These filters often stop legitimate marketing email, such as your broadcasts to opt-in lists, from getting through. Or they divert the email you're sending from the recipients' in-box to a separate bulk mail folder that people rarely open. That's why the test email accounts you’ve set up at AOL, Yahoo, and Hotmail sometimes don't receive your own marketing messages.

What can you do to make sure your legitimately sent, permission-based messages to prospects and customers make it past these filters to arrive in the in-box?

First, learn the rules:

Email filters can work in a variety of ways. Some automatically delete possible spam, while others divert it to a bulk folder. Some merely tag probable-spam messages either by color or by assigning a score to them based on content, and then allow the user or corporate data center decide what to do next.

The rules of e-mail filtering aren’t carved in stone, and they’re definitely a moving target, but there are some practical steps you can take to keep from seeing your campaign deleted before it arrives in inboxes.

Three things that alone will probably get your email filtered as spam:

1. Sending your messages by using BCC, rather than TO each individual recipient. This is probably the biggest spam signifier of all. (Note: many commercial PR services do this.)

2. Sending your messages using “open relay,” meaning that you send from a public relay, rather than using your own SMTP (outbound) mail server.

3. Any monkey-business with the TO or FROM addresses. Send from a real account on a real server to the individual recipient.

Unfortunately, you might not have control over some of the technical elements of the software that sends your messages, so send IT over to http://spamassassin.taint.org/tests.html to make sure that none of the technical components of the message violates the most commonly used rules.

Spam filtering software also views the words in your message to see if it resembles known-spam copywriting styles. So, try to avoid these five types of copy in your message:

1. Including bogus disclaimer text, such as “this is sent in compliance with …”. Legitimate opt-in lists don’t have to explain why they’re legal.

2. Including “This is a one-time mailing” in the body of the message. Even one-time mailings to lists that you don’t have permission to mail is SPAM. Doing this even once can get you added to blacklists.

3. Including $$$ in anywhere in the message or header.

4. Certain word choices can get your mail filtered as spam. For example, some filters look for specific terms, such as “millions of dollars,” “direct email,” and “bulk mail.”

So, for example if you are sending a legitimate press release for a company that is moving from bulk mail to direct email, hoping to save millions of dollars, the filters might tag your message as spam based on those keywords. So, find alternate ways to say these things, when possible.

5. Using the words, “Call now,” or “Fax in your order,” and including a phone number -- particularly 800 or 888 -- can also get you filtered as spam.

Test the rules for yourself …

Wayne Porter, VP of Product Development for Afftrack.com, suggests that you look at the distribution list for your own opt-in list. “Find the five or so domains with the largest percentage of recipients, then send test messages to test accounts at these domains to see what gets filtered.”