On 10/10/2009 11:27 AM, Marc Perkel wrote:
I'm thinking about starting a service to filter spam on outgoing email.
I was wondering if anyone has any experience doing this and has some
advice on how to do it. These customers will be businesses, not freemail
customers, and one of the only real threats is if someone gets hacked or
has some kind of web form that gets abused.
The advantages for customers would be that many of them have dynamic IPs
or static IP names that look dynamic and are worried about being
blacklisted. And I'm hoping that by tracking who they send email to that
I can match up replies and white list them.
Does this sound like a good idea? I'd like to hear from someone who is
doing this or people who have ideas about it.
Thanks in advance for your advice.
Some customers might find other aspects useful as well:
* If suspicious looking mail is outgoing, then it is highly likely an
infected host on their network. Part of a useful service would like an
intrusion detection system telling the customer the addresses of their
problem hosts.
* If outgoing mail is below a certain score you could auto-sign with DKIM.
Warren
