HOMING IN

Posted on by Chief Marketer Staff

It doesn’t take a genius to figure out who a cataloger’s best customers are: They’re the ones who shop most frequently and spend the most, of course.

But who are most likely to be the best customers next month? Will it be the same loyal group as last month, or will up-and-comers emerge with a little prodding? Analytics experts say that many catalogers need to spend more time nurturing their high-potential customers along with their high-performing ones. And thanks to the latest generation of demographics and geodemographics segmentation tools, finding those hot prospects is getting a whole lot easier, they suggest.

Chicago-based Chiasso, which sells modern home furnishings, is among the catalogers using regression tools at the zip-code level on non-catalog lists, such as magazine subscriber files, to find customers. Using a predictive zip model from Schaumburg, IL-based data and services provider Experian, the cataloger has seen as much as a 30% lift in response, says Chiasso partner Greg Kadens: “We’ll probably start taking a look at running it on more marginal lists.”

But Kadens knows Chiasso is just barely scratching the surface, considering all the precision-oriented geodemographics information now available on consumers. “We haven’t used it to the extent we can,” he says, partly because his small company doesn’t have the resources to devote to it. If you’re going to look at behavioral data, you need to be willing to spend the time to “correlate it,” he says, to determine how to change your catalog, Website, or business strategy accordingly. “That’s the next level of sophistication.”

Catalogers at that next level are using databases and modeling tools that provide more than basic age and income data. Products from Acxiom, Claritas, and Equifax, among other companies, offer detailed psychographics to provide insight into what makes consumers tick. And whereas in the past demographic data stopped at the zip-code or block level, today the information is available down to the household level.

Sophisticated catalogers use lifestyle, attitudinal, and precise geodemographic data for a variety of applications. “We’re seeing approaches that range from versioning their e-mail offers to different ink-jet messages being applied to different books or cover changes on a household level based on attitudinal information,” says Marc Fanelli, vice president of Experian’s Customer Insight Group. Such tactics can boost response 10%-20%, he says.

Getting better all the time

Geodemographics tools have definitely improved over the years, says David Hochberg, vice president of public affairs at Lillian Vernon Corp. The Rye, NY-based gifts, home goods, and children’s products cataloger uses the tools “extensively,” Hochberg says, to learn more about its house file as well as to target prospects more precisely. “We do rely very heavily on Experian because it’s essential to know as much as possible about our customers and their buying patterns and buying behavior.”

Catalogers’ acceptance of geodemographic data has been a long time coming. Mailers have traditionally shunned demographics and geodemographics tools as inferior to transactional data, Fanelli says. “These tools have been slow to gain in the catalog space, but we’re seeing that change now,” he continues. “Catalogers are embracing demographics, geodemographics, and attitudinal data to try to understand more about their customers.”

While this turnabout in attitude is significant, it’s not yet universal. “What really matters is an individual’s proclivity to buy direct, to trust a distant merchant. That trumps demographics,” says Bob Weinberg, president of Asheville, NC-based RW Consulting, which specializes in catalog and direct marketing. The information that catalogers already have in their databases from actual transactions is more accurate and generally more meaningful than the inferences they may draw from demographics or geodemographics, he contends.

Others, though, say that geodemographic and demographic data should enhance, not replace, existing transactional data. “You should explore all different technologies and combine them in ways the model providers do not,” suggests Bill Nicolai, senior consulting partner at Lenser, a circulation management consulting firm in San Rafael, CA.

For example, to find prospects, you might take a file of inactive buyers, model them using transactional data from a catalog cooperative database, then overlay lifestyle and life-stage household-level demographics data to identify clusters you might otherwise have overlooked. If life-stage data suggest that some of these names recently bought a home or had their first child, they would be good prospects for home furnishings or baby-oriented catalogs, even if they haven’t been big mail order customers in the past.

Indeed, knowing what life stage prospects are in and whether they are approaching a transition that may push them to make new purchases can be especially useful for catalogers, says Tiffany Weatherly, segmentation and analytics product leader for Little Rock, AR-based Acxiom’s InfoBase database. For a cataloger of baby products, for example, knowing that someone has just had a baby is more relevant than knowing her age and marital status.

Acxiom studies files for what it calls migration patterns — signs of lifestyle changes. For instance, having a child transitioning to the teenage years often brings new interest in cell phones and other electronics, Weatherly says. Life stage is also important, she suggests, in that many people’s shopping habits are determined in part by the cultural influences they’ve experienced. As a result, it’s likely a 65-year-old will relate differently to product offerings and marketing messages than a 30-year-old.

