Demographics Plus Distance Data Equals Improved Response

It may seem elementary, but it is difficult to succeed at target marketing if you don’t first clearly define your target market. This definition is best achieved through the consideration of demographic and distance data, yet this innovative combined approach is as yet a rarity.

Hopefully, you have joined the expanding ranks of marketers who incorporate demographic data to define their best customers. If not, a proven method of doing so involves using your current customer database, either in its entirety or with a suitably sized random sample, to identify those customers generating the top 20% of your revenue. You can then pull a similarly sized random sample of consumers from the same geographic areas—that is, from the same list of zip codes or other applicable geography—for the purpose of tracking how your customers compare with other consumers living in the same area. The random consumer samples are obtainable from any of the major national consumer database compilers.

Enhance your consumer and customer files with a range of demographic and lifestyle data—again, available from a number of data providers. These data include such details as age, gender, marital status, rent-vs.-own–characteristics that can relate the traits of your purchasing customers to your products. Cross-tabular comparisons of the two files can reveal how your best customers differ from those in the market at large, for instance demonstrating that your best customers tend to be older married homeowners.

Going further, you can develop logistic regression models to predict the probability of a consumer exhibiting best customer behavior. Logistic regression will typically yield 75%-80% correct classification rates. Applying the model to the original research sample will demonstrate the percent of cases correctly classified, and the model will generate a probability score that can be translated into consumer decile rankings. With the model built, target-marketing programs can focus on customers scoring in the top deciles.

If you leverage the modeling approach to identify your best customers, you are taking advantage of a vital tool. But while many marketers use demographic data at some level, few combine this information with distance data crucial for maximizing response rates. Yet distance data are among the top three predictors of customer response.

What are distance data? Simply put, these data indicate your customer’s geographic point of reference. Geographic proximity can be a key factor in the success of your target marketing program. For example, the farther a consumer must travel to reach your nearest brick-and-mortar store, if applicable, the less likely he may be to respond to your target marketing offer. This is seen even with products sold via the Web, telephone, or a print catalog, because the proximity can reflect the perceived distance a consumer may need to travel for service or returns. Moreover, your physical stores create a billboard effect; the nearer the store, the more likely the brand will occupy significant consumer mind share.

Traditionally, if marketers used distance data at all, distance was determined through a “radius selection”—that is, all consumers within 5- 10 miles of a store were considered to be within that store’s trade area. Customers rarely live their lives within neatly drawn circles, however, nor along straight-line paths from stores. The road network and dominant commuting routes can be far more predictive of the consumer’s ability and likelihood of responding to an offer from a given store.

Therefore, to make the best use of geographic information in identifying and reaching the target market, today’s multichannel merchant must

1) identify the customer’s geographic points of reference for the purpose of assessing the situation from the consumer’s standpoint; and

2) combine this information with demographic best-customer data for the optimal targeted marketing campaign.

Bear in mind that customers have multiple points of reference. These points may include

* impact of having a brick-and-mortar store in addition to your catalog or Website

* impact of competitors’ sites in proximity to your target customers

* impact of outdoor advertising near your consumers.

Essentially, to the extent that competitors’ options are available to consumers, response rates will decrease. To the extent that competitors’ sites are distant from consumers, response rates will increase. And if you have a store proximally available in addition to your catalog or Website, response rates will increase. Likewise, outdoor advertising increases awareness. As a general rule, response rates increase as the number of reference points increase.

The combination of demographic and distance data scoring works two ways. First, it helps you identify the areas most vital to reach. Second, it helps you identify areas that should be excluded based on a low probability of response.

In the example of best-customer modeling, both files should be enhanced with the measurement of the customer’s distance to your nearest retail location, if any, as well as to your nearest competitor option. Distances may be calculated either linearly (“as the crow flies”) or via the road network.

As a final but consequential point, many multichannel merchants may be concerned about the costs of such analyses. Today, however, we have cost-effective options for securing and processing data. Demographic and distance measures can be acquired and analyzed efficiently for your target marketing campaign.

