and they were in the process of developing a database to achieve marketing objectives.

Keri’s new supervisor had extensive experience in database marketing, coming to DSI from a major direct marketer of computer hardware. Keri realized there would be much to learn. However, she was confident that with her academic background—she was a marketing major and an infor- mation systems minor—and previous work experience, she would succeed.

At her first meeting in the consumer division, Keri discovered that database development had only just started. There was some customer information housed in various company files and databases, but this information was incomplete and scattered throughout several departments. DSI had just devel- oped a very basic Web site, and some data were also being collected there.

After DSI analyzed the opportunities and structure of the consumer graph- ics software market, they decided to use direct marketing approaches. The goal was to establish a database of nonbusiness users of graphics software programs so that the company could develop targeted promotions for their product lines. Because the development of graphics software is very dynamic, DSI felt that they could gain a competitive advantage in respond- ing to customers’ needs through a database. One of Keri’s first tasks was

Drozdenko03 2/26/02 6:11 PM Page 39

Drozdenko, Ronald G., and Perry D. Drake. Optimal Database Marketing : Strategy, Development, and Data Mining, SAGE Publications, 2002. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/nyulibrary-ebooks/detail.action?docID=996727. Created from nyulibrary-ebooks on 2020-06-09 19:14:53.

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

40 OPTIMAL DATABASE MARKETING

to gather all usable internal data and explore external data sources. She would be working with the marketing manager and the information systems manager on this project.

As we discussed in Chapter 2, the process of building a marketing data-base begins with a definition of what the database is expected to achieve in both the short and long term from a marketing perspective. A review of the organizational objectives and marketing plans must first be conducted in order to set the foundation for database development. Once determined, the details of the database can be outlined.

In particular, the following steps must be taken in the development of the database to ensure that marketing objectives will be met:

♦ Determine the data requirements needed to meet the marketing objective

♦ Establish guidelines for database maintenance and program coding ♦ Evaluate the database structure, including hardware/software

requirements ♦ Determine whether the database will be built and maintained inside

the organization or outsourced

This chapter discusses the first step—determining the data requirements. Subsequent steps are discussed in Chapters 4 and 5.

Data Needs Determination _______________________________ Database development begins by determining the types of data attributes (fields) required to support the marketing objectives. The amount of data residing on a database can vary greatly from company to company depend- ing on needs and how the database will be used to meet the marketing objectives. For instance, a food company might wish to use a database only to introduce a new line of organic breakfast cereals. To promote the cereal to a group of health enthusiasts, the company might use a mail-in response card in a diet and nutrition magazine. Responders will get a free sample of the new cereal. Because the product will be sold only in supermarkets, the company will not be able to establish a database that includes historical sales data. After the product is introduced, the database’s utility will be limited to a consumer loyalty program (e.g., coupons).

On the other hand, the direct marketer of a series of home improvement books might be striving to establish long-term relationships with cus- tomers. Customer retention and product development are important goals for the database. Therefore, an extensive database is needed to record and analyze the ongoing transactions of customers.

Drozdenko03 2/26/02 6:11 PM Page 40

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

Before proceeding any further, you need to know the difference between a marketing database and fulfillment and prospecting databases.

Any direct marketer currently taking and fulfilling orders of any kind must have, at the very least, a fulfillment database or fulfillment file. In all likelihood, this file is managed and maintained by an outside vendor. The sole purpose of this file is to preserve information on customers about their order, product shipment, and billing information and status. Most fulfill- ment files do not maintain historical data. As such, they cannot be used to conduct analyses of past customer purchase behavior.

A marketing database, on the other hand, is structured for efficiency and does maintain a history of all customer transactions over time. These databases are derived from the customer information in a fulfillment database. How often the fulfillment data feed the marketing database depends on the enterprise’s needs. This is discussed in greater detail in Chapter 4.

Marketing databases allow direct marketers to more easily obtain quick counts on active customers, select names for future promotions across the various divisional product lines, and track customer perform- ance over time. This book focuses solely on the use of a marketing data- base.

Prospecting databases are a type of marketing database but comprised solely of noncustomers. They are usually kept as a separate file because they have limited information on these noncustomers. Once prospects order and become a customer, their information will be transferred from the prospect- ing database to the marketing database.

Defining Customer Data Requirements 41

__________________ Data Residing on the Marketing Database To be successful, a direct marketer must know not only which data elem- ents to keep on the customer file but also how to use the data most effect- ively. The types of data residing on a marketing database can be classified as either internal or external, as shown in Exhibit 3.1.

