Zero False Positives

The key to achieving zero false positives is in PerfectMail's reputation engine. By watching and recording e-mail activity, PerfectMail learns the identity of your protected users, their regular e-mail peers, valid SMTP senders, black listed SMTP senders, which machines are conducting e-mail scans (harvesting e-mail addresses or looking for open relays) and much more.

For most organizations, the most important messages are messages from people your staff e-mail! Said another way, if I e-mail you, I certainly want you to be able to e-mail me! PerfectMail looks for these two-way e-mail relationships (A mails B and B mails A), and when if finds them it strongly favors them.

The result is that it is almost impossible for PerfectMail to reject an e-mail from someone who your staff has e-mailed (unless they send a virus). This is critical to business communications because:

• If you e-mail someone outside your company, their reply is almost guaranteed to be delivered back to you
• Your regular e-mail peers will never be rejected
• Consequently, PerfectMail provides consistently accurate filtering

Yes, it is possible for PerfectMail to reject a wanted message, but it is very rare. For PerfectMail to reject a wanted message, the message would have to:

• Come from someone with no prior history with your users
• Originate from a very poorly configured e-mail server
• Contain unverifiable information (Sender, Reply-To & Return-Path fields)
• Contain content that is very similar to content sent by spammers (heavily marketing oriented)

Since PerfectMail never rejects messages from known senders, and since legitimate senders would rarely ever send such a poor quality message from a poorly configured mail server, we say that PerfectMail provides effectively zero false positives.

The False Positive Dilemma

False Positives are the bane of antispam customers and vendors. The problem faced by false-positives usually plays itself out this way.

• Customer obtains an antispam solution hoping to cut down on spam and unwanted content.
• Customer installs and configures the tool according to their requirements. In many cases, the customer is forced through configuration dialogues that they don't fully understand (after all, isn't that the vendors job?).
• Customer set the spam filter to, what they belive to be, a reasonable setting that should eliminate most spam.
• To the customer's suprise, not only does the filter eliminate spam but it also blocks legitimate e-mails from known peers including customers and other important individuals.
• Blocked legitimate e-mails result in reduced levels of customer service and possibly lost business.
• Reluctantly, the antispam customer lowers the settings of their antispam tool. This ensures that legitimate e-mail is restored, but it also opens the gates to more spam.
• Customer is left with an expensive solution that is operating at only 10-30% efficiency. Spam is still a problem.

The PerfectMail Solution

At XPMSoftware, we know that correctly handling legitimate e-mail is even more important than blocking spam. We designed PerfectMail to quickly and accurately discriminate between unwanted e-mail from unknown senders and valuable e-mail from your established e-mail partners.

PerfectMail accomplishes this task through adaptive learning. Simply stated, PerfectMail watches all e-mail traffic and quickly learns who e-mails whom. Once e-mail relationships are established, PerfectMail treats the relationship like a White List (where each peer is automatically trusted by the other).

PerfectMail can discover e-mail peer relationships between active e-mail peers in as little as a few hours to a few days. Best of all, PerfectMail does this without any user intervention so your users won't have to waste their time trying to manage their e-mail partner list themselves.