What is the SPAM Score?

Mailix automatically checks all incoming and outgoing messages for characteristics of unwanted messages (SPAM) and assigns each message a numerical SPAM score, which is initially “0” and decreases if the message has characteristics that do not indicate SPAM or increases if characteristics are found that are often present in SPAM messages.

How is the SPAM value calculated?

The SPAM value of a message is determined by a number of complicated checks. The checks aim to detect as many unwanted message features as possible while avoiding false positives, so as not to classify wanted messages as falsely unwanted.

Internal methods for SPAM checking

We use both self-developed and commonly known and available methods and checking algorithms for SPAM checking, which are performed exclusively on our own servers. SPAM checking of the content of an e-mail message is always performed internally only.

External methods for SPAM checking

We also use external service providers for SPAM checking with so-called “blacklists”, which collect check results from several mail server operators and offer them to other mail server operators. This means that a SPAM sender who has attracted negative attention from one mail server operator can also be quickly discovered by other mail server operators and penalized accordingly.

Without exception, external service providers can only obtain the IP address of the mail servers involved in transporting the message, the sender e-mail addresses associated with the message, and domain names of the sender addresses and of included Internet addresses. As a rule, the external check for SPAM is anonymous. Under no circumstances will the plain text of messages be passed on to external service providers.

Storage of the SPAM value

The SPAM value is recorded for each checked message in the header data, both as a value and with an overview of how this value was determined, and is forwarded to the querying mail client or the accepting mail server.

The following header data is entered into the mail header by Mailix after the SPAM check:

• X-Spam-Status
  contains yes/no, the spam value as a comma number and the exact calculation of the SPAM value
• X-Spam flag
  YES / NO
• X-Spam score
  3-digit comma number of the SPAM value
• X-Spamd bar
  one “-” per negative value, one “+” per positive value
• X-Spam level
  One “*” per positive SPAM value

Using this header data, any automated action can be taken in an email software. For example, markings or moves to mail folders.