Internal or House Data

Internal or house data are obtained from a number of sources within the organization. Accounting, customer service, sales, research, and information systems departments can have data relevant for marketing purposes. Even within those departments, several different databases can exist. Marketers should be aware that the incompatibilities of existing databases can pose

__________ Fulfillment, Marketing, and Prospecting Databases

Drozdenko03 2/26/02 6:11 PM Page 41

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

42 OPTIMAL DATABASE MARKETING

problems when a database specialist attempts to consolidate data from various sources within the organization. A current concern for many marketers is integrating Internet databases with older legacy systems (Internet database integration issues are discussed in Chapter 15). Problems with incompatibility can result in project delays.

The types of customer data available from internal sources can vary widely, and the format and degree of completeness of these data will vary from organization to organization. The data include

♦ Previous contact information ♦ Past purchase records ♦ Product returns data ♦ Customer service data ♦ Responses to internal surveys ♦ Voluntary customer registration data (e.g., warranty cards, Internet

site registration)

Internal or house customer data found on a marketing database can be classified as either fulfillment, marketing, or customer contact data.

Int ern

al Da

ta External Data

MARKETING DATABASE

Exhibit 3.1 Sources of Data in the Marketing Database

Drozdenko03 2/26/02 6:11 PM Page 42

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

Fulfillment Data

Fulfillment data are basically raw or transactional-level data and are pri- marily used for billing and fulfilling orders. However, a marketer also wants to use such data. For example, the data field “Last Bill Effort Sent” is used by a marketer as an indicator of how quickly a customer pays.

Marketing Data

Marketing data are any piece of customer information used by marketers for the purpose of increasing the effectiveness or efficiencies of marketing activities. This includes data that can help marketers promote customers, develop relationships with customers, and establish marketing strategies and programs. For example, most past purchase information will be used by the marketing division to help predict a customer’s future purchase behavior. Most marketing data elements are based on summarized and “rolled-up” fulfillment data. For example, past information on customer payment patterns might be helpful for segmenting the best customers in the database.

In general, house marketing data can be classified into three types:

1. Recency data are related to the recency of a customer’s last promo- tion, order, payment, and so on. Examples of recency data created and maintained on marketing databases include time in months since a customer’s last order and time in months since a customer’s last payment.

2. Frequency data are related to a customer’s total number of promo- tions, orders, payments, etc. Examples of frequency data created and maintained on marketing databases include total number of promotions sent to a customer and total number of product returns by a customer.

3. Monetary data are related to a customer’s total dollar value of orders placed, payments made, etc. Examples of monetary data created and maintained on marketing databases include total dollars written off for a customer due to nonpayment.

Recency data are the strongest of the three data elements. As a result, the predictive strength of frequency and monetary data elements can be amplified by incorporating an element of recency in their definition. For example, the total number of orders placed by a customer is a strong predictor of a customer’s likelihood to order in the future, but the total number of orders placed by a customer within the past 12 months is an even stronger predictor.

Defining Customer Data Requirements 43

Drozdenko03 2/26/02 6:11 PM Page 43

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

44 OPTIMAL DATABASE MARKETING

If appropriate, bringing in an element of affinity to any of the above three classifications of data elements will also make these data elements even more relevant. For example, the total amount of money spent by a cus- tomer on investment books in the past 24 months is a very strong indicator of the customer’s likelihood of ordering investment books in the future.

According to Direct magazine’s 2000 subscriber survey, at least 75% of consumer and 52% of b-to-b direct marketers maintain standard recency, fre- quency, and monetary (RFM) indicators. Only 48% maintain promotional history, and only 35% keep track of promotional nonresponses—both of which are quite important in allowing a marketer to be effective in cross-selling and reactivating customers.

Customer Contact Data

Customer contact data are the foundation of the database. The marketer must have a means to reach customers. Basic contact information includes name, address, zip code, phone and fax numbers, and e-mail address. Marketing efficiency and effectiveness is greatly affected by the quality of the contact infor- mation. The required contact information must correspond to the organization’s marketing communication vehicle. A telemarketer must have an accurate tele- phone number, a mail marketer must have an accurate mailing address, and an Internet marketer must have an accurate e-mail address. Therefore, the strategy for developing and maintaining the database should encompass the importance of the information for marketing purposes (see Chapter 4).

