Identifying customers from both a customer service and a risk management perspective is vital in e-commerce, as I examined in a previous article. But customer/visitor identification is just one part of the online business equation—it’s just as important to understand which customers return to your site, how often they return, and what features and products lure them back for repeat visits.

Tracking online traffic isn’t easy or simple because it ties into the basic browser technologies in play and creates customer trust issues. And by simply refusing to accept your site’s cookie, customers can derail your good-faith tracking efforts. But some new privacy standards could go a long way in easing customer tension.

What a mouse click means
When Tim Berners-Lee was creating the World Wide Web in 1990, his desire was to make information and knowledge available to everyone. At that time, it was perfectly acceptable to treat each browser page load as an independent event. At that point, no one was concerned where the user had traversed before landing on the current site or whether a site visitor was returning for the second or 12th time.

But e-commerce business planning creates new demands for info about that page load. To better service a customer, whether new or returning, the online storefront needs to gather information from where that shopper came from—the site or mechanism (e-mail newsletter, friend, advertising, etc.)—to tracking every click that customer makes on the site, during both the current and future sessions. E-commerce operators should know what items customers bought and when, what they left unpurchased in their shopping baskets, and reach out to the customer the next day or week when a visit doesn’t elicit a purchase. One example of this is a coupon e-mail to lure back customers, or featuring special items in advertising e-mails to that customer base.

Gathering this type of data will not only help the commerce effort, it will go a long way toward justifying site operations costs.

Making necessary distinctions
How can we distinguish one visitor from another? How can we identify repeat visitors? The most effective solution is the much-maligned and terribly misunderstood cookie—that URL trail deposited by site visitors.

A host of experts have generated lists of all the terrible things that cookies can be used to do, such as thwart site user privacy. Sites can deposit cookies on customers’ machines only if the visitor allows it, but a big change in recent years is that many sites are now requiring users to turn on cookies so the storefront can track movement in the name of better customer service.

Since cookies are under customers’ direction and influence, it’s essential that sites have their customers’ trust. Otherwise, customers can (and sometimes do) impact the ability of the site to operate effectively.

Each major browser version has its own way of dealing with these site relations.

Internet Explorer, for example, has a Privacy tab that is actually Microsoft’s implementation of the Platform for Privacy Preferences (P3P) developed by the World Wide Web Consortium. The tab lets online visitors decide what general rules will be established in accepting cookies from the site or from third parties (such as banner ads vendors) present on the site. These choices range from accepting everything to accepting nothing. The tab also provides the customer the ability to override the user’s preferred baseline policy for specific sites.

Netscape, on the other hand, provides a less granular defense against cookies. In version 6.22, the browser is offered the basic authority to designate whether cookies are accepted or rejected in general. As in version 6.0 of Internet Explorer, the customer has the option to establish site-specific choices of whether cookies will be rejected rather than rely on the general rules. A nice Netscape feature is a Cookie Viewer that breaks down the content for each cookie loaded on the system.

It’s all a matter of trust
In today’s world, e-business sites need to convince visitors that cookies can be trusted. More precisely, they need to convince customers their site’s particular use of cookies can be trusted.

A key tactic is creating a strong privacy policy describing the information collected in the course of doing business and how that information will be used. A company should also explain how cookies are used and whether third-party cookies may be requested while at the particular site. Customers need to know that the company is not collecting data without a specific purpose. If customers don’t feel comfortable about the policies, there’s a good chance that they will refuse cookies, and this can severely constrain a site operator’s ability to track movements and visits. Basically, without that type of knowledge, there’s little personalization that can be done on-site.

Carving a better tracking path
In the future, the P3P technology now hitting the radar of many enterprises could prove to be a catalyst for effective cookie use. P3P provides a structure that requires companies to restate privacy policies in accordance with a specific format and syntax, which will clearly boost site visitor confidence levels.

When using a P3P-compliant browser, customers can assess published privacy policies and provide feedback on whether privacy expectations are met. These browsers sport a scripting language, a P3P Preference Exchange Language (APPEL), which allows customers to define privacy expectations and tolerance for information collected by the site.

While more refinement needs to occur with this evolving technology, the ultimate goal is to allow browsers and servers to exchange information about the individual customer with the customer’s full permission.

Though the technology does not explicitly address most of the debated issues of using today’s cookies, garnering customer support for a site’s privacy policy could boost cookie acceptance by visitors. This clearly would facilitate any e-commerce site’s attempts to offer personalized service.

In the next article, I’ll discuss tools and technologies that can be leveraged to measure the Web site activity and how cookies enhance the utility of that information.