Are you a service provider or a single company ?

Users as individuals, or users as a corporate email system ?

 

Who is having the problem :

Mostly a single user ?

Once you get on the list of spammers, there is no getting off, it will only increase. Even with the best filtering. Its like trying to catch water with a strainer, and a bunch of paper towels lining the strainer. Eventually you reach saturation and its starts flowing.

 

We have a few clients that complain about spam, but are constantly signing up for "free" things like daily inspirational quotes, publishers clearing house contests, and a bunch of other things.  They cant seem to understand the relationship about signing up for stuff and then getting bombarded with junk.

 

I put together a plan for one of our clients (a school) to migrate everyone to new email addresses on the domain as many of the staff were getting bombarded with junk. Its a strict plan that i can share with you the procedure. Part of the plan includes user training, and a list of new policies and in addition an "acceptable use policy" explaining that the companies email is to be used for company/business purposes only, and not personal things. ( "airline miles", "daily coupons", etc)  And if they do need to sign up for something a new temporary email address is created for it till it is determined to be legitimate.

Its a little extra work, but it makes a tremendous difference.

 

Curtis,

 

We use SmarterMail as our internal corporate email system. Our users are our employees.

 

We have multiple domains but only one that our employees use and it's the one that gets spam.

 

My boss (the CEO) gets the most spam, but his email has also been around the longest (since the 90s). I expect him to get more spam than our other users if for no other reason than how long his email address has been around, but we have a few other users that also get quite a bit of spam.

 

I started off using Bruce's antispam settings PDF, but when this wasn't enough I upped the weighting of both the bayesian filter and UCE Protect Level 2, so that failing either one would send a message straight to the spam folder.

 

We are getting a lot of messages that are obviously spam, that any human user would immediately identify as spam, but they're still getting through.

Curtis,
Take a look at Protected Sky (psky.me) This is a newer RBL that seems to have a low false positive and has worked well for us.

 

Name  Protected Sky - Yellow
Description Protected Sky - Yellow
Weight  10
Hostname bad.psky.me
Required Lookup Value(s) 127.0.0.3

 

Name  Protected Sky - Red
Description Protected Sky - Red
Weight  30
Hostname bad.psky.me
Required Lookup Value(s) 127.0.0.2