Smaller organizations often have poorly organized information about customers. For example, the sales staff of a small b-to-b marketer can have customer information on their PCs or in paper files. Centralization of marketing information in a common database will help to facilitate and coordinate marketing efforts. In the b-to-b market, redundant sales contacts can be avoided and interaction with multiple personnel in the organization can be documented to get a better idea of customer needs.

In an attempt to streamline the management of the sales process, some companies require salespeople to use Internet-based programs. This allows real-time updates on the time and location of their next appointment. In addition, the database provides the salesperson with market profiles and intelligence on customers and prospective customers. Although some salespeople were initially reluctant to use the Internet database for account management, they have found it reduces account maintenance tasks. With account maintenance task reductions, salespeople can spend more time on commission-generating activities. (Dana, 1999). Larger organizations with existing databases for nonmarketing purposes (e.g., operations, accounting) may need to restructure the database to make it suitable for marketing purposes. For example, existing databases may have customer contact and

Drozdenko03 2/26/02 6:11 PM Page 44

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

purchase transaction information. However, it can be difficult or impossible to relate specific purchases to a particular marketing program such as a mail or direct TV campaign. Remember that the goal for the organization is to develop a structured and systematic method of not only contacting customers but also tracking the individual account. Integration across functional areas of the business is a necessity for an optimal marketing database. Strategic priorities for developing the database should be placed on those areas that translate into effective and efficient customer contact and tracking.

External or Enhancement Data

External or enhancement data are information about a direct marketer’s cus- tomers obtained from a third party or vendor. This type of data is generally purchased by direct marketers to supplement their own house customer data to learn more about their customers. The marketers can then more effectively

♦ Determine current and future customer needs ♦ Enhance advertising ♦ Increase response rates ♦ Acquire new customers ♦ Extend appropriate credit

External data can be classified into three categories:

1. Compiled list data

2. Census data

3. Modeled data

How enhancement data are collected and how they can be used has recently become a topic of much debate due to privacy concerns of many citizens and the ease by which such information can be obtained via the Internet. For example, the Fair Credit Reporting Act forbids the sale of consumer credit report information to direct marketers unless authorized for specific purposes. The Driver’s Privacy Protection Act recently stopped the sale of all DMV (Department of Motor Vehicles) data. This was the single most accurate source of age information available. States must now obtain written permission from licensed drivers before selling the data to third-party marketers.

Compiled List Data

Compiled list data are individual-level data collected by list service bureaus or list vendors such as Polk, infoUSA, Experian/Metromail, Acxiom, and

Defining Customer Data Requirements 45

Drozdenko03 2/26/02 6:11 PM Page 45

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

46 OPTIMAL DATABASE MARKETING

Trans Union to sell it to direct marketers. It can be gathered from a variety of sources: telephone directories, voter registration files, birth records, housing purchase records, membership rosters, etc.

The information on compiled lists can be categorized as either demo- graphic or psychographic. Examples of the types of data elements you can purchase are shown in Exhibit 3.2.

Assume a major men’s shoe chain is considering expanding by selling upscale athletic shoes and exercise apparel. They have no information on their customer database that indicates who might be most interested in this new product line. Currently, they sell only expensive men’s dress and casual shoes. Before expanding, they would like to gauge demand for this new product line from their current customer base by determining what percentage of their customers are interested in fitness and exercise. What information could they use to enhance their customer database that would help them determine the demand?

The shoe marketer might consider some of the following enhancement data elements:

♦ Members of health clubs ♦ Recent purchasers of exercise equipment ♦ Respondents who checked “enjoy exercising” on a questionnaire or

warranty card ♦ Recent purchasers of exercise clothing ♦ Readers of exercise and fitness magazines

Assuming that a high percentage of their current customer base shows an interest in fitness and exercise, the shoe chain can use this enhancement data to help them select names from the database for a special promotion.

Direct marketers can also collect their own demographic and psycho- graphic data via questionnaires. Nordstrom, the upscale department store chain, recently augmented the information in its cardholder database from

Exhibit 3.2 Demographic and Psychographic Data Elements

Demographic Psychographic

Income Hobbies Age Reading interests Gender Exercising habits Marital status Music preferences (rock, country, etc.) Education level Movie preferences (action, drama, etc.) National origin, ethnicity Political orientation Family size Social (opinions on the environment, Occupation education, family, health care, etc.) Credit information Donor (the arts, health sciences, etc.)

Drozdenko03 2/26/02 6:11 PM Page 46

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

Defining Customer Data Requirements 47

a survey mailed to its 5 million cardholders. The survey has provided them with critical information for targeting purposes, including spending habits and price sensitivity (DM News, 2000b).

