Database Marketing: Mining for Gold

At the turn of this century, cataloging — and all retailing for that matter — was revolutionized by the emergence of e-commerce. Shopping at home became easier than ever, and consumers warmed up to the concept of wielding their purchasing power with the click of a mouse. Today, so many orders are coming in online that it’s becoming difficult for some catalogers to recall the times when they were all but completely reliant on their call centers for order-taking.

Your Web site, however, isn’t just a vehicle through which orders can be taken and processed more efficiently; it’s a powerful data-gathering tool that an increasing number of companies are using to build their businesses in myriad ways. It’s no secret that the quantity of consumer data that companies can gather through cyberspace is extensive, but some industry insiders say catalogers are just beginning to scratch the surface.

Ned Barrett, president of Direct Logic Solutions, a database marketing and business intelligence provider headquartered in Peoria, Ill., points out that many catalogers already are gathering a significant amount of marketing data. The issue is, when collected by an older enterprise legacy system upon which the infrastructures of many companies are based, much of this information is stored in several different places, making it difficult to access and transform into anything meaningful.

Variables, such as e-mail click data, Web surfing behavioral data and purchase history, aren’t that effective on their own. But combined, they can prove very useful. The best solution, Barrett believes, is to create a separate marketing database that externally runs to the legacy system, configured so that the data gleaned from the Web site can be converted into practical information that marketers can access.

Find Specific Elements

“The challenge is not to drown in data, but to look at the specific things that are really valuable to take action,” says Michael Baum, president of The Guild, a Madison, Wis.-based company that started out connecting artists with buyers through its Web site before launching The Artful Home print catalog of gifts to drive traffic online. “The data that come off a well-monitored Web site are valuable in different ways to different groups in the company,” he adds. The Guild uses San Mateo, Calif.-based analytics firm Coremetrics to facilitate the streamlining of information.

Phil Donahue, vice president of strategic accounts at King of Prussia, Pa.-based catalog design firm Catalogs by Lorel, illustrates an example of how this data can be applied: One Lorel catalog client, after studying the traffic on its Web site, recently determined that a product it had opted to exclude from its catalog was, in fact, proving popular on its Web site. Based on this, the cataloger folded the product back into the catalog, and as a result, both print and Web sales took off.

Another cataloger, Duluth, Ga.-based National Allergy Supply, gauges customers’ shopping habits through their page visits, explains Vice President of Marketing and Sales John Fry. He and his team focus on what visitors do once they enter the company’s Web site. Specifically, Fry and company look for the following from its customers:
u Do they visit multiple pages?
u Are they using the search function?
u Are they purchasing anything?
u How did they enter the site — through an unpaid organic search or a paid search?

Collecting data related to paid searches is extremely important when determining the return on investment related to this expense, Fry notes. If you want to be at the top of the list of sponsored links corresponding to a specific search term, such as “allergy supply,” for instance, you could pay anywhere from $.05 to $5 each time someone clicks the link to your site. “Some search terms are very expensive,” he says, “because the bidding on them has gotten out of hand.”

Organic searches don’t require payment, however. In order to be among those links at the top of the list, your Web site must be configured in such a way that the spiders — programs sent out by search engines such as Google or Yahoo! — identify the information you feature as relevant to certain search terms.

Both paid searches and organic or “natural” searches are important in the eyes of the consumer, Fry maintains. “Many people will not click on a paid ad,” he says. “When they’re looking for information, they’re not going to go to those paid links, because they know that these are people who want to sell something to them.” When they are ready to buy, however, sponsored links may be exactly what they’re seeking.

So what information, exactly, do you as a cataloger have the right to ask for? Virtually anything the consumer is willing to give up, provided you’re up front about why you’re requesting this information — and how you’re planning to use it.

Pat Kachura, senior vice president of ethics and consumer affairs at the Direct Marketing Association, says that for consumers, the most sensitive data is information about their health, family, children and finances. Most catalogers don’t need this type of information. Instead, input on their customers’ tastes and preferences regarding their products, or how often they receive or purchase from catalogs, is more relevant to the average cataloger.

“My perspective is always looking at that data from the consumer’s point of view,” Kachura says. “The data collected is useful if consumers want to share it, and if you’re going to use that data to contact consumers about matters relevant to their lifestyle.”

Best Use of Cookies

Tracking customer behavior traditionally is achieved through depositing cookies on their computers when they enter your site. Web surfers, however, have grown accustomed to this from the hundreds of spam e-mails they receive. And as a result, they erase their cookies with increasing regularity, severing their ties with the related sites. “In collecting data, the only challenge is the increase in peoples’ tendency to block cookies,” says Baum. “If someone has blocked cookies or is deleting cookies, you have less information as to who they are and where they’ve been.”

Barrett suggests that the best way to deposit a new cookie on a visitor’s computer is through e-mail campaigns. Once customers have registered for your e-newsletter, they receive it and, hopefully, click on one of the links in the message. A new cookie is deposited, and their behavior on your Web site can, once again, be traced.

