There are 2 main ways to reduce spam. The first is to block or filter spam messages as they are sent to your address, the second is to keep your address from having spam messages sent to it. The first way is a reactive method to try and separate legitimate email from spam. The second is a proactive method to keep spam databases from obtaining your address and is much more effective.
When it comes to blocking or filtering messages we rely on real time IP reputation services that are very expensive but can block the bulk of spam from even saying hi to our mail servers. Roughly 90% of all spam messages sent to addresses in our system are never seen because its blocked before it reaches the accounts. But some spam can be sent from legitimate sources that can’t be blocked by IP reputation. The worst offenders are free email services like Hotmail and yahoo. It’s very easy for spammers or phishing campaigns to create as many free accounts as they need to send their messages from legitimate yahoo and Hotmail servers. Other sources of spam that can’t be blocked by IP reputation are hacked email accounts of legitimate users or messages sent by viruses that get installed on a user’s computer. Those are the most desirable for a spammer because they appear to come from someone you know.
Spam that does make it past the reputation filters must then be filtered based on its content. This is the hardest part of spam filtering because we do not want to filter any legitimate or important messages by mistake. The feedback we get from users is that missing an important message because of a false positive from the filters is a more difficult situation and can be much costlier than manually deleting spam messages. Spam and phishing messages are purposely randomized from one message to the next including the spelling, phrasing, and content of messages plus what address or location the messages are sent from. This helps them bypass content filters that must rely on statistical patterns or consistent content to work.
The second method to keep spam databases from obtaining your address mostly involves not exposing your email address to Google. Other things help, but keeping Google from getting your email address is most important. If someone can find your email addresses in Google, then so can every spam database and over time it will get worse as your address is listed in more and more spam databases. Its best to remove those addresses as soon as possible and use contact forms instead or at least list the addresses using scripting methods rather than plain text. In our UnionActive system we automatically list email addresses in a scripted format for this reason.
We have tailored our system to block, filter, or tag as much spam as possible while trying to achieve a zero-false positive rate. As an individual user, you also have options available to further refine your own account filtering both at the webmail level and at your computer/software level. For example, if you notice that you get a lot of spam that contains the phrase “Miss. Charity Akika” you can create a mail filter rule in the webmail area that will delete any messages with that phrase or send them to the junk folder for further review. Or if you get repeated spam from the email address "email@example.com" you can create a quick mail filter rule in the webmail area that will delete any email from this exact address. If you are using a mail program like outlook you can also setup filtering rules there.
We take spam reduction very seriously and are constantly working to improve our efforts.