Census Data

Census data are obtained from the U.S. Census Bureau from geo-demo- graphic data (zip code, block group, census track, etc.) It is not available at an individual level as compiled list data. Rather, census data represent the average measurement of residents within the geo-demographic region. For example, people residing in zip code 12345 have an average income of $56,345. This income value will be appended to those residents on the database in zip code 12345.

Census tracts are subdivisions of counties. Today, there are approxi- mately 50,000 census tracts. Block groups are subdivisions of census tracts formed by grouping blocks (streets). There are approximately 225,000 block groups. Being too fine a split, some sensitive economic and personal data are not reported at the block group level; however, it is reported at a census tract or zip code level. This is because doing so can provide individ- ual information if, for example, it was known that in a block group only one family of a particular ethnic group resides.

However, it should be noted that with the 2000 U.S. Census, marketers have access to more than 63 single-race or multirace categories down to the block level. This segmentation can be valuable to marketers desiring to tar- get certain racial groups by modifying marketing communications or offers to the needs of particular groups.

The U.S. government gathers census data every 10 years, reestimating some data such as population growth estimates between updates. Census data available within a geo-demographic region include

♦ Median income ♦ Average household size ♦ Average home value ♦ Average monthly mortgage ♦ Percentage ethnic breakdown ♦ Percentage married ♦ Percentage college educated ♦ And even such measures as average daily commuting time

Exhibit 3.3 shows census information for three very demographically distinct census tracts. If you were to promote an expensive children’s book series and could only market to the names within one of the three census tracts, which would you choose and why?

Drozdenko03 2/26/02 6:11 PM Page 47

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

When lacking an abundant amount of individual-level data pertaining to customers, a direct marketer often relies on census-level data. The premise in using this type of data for determining a product’s target market, when no other data are available, is that all individuals living within a geo-demo- graphic region are similar to one another. Although this is true in some cases, it certainly is not in others.

For example, it might be true that most people living in Beverly Hills, California, are quite wealthy and drive expensive cars, but the majority of these residents will not have the same interests in reading, music, or hobbies.

Head-to-head, census-level data will never be as powerful as individual- level data in predicting customer behavior. However, as previously mentioned, when no other information is available, it will and can play a fairly strong role depending on your applications. Consider the following examples:

♦ Census data indicating a customer’s financial status (average income levels, average home value, average number of cars owned) will play a fairly strong role in helping determine who to promote with a free trial offer, ensuring a minimal number of written-off accounts.

♦ Census data indicating a customer’s financial status can also play a strong role in determining the customers most likely to respond to promotions from such companies as Lexus, BMW, KitchenAid Appliances, and Neiman Marcus Department Stores.

48 OPTIMAL DATABASE MARKETING

Exhibit 3.3 Census-Level Data Pertaining to Three Tracts

Percentage with Median Median Median Undergraduate

Household Monthly Household or Graduate Area Size Median Age Mortgage Income Degree

Chappaqua, NY 10514 (Block Group 2, Census Tract 013102, FIPS Code 36119) 4.1 37.2 $2,000 $148,649 38.5

Affton, MO 63123 (Block Group 1, Census Tract 219800, FIPS Code 29189) 2.7 37.2 $661 $41,579 22.4

West Frankfort, IL 62896 (Block Group 3, Census Tract 040900, FIPS Code 17055) 2.2 40.9 $490 $12,636 3.7

Drozdenko03 2/26/02 6:11 PM Page 48

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

♦ Using census-level data to help determine the customers most likely to purchase a general reference coffee table book will be much more dif- ficult. Selecting more educated and literate areas for promotion would be weak at best. In this case, the best bet is to append individual-level lifestyle data from a compiled list source.

Direct marketers can obtain free census data to target their mailings from the data collection company SRC by logging on to their Web site at www.FreeDemographics.com.

Modeled Data

Modeled data are generated from statistical analysis such as customer clustering. These data are often used to classify or segment customers based on purchase patterns, demographics, and psychographics. On the basis of certain geographic data, for example, income or educational levels can be predicted.

