I’ve used Cloudmark’s SpamNet since it was in early beta. I think it is one of the best anti-spam products out there. In a typical day, I get about 150 spam messages and SpamNet removes all but 3 or 4 of these. I don’t have a spam problem anymore.
Well, as a consumer I don’t have a spam problem. But as someone who sends e-mail newsletters to people who have subscribed at my site, it is a very big problem.
Overaggressive spam filters continually block legitimate e-mail communications, primarily newsletters and other corporate communications which can look “spammy” even if they are not.
In fact, the only problem I’ve had with Cloudmark is that it traps a fair number of legitimate newsletters I’ve signed up for as spam. This happens because Cloudmark users “vote” on whether messages are spam or not and Cloudmark then uses Bayesian filters to block similar messages from other users’ inboxes. This works well until a bunch of people decide that news.com’s newsletters aren’t worth reading and they “block” them.
Cloudmark got one step closer to the perfect solution this week when it introduced the Cloudmark Rating System which is effectively a global whitelisting process to avoid the blocking of mailings from people who are willing to identify themselves.
“The breakthrough email reputation system solves the industry-wide problem of false positives, or good email getting caught in spam filters. In the race to stop spam, false positives are crippling email as a viable way to do business. Ferris Research estimates the cost of false positives to businesses could be as high as $3.5 billion. Consumers, legitimate e-mailers and ISPs are all becoming collateral damage in the war against spam.”
This is good news. Now if only we could get everyone to switch to Cloudmark we’d have this problem licked!