Are Unique Visitor Counts Over?

Are your unique visitors truly unique? As many as 39 percent of online users delete cookies from their primary computer each month, according to a new Jupiter Research (a Jupitermedia Corp. division) report. The survey referenced in this report found 12 percent of consumers delete cookies monthly; 17 percent weekly; and 10 percent daily.

This indicates traditional unique visitor counts are increasingly inaccurate, and possibly irrelevant.

It’s imperative to understand this trend’s effect. A user deletes or rejects a cookie. When she returns to the site later, she’s registered as a unique visitor. An inaccurate unique visitor count can potentially skew the accuracy of your Web analytics metrics, including information you may be using to sell advertising or maintain relationships with affiliates. In previous columns, I’ve discussed the importance of having accurate data for decision-making.

Look at the data from your Web analytics tool. If you have fewer returning visitors and more new visitors when nothing else has changed in your business, you may be experiencing the effects of this cookie blocking and deleting trend. All our clients experience it, and an increasing percentage of their site visitors aren’t accepting cookies at all.

Typical Visitor Tracking Methods

There are a number of different ways to track visitors. The three most common are:

• IP address tracking. This was one of the original ways companies tracked visitors in the early days of Web analytics. Though many organizations still track site visitors this way, it’s very inaccurate. It identifies visitors and visit sessions based on their IP address. Today, there can be more than one person using one IP address simultaneously, and that address can easily change from session to session. (If you still track visitors via IP address, let me know and I can give you some pointers. If you aren’t sure, check now!)
• Cookie tracking. Problems with IP address tracking gave rise to cookie tracking. Here’s how it works: When a visitor accesses a site, a cookie is placed on the individual’s machine. All movements through the site are tracked back to that cookie. Web analytics tools can string together a single session, or multiple sessions, based on the cookie. Cookies are assigned per machine, or by individual user per machine, but only if each user logs in and out of the PC. Moreover, if a visitor accesses a site from his home, office, PDA, and a public terminal, the tracking tool can’t combine those four visits into one visitor experience.
• Registration/login tracking. This type of tracking is based on an actual site login, and all movement through the site can be tracked back to that visitor through the login. For example, once you log in to E*TRADE, eBay, or Amazon.com, you not only see personal information, but can also access your account. With this approach, visitors and visit sessions can be tracked based on the login. Sounds like an ideal way to track all site visitors, right? But there’s information on all the sites mentioned above visitors can access before they register or log in.E*TRADE’s current ad campaign is trying to drive new customers to its site. That makes it important to understand user behavior prior to registration. By tracking preregistration behavior, E*TRADE can identify opportunities to improve conversion and registration performance. The company may be just as interested in customer behaviors before they open an account as after. Login tracking is ideal, but it’s typically used in conjunction with cookie tracking. (Note: If you us an ASP Web analytics solution, such as Omniture SiteCatalyst, WebTrends On Demand, HBX, or Coremetrics, it’s likely tracking based on cookies or unique log-in IDs).

First-Party vs. Third-Party Cookies

There’s an important distinction between first- and third-party cookies, especially when it comes to how spyware, browsers, and users treat them. This how Microsoft defines first- and third-party cookies:

• A first-party cookie either originates on or is sent to the Web site you are currently viewing. These cookies are commonly used to store information, such as your preferences when visiting that site.
• A third-party cookie either originates on or is sent to a Web site other than the one you’re currently viewing. Third-party sites usually provide some content on the Web site you’re visiting. Many sites’ advertising is served by third-party sites, and these third-party sites may use cookies. A common use for this type of cookie is to track Web page use for advertising or other marketing purposes. Third-party cookies can either be persistent or temporary.

Different browsers and spyware programs have different settings for how to handle first- and third-party cookies, whether rejection or deletion.

Jupiter Analyst Eric Peterson believes businesses must reexamine their use of metrics and determine which long-term measurements could be affected by cookie blocking and deletion.

“What we define as a ‘visitor/customer lifetime’ is greatly accelerated in the online space,” said Matt Jacobs, the senior analytics manager at ZAAZ (where I work). “When you couple that fact with a high rate of cookie deletion, ad networks and most Web analytic vendors are presented with some interesting and difficult challenges for RFM (recency/frequency/monetary) and lifetime value analyses.”

The full outcome of this trend on site analytics, ad networks, affiliate programs, and personalization is yet to be seen. Also of interest is the future effect of the trend: the ability to act on visitor data and use it to tune and optimize site performance. Skeptics may seek ways to minimize this issue, but cookie blocking and deletion is a reality and it merits attention. It can potentially skew data you may be relying on to make important decisions.