Claritas, a company that analyzes and develops databases, offers a classification scheme called PRIZM. PRIZM is based on the premise that people with similar lifestyles tend to live near one another. PRIZM describes every U.S. neighborhood in terms of 62 distinct lifestyle types, called clusters. Clusters have been given names such as Second City Elite, Upward Bound, Boomtown Singles, Starter Families, Smalltown Downtown, Pools & Patios, Country Squires, God’s Country, Greenbelt Families, and Middle America. Each cluster has certain characteristics. For example, the Elite Exurban Family cluster is aged 45 to 64, professional, and has a median household income of $89,000. This PRIZM cluster is most likely to go cross-country skiing, bank online, own a fax machine, watch Frasier on TV, and read Forbes. The Elite Exurban Family cluster live in neighborhoods such as Prospect, Kentucky, New Fairfield, Connecticut, and Belle Mead, New Jersey. Marketers of cross-country skiing equipment or home fax machines can enhance their database with this cluster information or rent such names for promotion.

The use of this data to segment customers is discussed in more detail in Chapter 8.

Defining Customer Data Requirements 49

________________________________________ Lists Versus Data Names residing on lists can also be rented for promotions or purchased and added to a direct marketer’s prospecting database. Either way, the sole purpose is to acquire new customers. Depending on the agreement made with the list owner, you may or may not be able to promote the names

Drozdenko03 2/26/02 6:11 PM Page 49

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

multiple times. Typically, contracts are set for a one-time use or a one-year agreement with unlimited use. Outside intermediaries such as list brokers or managers often handle list rentals and maintenance for an organization.

Regarding names found on compiled lists, the main issue is that they are not proven direct mail responsive. Remember, these names came from var- ious sources such as questionnaires, registration lists, and warranty cards.

Acxiom, a major provider of compiled lists, offers InfoBase, a large collection of U.S. consumer, business, and telephone data for prospecting or enhancement. A few of their offerings include names and lists dealing with

♦ New movers ♦ Young families ♦ Active seniors ♦ Working mothers ♦ High-tech industry ♦ Banking industry

INSOURCE is a composite of the independent databases of Metromail and Experian. INSOURCE enhancement data include the following infor- mation from various sources:

♦ A broad range of demographics ♦ Consumer interests and lifestyle ♦ Telephone numbers ♦ Mail order responsiveness indicators ♦ Automotive ownership ♦ Real estate holdings ♦ Segmentation and clustering systems ♦ Census geo-demographics

There are many vendors of compiled lists and data, many of which advertise in industry publications such as DM News and Direct magazine.

Response lists are lists of names or businesses put on the market by direct marketers for rent by other direct marketers. They are called response lists because they have responded to past offers from, for example, mail order catalogs, magazine subscriptions, or charities. Response lists have a higher potential response rate than compiled lists because past behavior is a good predictor of future behavior. Therefore, marketers seek response lists composed of purchasers (or donors) in the same or a similar category to their own.

Some lists are further specified as being comprised of recent purchasers. These are called hot-line lists, and individuals on these lists are often more likely to purchase. Direct marketers can retain any name of a person who

50 OPTIMAL DATABASE MARKETING

Drozdenko03 2/26/02 6:11 PM Page 50

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

orders or responds in some manner (e.g., makes a purchase or requests more information). However, the use of the names of the people on the response list who did not respond is restricted by the rental agreement.

The Reader’s Digest Association offers more than 30 million of their names for rent. You can select names based on recency of order in addition to various demographic and psychographic attributes. They also offer response models and “best customer” models as other options in selecting names from their file that will be most responsive to your offer (Levey, 2000).

You cannot enhance your file with names rented from a response list or add them to your prospecting database. You rent the name for a one-time use only—a promotion. When you rent a response list, you first match it to your house file. This allows you to eliminate any name on the response list that matches a name on your house file that, for example, already ordered the product or service you will be offering or currently owes you money. In addi- tion, when you match the response list to your house file, a certain portion of the matches will be to your older inactive customers. Therefore, the response list provides you with valuable information for some of your older inactives. This valuable piece of information tells you which of your older inactives are active on the list owner’s file. Take advantage of this information and promote these names. However, remember, you cannot retain this data on your customer file indicating that they matched the list owner’s file unless you have made such an arrangement with the list owner in advance.

Defining Customer Data Requirements 51

If direct marketers wish to have enhancement data (census, compiled, or modeled) overlaid on their customer file, they often follow these steps:

1. The direct marketing company makes a copy of their customer file (or whatever portion they wish to enhance) and sends it to the to the enhancement service.

2. The service bureau matches their file to the direct marketing company’s file using a name and address-matching algorithm (Chapter 4).

3. Once matched, the service bureau appends the desired information (e.g., age, income, lifestyle indicators) to the copy of the file given to them.

4. The direct marketing company matches the file back to their data- base (via a unique customer number) and appends the enhance- ment data to their file for future use.

