How Spam Scores Work
The Spamfire approach
Most anti-spam solutions apply a simple, black-or-white test to incoming messages. If a message violates a single filter, it is marked as spam and the remaining filters are skipped.
Spamfire takes a different approach. Spamfire applies every filter to build an intelligent Spam Score for each message. The more filters a message violates, the higher its Spam Score will be. You can decide how high the Spam Score must be for the message to be considered spam.
How the Spam Score is calculated
The Spam Score can range between 0 and 100. Incoming messages start with a Spam Score of zero. Each time a message violates a filter, its Spam Score increases.
Different filters can increase the Spam Score by different amounts. See this example.
Some messages have Spam Scores even higher than 100. They are almost certainly spam (like an MLM business opportunity to lower your interest rates by working from home and accepting credit cards to sell pornography and save the wife of the former commander in chief of Nigeria).
The Spam Score Threshhold
After Spamfire has calculated the final Spam Score for a message, Spamfire checks the Spam Score Threshhold to decide whether to accept or reject the message.
The Spam List
Spamfire stores the spam on the Spam List. You can sort the Spam List by Spam Score. This makes it easy to check Spamfire's accuracy, because the messages Spamfire is least certain about appear at the top of the list.
Showing Filter Violations
If you are wondering why a particular message's Spam Score is so high, you can easily see its Filter Violations.
Using the Friends List
Messages from senders on your Friends List always have a Spam Score of zero, and so are always accepted. The Friends List is a crucial part of the Spamfire filtering strategy. Keep it clean and up-to-date.
Using Negative Score Filters
You can create filters with negative scores. When a message violates a negative score filter, its Spam Score is lowered and Spamfire is more likely to accept the message. See How to Make Negative Score Filters.