The secret, of course, is to avoid coming off like Big Brother. Every Amazon.com customer is familiar with the marketer’s practice of suggesting titles that may correspond to customer preferences with its “customers who bought this book also bought these books …” suggestions. Amazon matches individual consumers’ behavior with other customers to provide the best service. “It’s up to the marketer to present those kinds of things in such a way that won’t threaten consumers,” Barrett says.

Make it ‘Coincidental’

Catherine Paschkewitz, manager of consumer marketing at HP Home and Home Office Store, says that HP makes these types of suggestions in a “coincidental fashion. For example, if you left something in your shopping cart, we might send a gentle reminder saying that you left something in your shopping cart and perhaps offer a small deal if you were still interested in buying that product,” she explains.

Following up by saying, “We know you looked at this [notebook computer],” is not the right approach, Paschkewitz says. A better way: a general, gentle e-mail alerting them about current notebook products.

At The Guild, the customer care group periodically reviews online behavior data. If it’s noted that a potential client was in the process of placing an order, and then suddenly stopped, the prospect receives an e-mail offering assistance. “When we see that, we send the customer an e-mail saying, ‘You seem to be having trouble placing your order. Let us know if we can help,’” Baum explains.

In some cases, the transaction winds up becoming a sale. “We’ve converted a number of orders by doing that, because people had just been having some kind of technical issue, or they didn’t understand what they were doing, or they had a question. And when we answer it, they’re happy for the assistance.”

National Allergy Supply outsources its data collection, which, while potentially more efficient in some ways, puts the company at the mercy of its provider. The information is supplied on the vendor’s schedule rather than on the cataloger’s. “When we did this in-house, I used to be able to pull up our back-office administration and very quickly see anything I needed to see, such as how certain e-mail campaigns were working,” Fry says. “I can still see some of that, but I only see the details once a month, on the vendor’s schedule.”

Consumers on Auto Log-in

While print catalogs continue to drive sales to a large degree, catalogers must consider how to leverage the relationship between what’s in print and what’s on screen. Today’s consumers, Fry argues, are conditioned to log onto the Internet the minute they require purchase information related to a specific product.

The more access catalogers have to the details generated by their sites, the better they can assess not only how their print catalogs are performing, but also their Web sites, as well. And, because Web sites easily can be modified, shifts in sales strategies can take place with the click of a mouse. For example, a company may load alternating homepages onto its site, wherein visitor A sees one version and visitor B views another. The details gathered from the outcome show which layout is more effective, which contributes to future marketing efforts.

Donahue emphasizes that Web data shouldn’t just be used for couponing and sales promotions, but also for timely and relevant offers. “Catalogs, by their nature, aren’t sales promotion vehicles,” he says. “So the marketer has to make sure the digital, retail and catalog aren’t eating away at the brand equity by just being another discounter. Marketers have to pursue the testing of timing of the proper product or solution that’s a good value.”

A Customer is a Customer … is a Customer …

Madison, Wis.-based The Guild, which began as a Web business and later developed a print catalog of gifts called The Artful Home, regards Web clients no differently than those who order via catalog.

“The result of our being a Web site with a catalog advertising outreach is that all of our customers are treated the same,” Guild President Michael Baum says. “We don’t really draw a distinction between Web customers and catalog customers, except in that we track who had a catalog and who didn’t. Regardless of whether a particular customer is responding to a catalog or arrives via the Web or telephone, the same data is collected.”

Baum concedes that Web technology does allow companies to view information that catalogers wouldn’t have dreamed of accessing before the Internet revolutionized the industry. The Guild works with San Mateo, Calif.-based analytics firm Coremetrics to review elements such as what pages get viewed, what products get viewed, and what items get put in shopping baskets only to be taken out. “We’ll also look at what items are bought when they go up on the Web site, before they go into a catalog,” Baum explains.

Multichannel marketers, such as The Guild, “have to realize that customers don’t think of themselves as Web customers or catalog customers,” Baum says. “They just want to get something somewhere. No matter how they come in the door, you need to treat them well and the same. And you need to be looking at the same data.”

Categorical Collection

At HP Home and Home Office Store, consumers’ total online store behavior isn’t just what the company’s marketers are interested in. They also want to know what consumers did while visiting different categories within the Web site and how they used the site’s functions. This, according to Catherine Paschkewitz, HP Home and Home Office Store’s manager of consumer marketing, enables the company to better optimize the technical features of the site while enhancing its marketing efforts.

Paschkewitz believes companies aren’t taking full advantage of what today’s Web analysis technology has to offer. “You need to think through how you need your data categorized,” she says, “and have it in a way that is easily accessible to the average person on your team. This way, it’s not something that has to be data-mined and extensively analyzed before people can see it.”

To achieve this, companies must invest in implementing the systems and enlisting individuals or teams that understand the technicalities behind data collection, and how this information impacts business objectives. “If you haven’t invested in having a dedicated person or team around your Web analytics,” she notes, “that makes it a lot more challenging.”

Carolyn Heinze is a freelance writer/editor based in Vancouver, British Columbia, Canada. You can reach her at carolynheinze.blogspot.com.