Nearly 20 percent of the emails we track either contain bad links or do not comply with accepted email standards in some way. Being non-compliant means the emails are sent in a format that does not conform to specifications developed by industry email standards bodies. Some of these errors may not be fatal, depending on how lenient a particular email program is, but it does mean that some email browsers may not be able to view the email at all. These errors include two @ signs in the sender's address, dates in the wrong format, and other formatting errors.

Some non-compliance errors may be done on purpose and actually fall more in the area of deceptive business practices. These errors include dating the list managers' email servers a year in advance. For instance, an email may be dated Jan. 15, 2005, instead of 2004. This ensures that anyone who sorts their emails by "date received" would see these emails at the top of their inbox, regardless of when they were sent. Some email providers do one better and "post date" their email servers a year and a day ahead, in order to beat the competition.

Some other errors are just plain embarrassing and fall into the area of bad quality assurance testing. Recently, our email tracking system flagged emails from Accenture as having bad links. The welcome letter that Accenture sends to thank recipients for subscribing to their newsletters contains a link to learn more about their account and to unsubscribe. Here is the text of the email:

"Thank you for registering for email alerts on accenture.com. We will email you an alert when we have a new story in a subject area you have selected. At any time, if you want to change some part of your registration or sign up for new features, you can do this here.

Unfortunately, the link to the preferences page contains the period at the end of sentence, which makes the link unworkable. If you click on the link, you get a "Page Not Found" error. Of course you could cut and paste the url into a browser and erase the period, but for a company that prides itself on being a top IT consultant, this lack of attention to its own communications is a poor introduction to its services, in my opinion.

By applying to certain sweepstakes list managers you may find yourself "opted-in" to literally dozens and dozens of their other "newsletters," each requiring its own separate opt-out. Newsletters is in quotes here, because many are not actual "news" letters in the sense that they provide editorial content, but exist only to send email marketing offers.

The Vendare Group, owners of Jackpot.com and UpRoar, have 51 such newsletters including Digital Know How, ePinPoint, Bargain Times, and many others. Without too much effort, you can get subscribed to quite a list of emails each day. Back in the fall of 2003 when I tested those newsletters to see if you could opt-out you needed to opt-out of each newsletter individually. Recently, however, I rechecked their opt-out policy in light of the new Can-Spam rules and now the unsubscribe page clearly tells customers how to get off all the lists all at once by going to WinSweepstakes.net or Jackpot.com and updating your account, which is certainly a step in the right direction.

Domain registration services such as "Domains By Proxy" allow companies to hide their identity behind an anonymous proxy account. Brands that have utilized these proxy services include online short-term loan providers such as "Payday Right Away" and "Loan-Till-Payday." Email owners who have hidden their true identity include Direct Deals and Online Offerings. Fortunately, these cases seem rare. More common, is the case of the clustering P.O. Boxes. For some reason, certain post office boxes in Manhattan and San Francisco are home to dozens of email list managers.

Witness, if you will, P.O. Box 7361 in San Francisco. The 5-Day Can Group has P.O. Box 7361 - 101480, Bestted.com Group has P.O. Box 7361-101489, Capital Lenders has P.O. Box 7361 - 101438, Gossip and News Group has P.O. Box 7361 - 101483, and on and on. Not to be outdone, P.O. boxes in Ansonia Station in New York demonstrate the same clustering phenomena. Currently, I don't have the answer for why this is, but keep reading this column. I'm working on it.