Every user customizes their own mailbox

People don't agree on what counts as spam or where a message belongs. A newsletter one person reads every morning is junk to the person in the next office. SpamFoo learns each person's choices and applies them to their mailbox only.

A correction takes effect right away

When a user marks a message as spam, rescues one from junk or moves it to another Email Classification, SpamFoo applies that choice to their mailbox at once. The next similar message goes where they put the last one.

  • No waiting for an admin or a model update
  • It doesn't change anyone else's mail
  • Works for spam and for Email Classification
Before
InboxRewards Desk · Claim your points before Friday
After Dana marks it as spam
JunkRewards Desk · Claim your points before Friday
New Trained Rule in Dana's account
Emails from rewards-desk.exampleSpam
Before
JunkMaple Street Pharmacy · Your refill is ready
After Dana rescues it from junk
InboxMaple Street Pharmacy · Your refill is ready
New Trained Rule in Dana's account
Emails from maplestreetrx.exampleNot spam
Before
PromotionsNorthwind Supply · Receipt for invoice 4471
After Dana moves it to Transactions
TransactionsNorthwind Supply · Receipt for invoice 4471
New Trained Rule in Dana's account
Emails from billing.northwind.exampleTransactions

Spam Rules for the senders people name

Users add their own Spam Rules on their My Rules page to always treat a sender as Spam or Not Spam. A rule can name an email address or a whole domain.

Domain admins add rules for their whole domain, and system admins add rules for the whole server. Users see those rules next to their own, but only an admin can change them.

Every correction becomes a Trained Rule

Users can see what SpamFoo learned from them. Each correction shows up as a Trained Rule on their My Rules page, with the result it gives and how many times it has been used.

Users can remove any Trained Rule, or clear all of their personalization and let SpamFoo start learning their mailbox again. If a user later corrects a message the other way, older choices that disagree lose weight and are eventually removed.

Phishing checks still come first

A user's own rules and their domain's rules are checked first. What SpamFoo learned from the user is checked after the phishing checks and before sender reputation and the models.

That way a user can't accidentally teach SpamFoo to trust a phishing message, but their choices still come before a general score.

One person's correction helps everyone

A correction also helps train the models that every SpamFoo installation uses. When users at many installations report the same new scam, the next models learn to catch it, and so do the installations that never saw it.

Before a correction is shared, SpamFoo removes personal information from it. Read how one correction teaches a mailbox for the details.

User Customization comes with every SpamFoo license

It works for antispam and Email Classification, and each user manages it from their own dashboard.