Larry Daniel is president/CEO of Austin, TX-based database marketing services firm Conclusive Strategies.

Demographics Plus Distance Data Equals Improved Response

It may seem elementary, but it is difficult to succeed at target marketing if you don’t first clearly define your target market. This definition is best achieved through the consideration of demographic and distance data, yet this innovative combined approach is as yet a rarity.

Hopefully, you have joined the expanding ranks of marketers who incorporate demographic data to define their best customers. If not, a proven method of doing so involves using your current customer database, either in its entirety or with a suitably sized random sample, to identify those customers generating the top 20% of your revenue. You can then pull a similarly sized random sample of consumers from the same geographic areas—that is, from the same list of zip codes or other applicable geography—for the purpose of tracking how your customers compare with other consumers living in the same area. The random consumer samples are obtainable from any of the major national consumer database compilers.

Enhance your consumer and customer files with a range of demographic and lifestyle data—again, available from a number of data providers. These data include such details as age, gender, marital status, rent-vs.-own–characteristics that can relate the traits of your purchasing customers to your products. Cross-tabular comparisons of the two files can reveal how your best customers differ from those in the market at large, for instance demonstrating that your best customers tend to be older married homeowners.

Going further, you can develop logistic regression models to predict the probability of a consumer exhibiting best customer behavior. Logistic regression will typically yield 75%-80% correct classification rates. Applying the model to the original research sample will demonstrate the percent of cases correctly classified, and the model will generate a probability score that can be translated into consumer decile rankings. With the model built, target-marketing programs can focus on customers scoring in the top deciles.

If you leverage the modeling approach to identify your best customers, you are taking advantage of a vital tool. But while many marketers use demographic data at some level, few combine this information with distance data crucial for maximizing response rates. Yet distance data are among the top three predictors of customer response.

What are distance data? Simply put, these data indicate your customer’s geographic point of reference. Geographic proximity can be a key factor in the success of your target marketing program. For example, the farther a consumer must travel to reach your nearest brick-and-mortar store, if applicable, the less likely he may be to respond to your target marketing offer. This is seen even with products sold via the Web, telephone, or a print catalog, because the proximity can reflect the perceived distance a consumer may need to travel for service or returns. Moreover, your physical stores create a billboard effect; the nearer the store, the more likely the brand will occupy significant consumer mind share.

Traditionally, if marketers used distance data at all, distance was determined through a “radius selection”—that is, all consumers within 5- 10 miles of a store were considered to be within that store’s trade area. Customers rarely live their lives within neatly drawn circles, however, nor along straight-line paths from stores. The road network and dominant commuting routes can be far more predictive of the consumer’s ability and likelihood of responding to an offer from a given store.

Therefore, to make the best use of geographic information in identifying and reaching the target market, today’s multichannel merchant must

1) identify the customer’s geographic points of reference for the purpose of assessing the situation from the consumer’s standpoint; and

2) combine this information with demographic best-customer data for the optimal targeted marketing campaign.

Bear in mind that customers have multiple points of reference. These points may include

* impact of having a brick-and-mortar store in addition to your catalog or Website

* impact of competitors’ sites in proximity to your target customers

* impact of outdoor advertising near your consumers.

Essentially, to the extent that competitors’ options are available to consumers, response rates will decrease. To the extent that competitors’ sites are distant from consumers, response rates will increase. And if you have a store proximally available in addition to your catalog or Website, response rates will increase. Likewise, outdoor advertising increases awareness. As a general rule, response rates increase as the number of reference points increase.

The combination of demographic and distance data scoring works two ways. First, it helps you identify the areas most vital to reach. Second, it helps you identify areas that should be excluded based on a low probability of response.

In the example of best-customer modeling, both files should be enhanced with the measurement of the customer’s distance to your nearest retail location, if any, as well as to your nearest competitor option. Distances may be calculated either linearly (“as the crow flies”) or via the road network.

As a final but consequential point, many multichannel merchants may be concerned about the costs of such analyses. Today, however, we have cost-effective options for securing and processing data. Demographic and distance measures can be acquired and analyzed efficiently for your target marketing campaign.