Traffick - Search Engine Enlightenment

Search Engine Enlightenment

Search »
 
    Home  |   About Traffick   |   Traffick Directory   |   Article Archive   |   Internet News   |   RSS   |   Contact Us

Thursday, June 02, 2005

Cookie rejection: who's right?

Webtrends, in response to a growing controversy about cookie rejection affecting online marketers' ability to track user sessions, recently released some statistics of its own based on a sample size of five billion user sessions. The company notes that cookie rejection has increased fourfold in the past sixteen months, to 12.4% across all industries. This is significantly below what a previous Jupiter Research study had shown, but is still cause for concern. It's also below the number cited in the latest study to be released. This one, by Burst Media, says that 30% of users "claim to" delete cookies.

First of all, as Webtrends Director of Product Marketing Jeff Seacrist told me last week, it's important to distinguish between cookie deletion and cookie rejection. In reality, few users wipe cookies off their systems as often as they say they do. And eliminating cookies from one's computer periodically isn't disastrous for ecommerce tracking.

Cookie rejection is actually quite common, given the proliferation of anti-spyware software that is set on high alert to reject all third-party cookies. Amazingly, some browser versions, such as IE XP SP2, default to rejecting third-party cookies.

To recap Webtrends' breakdown of cookie rejection as it affects marketers in key verticals:

  • Retail: 16.9%
  • Telecommunications: 15.4%
  • Healthcare: 14.7%
  • Manufacturing: 13.3%
  • Transportation: 13.0%
  • Media/entertainment: 12.0%
Five billion user sessions don't lie. In terms of who seems to have the most accurate stats about cookie rejection rates, I'm going with Webtrends' numbers.

More controversial, perhaps, is the needed response to this trend as Webtrends sees it. Webtrends is promoting its ability to help its clients install first-party-cookie based tracking. Good idea, but Webtrends doesn't have a monopoly on it.

Seacrist makes a number of compelling points, though. Analytics solutions vary in their ability to employ "backup sessionization methods" to analyze user behavior. Again, here, Webtrends seems to be doing a great job of that, but some vendors of low-cost analytics solutions also work on this. Kent Davidson, CEO of ConversionRuler, mentioned to me that this is part of his product's mix but it isn't touted, probably because it would confuse many customers who assume that tracking is tracking. And there are other analytics solutions which do not need to set a cookie to work. These range from advanced logfile analyzers to "rogue" methods such as installing tiny flash graphics files on a site. Seacrist points out that unusual methods such as the latter pose a danger insofar as they seem roguish enough to attract the attention of anti-spyware software. The clear message is that analytics are so important to web marketers, if there is any doubt about the accuracy of the stats, it can lead to paralysis. It makes sense to go with established analytics vendors who have a long history of working through measurement issues.

Bryan Eisenberg of Future Now Inc. and author of Call to Action: Secret Formulas to Improve Online Results, seemed to agree with the general thrust of the Webtrends news release. One key point he reminded me of is that simple tracking solutions provided by companies like Google (Google's free Conversion Tracker for AdWords advertisers) are susceptible to cookie rejection because they place a third-party cookie.

All in all, marketers should expect some inaccuracy in their conversion data. However, if they adopt a reasonably good analytics solution that includes backup sessionization methods, the distortion should not be severe in most cases. But for those who take their numbers very seriously, it's worth taking a second look at the top-tier vendors with the deepest expertise in the field. "If people start believing their site stats aren't accurate, that becomes the biggest barrier to adoption [of analytics solutions in the enterprise]," says Seacrist.

He might have added: "Just because a lot of people say they delete cookies, doesn't mean your site stats are necessarily inaccurate."

Posted by Andrew Goodman
| | Permalink

Digg this Traffick post Grab the Traffick RSS feed  

 

View Recent Posts

 

The Traffick Search Engine Directory ::
» Internet Marketing
» Internet Tools
» Search Engines
» Web Browsers
» Web Portals
» Webmaster Tools
» About the Directory
» Add URL
» Traffick Report: Flock

Traffick RSS feed

:: STAY CONNECTED ::


:: SEM 2.0 GROUP ::


Join the SEM 2.0 discussion group
1,500 high quality members, and growing!

 


:: PREVIOUSLY ::

 Recent Posts

:: FRIENDS O' TRAFFICK ::


» Battelle's Searchblog
» HighRankings
» IE Blog
» Inside AdWords
» Matt Cutts' Blog
» MozillaZine
» PaidContent.org
» Search Engine Blog
» Search Engine Guide
» Search Engine Watch
» SEM 2.0 Group
» Seth's Blog

» Yahoo! Search Blog




© 1999 - 2007 Traffick.com. All Rights Reserved

Home - About Traffick - Newsfeeds - Directory - Articles - Site Map - Send to a Friend - RSS Feeds