When we learn more about how anti-spam filters operate, we can get much closer to an answer to this question.

Email messages are reaching the mail server of our addressee. However, before the messages are saved in a particular inbox, an anti-spam application will analyse the message, and categorise it. The most popular solution used by anti-spam software is to carry out some tests, and each test ends with an allocation of a specific number of penalty points. Email will be considered spam when the anti-spam filter will allocate a certain number of penalty points to it – equal to or higher than a set limit (most frequently, the limit is 5). There are a number of tests available. They are of different significance (by number of allocated points).

Currently the most popular tests, and of the highest significance, are testing the email source; a so-called Baesian filter is also widespread. The Baesian filter is a remarkable tool, as it is a self-learning filter. It analyses the content of emails by words that frequently appear in spam messages. Each Baesian filter is operating independently – as a result, there are no email marketing mechanisms that could evaluate the probability if an email will be classified as spam.

In addition, anti-spam software will perform some of the following tests:

 - it will check the subject of the email (there are more penalty points if e.g. the subject is written in capital letters, or if there are many exclamation marks, or if the subject contains words typical for spam messages: e.g. buy, free, promotion, special offer, guaranteed, bonus)
 -  it will check the address of the sender – some senders are blacklisted (i.e. their addresses are included in spammers’ lists)
 -  it will check all headings of the email – e.g. incorrect dates are typical of spam
 -  it will analyse the email by phrases that are popular in spam messages and by the number of www links (it is also possible to check whether the links are not suspicious – if they are blacklisted)
 - it will check the structure of the email – it is not recommended to send emails composed entirely of graphics; HTML email messages should be complemented with text versions
 - it will verify the HTML code – when we are working on a newsletter, we should not paste extracts directly from MS Word, as we will automatically paste all formatting markers.

If our objective is to ensure that our emails are not blocked by anti-spam software, we need to structure our emails properly, steer clear of phrases used in spam messages, and avoid spam-like structures of the message contents. However, any of these measures will not ensure full protection. If it turns out that a specific server has blocked our message, the best solution is get in touch directly with the server administrator. The administrator will be able to provide us with a detailed explanation on why our email was considered spam. As a result, we will be able to prepare our next newsletters better – and they will not get blocked. The efficiency of our emailings will increase.

Tagi: spam