Note: this feature requires a commercial DeskNow license. It is available in DeskNow Lite only for a trial period - more information
Spam emails are all those unsolicited emails that are usually sent to millions of addresses at a time, often containing commercial offers. They usually cause a great waste of your time and energies, because every day you have to sort through them to delete them, in order to access your "proper" legitimate emails.
DeskNow helps you filtering out spam email in many different ways. It is very important to understand how DeskNow helps you in fighting spam, to improve your mail experience.
Spam detection
It is usually difficult for a computer program to classify a message as spam or non-spam without uncertainty.
DeskNow uses two mechanisms to determine if a message directed to you is spam:
DeskNow analyzes the text of the message. Spam messages usually contains common terms (like 'life insurance' or 'viagra'). DeskNow uses a statistical algorithm (called Bayesan) to calculate the probability that a message is spam. This probability is a number between 0 (very unlikely to be spam) and 1 (very likely to be spam).
DeskNow checks the address of the server that has sent the message. If this is server is known to be used/open to send spam, the message will be classified as spam. In probability terms, a message that comes from a spam server is assigned spam probability 1 (highest).
Spam filtering
Once DeskNow has calculated the spam probability of a message, if the probability is higher than a certain threshold, it will store the message directly into your 'spam' folder. The message will appear as read, so that you won't be bothered by it.
You can set your preferred probability threshold in Preferences/Mail . The default value is 0.7, which means that all messages with spam probability greater than 0.7 will be moved to the 'spam' folder.
Messages are removed from the spam folder automatically after 14 days (you can change this in the Preferences).
Tip: to understand why a particular message was or was not classified as spam, check the message Headers. DeskNow adds the following headers to all the messages that it analyzes:
X-DN-Spam-Blacklisted-By: xyz - this header appears if an external DNSBL service has blacklisted the server that sent this message
X-DN-Spam-Probability: x - this is the spam probability as calculated by the Bayesan tatistical analysis (if the message was blacklisted, the probability is implicitly 1)
X-DN-Spam-Blacklisted-By-SURBL: x the domain or internet address x is blacklisted by the SURBL service
X-DN-Sender-In-Whitelist: YES - this header is added if the sender belongs to your whitelist or contacts (see below)
X-Spam-Flag: YES - this header is added if DeskNow has classified the message as spam
Important: as mentioned before, spam classification is not an exact science. We recommend checking periodically your spam folder to see if any legitimate message has been wrongly classified as spam. Please read below on how to reduce the risk of this happening.
Whitelists
To minimize the risk of false positives (legitimate messages that are wrongly classified as spam), you can tell DeskNow to never classify as spam messages that are sent from particular people or domains. This is called a whitelist, and can be configured from Preferences/Mail/Whitelist.
Tip: if you often receive emails from a customer company abc.com, you can add '@abc.com' to your whitelist to make sure that their messages will never be classified as spam.
Note that all the people in your contacts are automatically whitelisted, which means that mail received from them will never be considered spam.
Improving the statistical filter
The statistical filter of DeskNow can learn from your feedback. If you receive in your inbox a spam message, click on the 'This is spam' link. DeskNow will then analyze the message again, and adjust its internal parameters learning so that the next time a message with similar words arrives, it will be more likely considered as spam.
In the same way, if you receive a legitimate message but DeskNow wrongly considered it as spam, click on its 'This is not spam' link. DeskNow will:
add the sender to your whitelist
analyze the text of the message and adjust its internal statistical parameters
move the message back to your inbox