Multichannel Strategies Win Confidence
Business intelligence, the technique of discovering meaningful marketing trends from a mass of raw data, gave rise to customer relationship marketing, which is a way of looking across the spectrum of a company's offerings from a customer's point of view. These days, high profitability demands that marketers engage their customers fully by rewarding their loyalty beyond what is attainable through traditional CRM models.
To win customer confidence, marketers are abandoning single business model approaches such as stand-alone brick-and-mortar stores. They are turning to multichannel strategies and optimization, linking their store operations with e-commerce, catalogs and call centers.
This customer-centric approach lets companies build a complete picture of how each consumer interacts with the business across its spectrum of product or service lines. Multichannel strategy takes every possible synergy into account, building marketing initiatives to implement more effective, targeted campaigns; and business intelligence permeates all aspects of this new marketing.
The following is an example of one multichannel approach. Saturating good customers with paper catalogs is becoming ineffective. But by complementing catalogs with a second channel - the Internet - the standard catalog regains its value. Here is what happens: People open their paper catalog and order through the Internet by typing the product number into the vendor's Web site. This trend seems to show that consumers demand the satisfaction of perusing a "real" catalog to buy through the Web. We know this because a merchant that mails a catalog experiences a spike in sales via the Web.
Another surprising discovery is that a customer ordering from a catalog through traditional channels returns 20 percent of purchases. If a customer orders from the same catalog and types the product number into the vendor's Web site, the rate of returned merchandise drops to 10 percent.
The multichannel approach. To leverage this sort of knowledge into additional profits, major merchants are implementing multichannel optimization strategies. Fingerhut Cos. Inc., Minnetonka, MN, for example, is moving from a product-centric to a customer-centric approach. Large data warehouses track, cleanse, store and analyze each customer's online, catalog and store-based retail habits. The result is a greater understanding of buying patterns and a higher level of customer service.
However, reaching that level of sophistication is not easy. For example, in 1997, Fingerhut, a wholly owned subsidiary of Federated Department Stores Inc., Cincinnati, mailed 467 million catalogs in 120 editions, selling an enormous range of products to 7 million people.
Fingerhut's information technology team created more than 200 finely tuned models of consumer behavior to serve these customers. For example, it discovered that one group of customers buys more jewelry than others do, so Fingerhut created a catalog to serve that group. It also learned that couples who change their address buy more household wares within 12 weeks of moving. As a result, Fingerhut mails a ready-made, tightly targeted catalog to the "movers" group.
Optimization cuts costs, boosts profit. Sales spike after a vendor mails a catalog. In the past, that often resulted in the same profitable customer segments being targeted for multiple mail promotions.
New, multichannel CRM strategies eliminate this as vendors learn to merge promotions, thereby translating cost savings into more profit. The industry refers to this as "avoiding saturation." In the case of Fingerhut, IBM business intelligence consultants assisted the company in refining its horizontal marketing (customer-centric) approach.
The first step was for IBM consultants to thoroughly understand the issue of overpromotion. Using Fingerhut's "as is" model as our base line, we subjected it to advanced analytics, adding a number of processes as work progressed. Fingerhut subsequently increased its returns while reducing its mail stream.
The company demonstrated to its satisfaction the effectiveness of horizontal marketing. At that time, the president of Fingerhut was quoted as seeing the trend to advanced business intelligence analytics as transforming his business. Fingerhut has been able to shift the emphasis of its business from catalog to customer, featuring merchandise that specific customers are likely to buy.
Multichannels, different desires. We need to satisfy two sets of customers in the multichannel world. New dot-com-only vendors want to pull the many click streams together, tracking anonymous trends to help draw browsers to the center of their own sites. Then we have traditional channel vendors - catalogers, stores and mailing list database marketers.
These vendors are going multichannel, establishing their brands on the Internet, and refining data-handling techniques so they are consistent with their other, established channels. To achieve this, multiple-channel marketing requires not only the best science that data mining can supply, but also the best optimization.
The Internet: A valuable online test vehicle. The Internet is already a valuable tool for conducting online market surveys, giving rapid turnaround between experiment and result. For example, some major merchandisers build Web sites in such a way that if you open them several times in rapid succession, you may not see the same home page twice. These merchants use testing control to sample consumer response to different appearances and pathways. Efforts are under way to automate this process.
The science of database marketing has become so complex that IBM has pressed into service a powerful parallel supercomputer. In essence, the computer is trying to establish discernible trends from a bulk of anonymous click-stream data left by people browsing the Internet.
We have a way to go. Eventually, we hope to predict where markets will go. Traditional store designers have known for years that when customers walk into a "real" store, they turn right. E-business research has a way to go before we can predict a customer's next move in a multichannel world. But we are getting there.