Match rates from a single source vary, depending on the makeup and size of the customer file and the specific data elements being appended.

____________________ Applying and Using Enhancement Data

Drozdenko03 2/26/02 6:11 PM Page 51

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

52 OPTIMAL DATABASE MARKETING

To get better coverage on age enhancement data, for example, consider going to several sources. You can usually work out appropriate net name/match arrangements with vendors when pursuing multiple sources of the same data field.

We end this chapter with three case studies illustrating the use of enhancement data.

Case Study 1: A Tire Manufacturer

Company: A leading manufacturer of tires, automotive parts, and services with retail locations throughout the United States. Situation Analysis: With a proprietary cardholder base of 1 million customers, the company wanted to capture consumers in the growing segment of the female mar- ketplace: women purchasing automotive accessories and services. Objective: Widen existing customer base to include more female consumers. Strategy: Using INSOURCE, the retailer overlaid demographic and automotive information onto its credit card member file, creating a more complete picture of its customer base. Focusing on the retailer’s female cardholders, a profile was developed. This was used to select a compiled list of prospects who closely resem- bled what the retailer now identified as its female “best customer.” This list was run against the client file, removing names duplicated on the database. A customized, direct mail campaign was developed with offers designed specifically for the above audience. To enhance response rates, customized offers were ink-jet- ted onto direct mailers in regions where retail locations were offering discounts. Result: The manufacturer generated an 18 percent response rate from the promo- tion and converted 45 percent of those respondents into cardholders.

Source: http://www.experian.com/direct_marketing/products/direct.html#insource

Case Study 2: A Direct Marketer of Sporting Goods

Company: A major direct marketer of sporting goods. Situation Analysis: Learn more about customer product needs. Objective: Expand product lines. Strategy: This company currently has all necessary information to contact cus- tomers (name, address, phone) and past purchase information. They want to learn more about the customers for the purpose of product development. They will send a questionnaire to customers as well as enhance the customer file with information pertaining to sporting and lifestyle activities. Result: The sporting goods marketer determined what additional products to offer their current customers based on an assessment of their needs and lifestyles via questionnaire and enhancement data.

Drozdenko03 2/26/02 6:11 PM Page 52

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

Defining Customer Data Requirements 53

_______________________________________ Chapter Summary Defining data requirements is an important early step in the development of a marketing database. Marketers have to determine what information about customers is necessary to achieve their objectives. To optimize the utility of a customer database, marketers gather data from a number of internal and external sources. Internal sources can provide data on trans- actions and any other information collected on individuals by the company. External sources can enhance this internal or house data by providing demographic, psychographic, and transaction data. There are also external lists of potential customers based on demographics, psychographics, and actual responses to offers. Some of these lists are compiled from sources such as telephone directories, birth records, and membership rosters. Response lists are distinguished from complied lists because individuals or businesses on response lists have previously responded to some type of offer. Data from the U.S. Census Bureau can be used to specify the charac- teristics (e.g., median income, household size, racial group, education level) of people living within geographic segments of the country. Modeled data generated from statistical analyses classify individuals in certain geographic areas by lifestyle categories such as Starter Families and Upward Bound. The chapter concludes with examples of how enhancement data are used to reach organizational objectives.

_______________________________________ Review Questions

1. What are some of the sources of data that can be included in the database?

Case Study 3: A Magazine Publisher

Company: A magazine publisher. Situation Analysis: Profile subscriber base. Objective: Increase advertising revenue. Strategy: Using demographic and psychographic data overlays, the publisher created a profile of their current subscriber base. In particular, they appended information regarding age, income, household composition, car ownership, and occupation. Profile reports were generated for the advertising sales staff to assist them in their efforts. Result: By profiling the subscriber base, the advertising team was successful in obtaining new accounts and renewing old accounts. As a result, advertising rev- enue increased over the prior 12-month period by more than 19 percent.

Drozdenko03 2/26/02 6:11 PM Page 53

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed

54 OPTIMAL DATABASE MARKETING

2. Give some examples of demographic and psychographic data elements.

3. How are compiled lists different from response lists?

4. Why are house files (internal databases) enhanced with supple- mental data?

5. Describe how marketers can use U.S. Census Bureau data.

6. What steps are involved in using an outside service to enhance a house file?

Drozdenko03 2/26/02 6:11 PM Page 54

C op

yr ig

ht ©

2 00

2. S

A G

E P

ub lic

at io

ns . A

ll rig

ht s

re se

rv ed