Advertising: What constitutes a unique visitor?

I am going to post a bit more about the advertising side of online in the weeks to come. In my opinion this is where the heavy lifting is being done. Once standards are established and the ad world feels in control again, I think, you will see a rush to the net like never before.

This is a great overview of what some of the challenges are at the moment.

From Cynthia Cynopsis

The IAB issued revised Ad Unit and Rich Media Creative Guidelines yesterday, asking for input from ad agencies, advertisers, online publishers and technology vendors. The guidelines cover formatting standards for banner ads, buttons, floating ads and other rich media products. Companies have 30 days to comment via the IAB site.

The IAB's Audience Measurement Working Group also expects to release guidelines soon that will finally define many of the metrics terms publishers throw around. If only standards could be set in time for the upfronts where web metrics are expected to play a major role. Experts say inconsistencies and confusion over web measurement issues have a more adverse affect on digital ad spending than practically any other issue.

Some of the issues that need reconciling include:

* What constitutes a unique visitor? A recent comScore study found that approximately 31% of U.S. internet users clear their cookies once a month or have them cleared by default. Some users clear cookies after each use, and these users have a profound effect on web metrics. Why? A cookie is a string of text counting programs use to determine unique visits, and if they're cleared monthly or more frequently, these users are then counted as "unique visitors" each time they return to favorite sites. Measurement firms say they factor in cookie deleters when counting unique visits, but estimates vary dramatically and accuracy remains a deep concern.

* How is time spent on a site measured? The explosion of widgets, blog posts and online video has lessened the impact of page views and put more weight on the average time spent metric. Nielsen Online announced last July it was scrapping page view rakings in favor of time spent. Time spent may be an improvement, but it's not without its own issues. Many browsers allow multiple windows to remain open simultaneously and indefinitely. Nielsen says it solves this by only measuring which tab is "in focus," but who's to say you're reading what you last clicked on? It's like the TV being on, but nobody's in the room watching. Time spent rankings also tend to reward sites that are clunky and hard to navigate, thus taking more time to wade through.

* How can panel based and server based measurement be reconciled? Advertisers get confused when web metrics vary considerably between third party estimates (usually based on a panel of users) and internal estimates (based on internal server log data). When numbers diverge by a factor of 2 or more, serious trust issues ensue. Media Ratings Council audits of both Nielsen Online and ComScore should help allay some concerns.