Traditional anti-spam appliances rely on pattern-based filtering which sometimes detects genuine messages as spam (so-called false positives). Therefore, they usually implement quarantines or spam folders, thus transferring the responsibility of managing spam to the customer, or end-user. This is of course error-prone and time-consuming.
There are two techniques which together improves anti-spam filtering to the extent that people literally forget about spam. The first one is in-line filtering with the possibility to reject spam, and the other is to detect spam based on volume rather than content.
Because our e-mail security gateway was designed with hosting and service providers in mind, it's essential that the product requires absolutely no maintenance. This is accomplished by implementing the entire message handling in-line; that is in the process of receiving the message. In that way, messages can be deferred or rejected in case they are detected as spam, virus, or the delivery fails because of non-existing or full mail boxes. Further, the gateway uses so-called real-time outbreak technology, detecting spam based on the volume which it's sent in rather than the content of the message. In this way, no genuine messages, which are typically sent in low quantities, are blocked.
The result is that end-user can forget about spam, and genuine sender's messages are delivered. Should a message be erroneously blocked, the sender gets an informative and polite error message, and can thus contact the recipient or the recipient's provider.
The product is used by many large hosting providers, in larger cluster processing hundreds of thousands messages per hour. It has also been awarded by Virus Bulletins's anti-spam award every month since we first participated, and ruled winner in respectable magazine's product comparison reviews.