Lifestyle information can be particularly useful when determining which products to promote, adds Jim Wheaton, principal at Chapel Hill, NC-based Wheaton Group, which specializes in direct marketing consulting and data mining. It’s also important for content tailoring, allowing copywriters to test different approaches for different segments.

The tools in action

Most catalogers conduct basic segmentation practices on their house file using the tried-and-true recency/frequency/monetary value (RFM) or recency/frequency/monetary value/product type (RFMP). Bob Soljacich, president/chief executive of Latham SRM, a marketing agency in Oak Brook, IL, advises adding geodemographic data to the formula to ferret out “underperforming” customers.

“Customer valuation mapping,” Soljacich says, cross-references customers’ actual spending levels with their propensity to spend. While conventional wisdom might suggest your best customers are the ones spending the most with you, you might have more to gain by paying attention to customers who currently spend significantly less with your catalog but who, based on demographic or geodemographic data, resemble those top buyers and therefore have the potential to spend $1,000.

“If a customer is at the top end of the spend level, that particular catalog has a high share of his wallet,” Soljacich says, so getting that customer to spend even more might be difficult. By scoring each customer based on both actual spend and propensity to spend, you can determine loyalty and adjust your contact strategy to make it more appropriate. “You may want to spend more money on those individuals,” he says.

Using the customer valuation approach to mold your marketing strategy can lift response from your customer file by at least 20%, according to Soljacich. The process can be used with acquisition lists as well. “It helps us to choose our selects along clearly defined profiles,” he explains. “You can drill down with these lists so that you’re buying only the names that match up with your customer file. Then you can really notice a difference.”

David Huffman, managing director of ESRI Business Information Systems in Vienna, VA, suggests a similar use for his company’s new Tapestry segmentation tool. By applying the tool to a house file, Huffman says, a cataloger can determine which “clusters” of customers are more likely to purchase from the catalog or the Website. With this knowledge, the marketer can narrow its prospecting efforts.

Tapestry’s wealthiest segment, called Top Rung, consists of households with a median income of more than $168,400. “If Top Rungs are likely to buy my product or service,” Huffman explains, “I can then go buy an Experian or Acxiom file that’s coded with Top Rung [it’s not uncommon for companies to apply codes from competitors to their own data]. So rather than look at age, home value, and marital status separately, you’re looking at one code that applies all this other information about this neighborhood and the people who live there.”

In the brave new world of multichannel marketing, some demographic tools, such as MapInfo’s PSYTE U.S. Advantage, can identify customers and prospects who prefer buying online, ordering via phone, or shopping in a store. You can then modify your contact strategy and allocate resources accordingly. “You really get a different set of customers and a different set of attributes depending on how they’re shopping,” says Chris Michels, product manager for PSYTE U.S. Advantage.

Differentiating retail, catalog, and online buyers “is suddenly becoming tremendously important,” adds Experian’s Fanelli. Knowing which prospects and customers are more likely to respond to an e-mail solicitation vs. a catalog or store offer can boost response while minimizing cost. “If I want to drive store traffic, it might be a different communication than if I’m driving an Internet purchase,” he says.

Experian’s new TrueTouch product combines demographic and geodemographic data with the company’s own attitudinal information, based on surveys with 100,000 consumers nationwide, to help marketers better understand which channels customers and prospects prefer as well as what motivates them, Fanelli says.

You can also use the data to push content to customers. If, for instance, you know a customer falls into an audience segment that shows a proclivity to be an early adaptor of technology, you might try to push offers for high-end electronic products or computer equipment to that customer, suggests ESRI’s Huffman.

The same but different?

Many segmentation products now have a lifestyle component. In fact, some of the products available look so similar it’s hard to discern which one may be best for a particular cataloger. Most are based on the same or similar data, drawing on the 2000 Census, information from research firms Mediamark Research (MRI) and Simmons, and each other’s proprietary data (Acxiom’s Personicx household level life-stage data, for instance, or Equifax’s data derived from product registration cards).

At the same time, they are trying to get beyond the nuts-and-bolts information provided by the Census. “Segmentation tools that rely on the Census are good at profiling, but they can’t connect the ‘why’ someone purchases,” Fanelli says. “They don’t establish a cause and effect as to why I buy from a particular catalog.” This is where attitudinal surveys, for instance, can flesh out demographic information.

There’s no mistaking, however, that “it is a very incestuous business,” Wheaton says. “They all supply data to each other.” At the same time, he adds, “each has a specialty.” For example, one company might have more recent data, while another provides more breadth.

