Harvesting Customer Data From Your Site

So I was online looking for tickets to the Britney Spears show for my wife and myself. Why, you ask, would a couple in their late 30s want to see Britney Spears? Long story. I’ll explain later.

It occurred to me that if it weren’t for the Web, I wouldn’t be buying the tickets. My wife and I like concerts, but we only go if we can find great seats. On the Web I can view the theater layout and see exactly where my prospective seats are located. I absolutely spend more money because of the Web.

Scott Silverman, executive director of Shop.org, says I’m not alone.

“We have discovered that a full 36 percent of retailers’ online customers are entirely new to the retailers whose sites they are shopping,” he said. “That’s a very strong statement that the Web channel is generating incremental sales.”

U.S. Web channel sales increased 51 percent to $114 billion in 2003, according to Shop.org and Forrester Research. With more people buying online every year, the Web-buying public is no longer the early-adopting, tech-savvy bunch it once was. The broader online marketplace presents challenges when it comes to harvesting accurate, high-quality customer data from a retail site. It also makes inaccurate data a more expensive prospect.

Beam me up. site visitors have been known to use fake names and addresses when registering on commercial sites. I call this the “Captain Kirk of the Starship Enterprise” phenomenon. False or inaccurate customer records can be an enormous drain on profits.

A few years ago, I was working with a European car manufacturer that ran an innovative promotion: register on their site and access artistic shots of the latest car models. They followed up with a direct-mail campaign to all registrants.

They were shocked to discover that a full 40 percent of the registrations they received were bogus, from people interested only in accessing the photos. Unfortunately, management didn’t discover the problem until they received hundreds of returned mail packages containing the expensive four-color brochures they had shipped.

Address errors occur by accident, too. America is the great melting pot, which is reflected in its multilingual street and town names. With a full 17 percent of the U.S. population moving each year, some customers are unfamiliar with the address details of their own homes. Hurried customers may leave directionals (“Main Street North”), fractions (“50 1/2 Second Street”) or apartment numbers off of Web forms when they buy online. Gift-givers may guess at critical portions of recipients’ addresses, such as ZIP codes.

Linking cyberspace and profits. Simply put, the address is the bridge, the physical link, between cyberspace and profits. When shipping products, you don’t have a sale without a valid address. We should be doing everything we can to make it easy for customers to provide us with accurate addresses.

Are we succeeding? Clearly not.

According to a 2002 U.S. Postal Service and PricewaterhouseCoopers report, 23.6 percent of all mail sent in the United States is inaccurately addressed, 17 percent is delayed and 2.7 percent, or over 2 billion pieces, is ultimately destroyed because of address problems.

Consumer preferences in shipping, packaging and product selection also make accurate addresses critical. Free shipping is a hugely popular Web site promotion that increases sales, but also transfers the direct cost of address errors to the retailer. As the buying public becomes more comfortable with online shopping, Web sales of jewelry, other luxury items and apparel have grown.

“The bar is being raised when it comes to shipping, too,” Silverman said. “Nicer packaging is becoming critical.”

Losing expensive products, packed in expensive boxes, through the mail gets, well, expensive.

How expensive?

The Data Warehousing Institute estimates that poor-quality customer data costs U.S. businesses a staggering $611 billion annually in postage, printing and staff overhead.

Why do they persist? With customers entering their own addresses over Web sites, address errors in the company database may be even more likely to occur now than they were a decade ago.

We’ve all tried to tackle the issue. Microsoft Passport and the “Digital ID” project attempt to store customer data in central databases, accessible by all commercial sites. Yet, with users concerned about security and privacy problems, central repositories face an uphill battle.

Online retailers also are creating their own contact information repositories: customer profiles that site visitors retrieve by entering usernames and passwords. Unfortunately, users tend to forget their numerous name and password combinations and may fill out the same account information every time they buy. If a customer makes even slight changes to an address abbreviation or punctuation, he or she may create profit-gobbling duplicate records in your database.

Some companies are still trying to address the problem from the back-end with “cleansing” software. Yet the moment the cleansing process is completed, dirty address data starts re-entering the database all over again.

Automatic, real-time address validation is a solution. By validating an address as the customer enters it via your site, you ensure that you get valid, accurate data without asking customers to work too hard or provide too much information. Standardized addresses help you avoid duplicate records in your database and maintain a single view of all transactions for a given customer, making it easier to market to them in the future.

More than a destination. An address tells you much more about a person than where to send his mail. It’s also the easiest, simplest way to get a fix on your customer’s demographic. A Texan visiting a hardware site probably doesn’t need a snow blower. A homeowner from the Main Line in Philadelphia is probably not interested in used furniture. Getting accurate addresses into your database can help you organize your site around sales-centric principles.

Taking the profiling-through-addresses approach even further are new services and products that provide geo-demographical information based on addresses. Whether you’re customizing content for a customer or using third-party software to analyze your customer base, don’t bet the shop on address data that in all likelihood is nearly a quarter inaccurate. Because the address is the key to selling from the Web, adopting a simple address validation program leverages all your other investments in a Web channel.

As Web sites grow in sophistication, the need for address validation at the point of entry becomes even more urgent. The more a company “bets” on a site, the more there is to gain from taking the simple step of ensuring that the addresses it collects are valid, accurate and formatted in a standardized way.

And as for Britney Spears? Hey, my wife likes to dance.

Related Posts