There are other differences, including price. Several providers home in on data at the household level, which can cost $100,000 or more for comprehensive information. Data drawn from block-group, Census-tract, or zip-code level is less expensive. A report on a single zip code can cost as little as $50.

Not surprisingly, data providers whose primary tools aren’t at the household level suggest marketers don’t really need that level of precision. For example, in family-oriented suburban neighborhoods across the U.S., Huffman says, “people do behave similarly. They’re going to soccer games on the weekend.”

But others say household-level data can make a big difference, particularly in diverse neighborhoods where age, income, and life-stage variables differ markedly from one address to the next. “We tried to attack those differences head on — by not saying all people there are the same,” says Tom Exter, director of data development for MapInfo. “We didn’t want to perpetuate stereotypes. We didn’t say, ‘Okay, if you want a Mexican cluster this is the one.’” Instead, he says, “we let the numbers speak for themselves.”

That wide range of options is nonetheless good news for smaller marketers for whom spending hundreds of thousands of dollars is out of the question. Running reports on select areas is one option, experts suggest; drilling down to, say, zip + 4 level rather than household level is another. “On average, it describes just a handful of households,” Wheaton says of the latter. “For certain offers, it can be just about as powerful as individual/household-level data, but less expensive.”


Ann Meyer is a freelance writer based in Wilmette, IL. She writes frequently on business topics for the Chicago Tribune, among other publications.

What’s Available

A selection of the latest demographic tools available to marketers:

  • Launched in 2002, Acxiom’s Personicx is a household-level segmentation system with 70 segments and 21 life-stage groups. This focus on life-stage behaviors can be useful in predicting purchasing behavior, says Tiffany Weatherly, segmentation and analytics product leader at Acxiom’s InfoBase division. Costs depend on how much data you’re seeking, but they generally start at $2,000 and can run as high as $100,000.
  • Claritas’s PRIZMNE, released in fall 2003, segments customers based on demographic and behavioral traits. It’s the “new evolution” of a geodemographic tool that dates back to 1974, says Stephen Moore, spokesperson for the San Diego-based information resources provider. The new version consists of 66 segments, including 26 new groupings. Information, based largely on the 2000 Census, is available on the household level as well as for broader geographic assignments. Prices range from a few hundred dollars for segmentation distribution and profile reports to more than $20,000 for directory licenses.
  • ESRI Business Information Solutions’ Tapestry, released in October 2003, is a step up from the company’s earlier Acorn segmentation. It’s designed to more accurately profile a company’s customer base, find similar customers, and target marketing messages. It offers Census data with Mediamark Research’s survey data and Acxiom’s InfoBase database to provide information on a block-group, Census-tract, or zip-code level. While Acorn provided a socioeconomic and demographic look at neighborhoods throughout the nation using 43 segments, says managing director David Huffman, Tapestry divides 65 segments into 12 “lifemode” summary groups based on lifestyle and life stage, and 11 “urbanization” summary groups based on geographic and physical features plus income, as well as offering custom groupings by company or industry. Prices start at $50 for a report on a single area and run up to $32,000 for data files on all geographies. A Web subscription is also available.
  • Launched last summer, Experian’s TrueTouch is a multichannel targeting tool on the household level designed to help marketers understand what motivates customers. Information is drawn from Experian’s own primary attitudinal research based on more than 100,000 completed surveys as well as from more-traditional data sources, such as Census and transaction data. Fees for analysis, which involves profiling a customer base to determine predominant triggers and most receptive channels, run about $7,500. To apply the findings as an append costs about $25/M-$50/M.
  • MapInfo’s PSYTE U.S. Advantage, available since September, is a complete update of PSYTE U.S. It combines consumer lifestyle and spending information from research companies such as MRI and Simmons with demographics data from the Census to produce a neighborhood profiling system. The tool defines 72 neighborhood clusters, up from a previous 65, based on demographics, consumer behavior, and location. Among the newest clusters are Hispanic Hopes, young, up-and-coming professionals living in a diverse culture neighborhood, and East Meets West, made up primarily of Asian-American families with children. The company also provides tailored clustering options based on a company’s specific market. Prices range from $500 for a report to $15,000 and up to purchase a software system.
    AM

More

Related Posts

Chief Marketer Videos

by Chief Marketer Staff

In our latest Marketers on Fire LinkedIn Live, Anywhere Real Estate CMO Esther-Mireya Tejeda discusses consumer targeting strategies, the evolution of the CMO role and advice for aspiring C-suite marketers.

	
        

Call for entries now open

Pro
Awards 2023

Click here to view the 2023 Winners
	
        

2023 LIST ANNOUNCED

CM 200

 

Click here to view the 2023 winners!