The Mark as spam functionality allows you to mark unwanted or unsolicited emails as spam, and teaches the Jellyfish anti-spam system about preferences for your particular mailbox. Once you mark a message as spam in your Private Email webmail, it is moved to your Spam folder to keep it separate from your legitimate emails in the Inbox folder:
Spam filtering system learns from this action and analyzes the spam email to identify similar emails in the future. Later, such emails may be automatically filtered into the Spam folder for your particular mailbox.
Please note that the system is trained per mailbox, meaning that if some other users mark an email as spam, this won’t affect the preferences of your particular mailbox. This is done specifically to ensure that the preferences of one user don't impact others - even within one domain (i.e. organization, team, etc) - and users can tailor their email flow to their own preferences.
What happens next?- When an email is marked as spam by a user in Private Email Webmail, it is passed to Jellyfish filtration system.
- Jellyfish then breaks the email into several tokens (small encrypted pieces that the system defines itself automatically).
- Those tokens are saved into the database with the corresponding label - spam or not spam. The system also links them to the mailbox they came from.
Therefore, even if you delete emails from your Spam folder, this will not affect your markings as the system already saved the emails marked as spam as tokens.
What happens when you receive a new email?- When an email is received to your mailbox, Jellyfish checks the database for saved tokens for your particular mailbox.
- If there are any tokens found, the system then compares the existing tokens in the database with the ones in the new email.
- However, if there are no known tokens found in the database, the system adds new tokens to it with the corresponding label (spam or not spam) depending on the user markings.
The more tokens in the new email match the ones in the database, the higher the probability that the new similar emails are marked as spam by the Jellyfish system. This is because it affects the spam score of the email.
NOTE: A spam score is assigned to an email after it passes Jellyfish filtration and is based on a lot of different anti-spam checks. The higher the score - the higher possibility that an email is delivered to the Spam folder or even rejected by Jellyfish filtration.
If a lot of tokens from the database match the ones in the new email, this adds points to the overall spam score of the email. This means that future similar emails will most likely be defined as spam by the Jellyfish system and, therefore, placed in the Spam folder.
On the other hand, the Not spam button allows you to unmark emails wrongly marked as spam. You can also mark emails as not spam if the system incorrectly placed an email in the Spam folder. Once you mark some particular email as not spam, it is moved to your Inbox folder:
Frequently Asked Questions
Will marking an email as spam prevent similar emails from reaching my Inbox in the future?
Yes, by marking emails as spam, you can influence whether similar emails might be delivered in the future. However, this process is not an instant and does not guarantee that the next similar email will be delivered to the Spam folder. The more you train the system - the more you increase the chances that similar emails are delivered to the Spam folder in the future.
This is done specifically to ensure that the anti-spam filtration works effectively with new spam patterns or even if spammers use different domains/emails to bypass the filter.
What should I do if I want an instant block for some sender/email?
If you need to instantly block some sender or subject, you can opt for creating blocklist rules in the Jellyfish admin instead. This will ensure that emails from this sender or with subject are not delivered to you in the future. You can check this guide to know how this can be done in your Jellyfish admin: How to manage Jellyfish rules.
What should I do if I accidentally mark a legitimate message as spam?
If some email was mistakenly marked as spam, don’t worry that the system will now treat it as a spam one. It’s possible to correct your labels by marking the email as non-spam. The system will remember that the email is actually a non-spam one and add necessary ham (not spam) tokens. This, in its turn, will ensure that similar messages are sent to your Inbox in the future.
NOTE: Instead of clicking the ‘Mark as spam/not spam’ button, you can also use the drag-and-drop option. If you’d like to mark an email as spam, move it to the Spam folder and if you’d like to mark an email as not spam, move it from the Spam folder to Inbox (or any other folder except Trash).
What should I do if I receive false-positive or false-negative emails?
Sometimes, you may face cases when valid messages are incorrectly identified as spam, or vice versa - actual spam messages may be delivered to your Inbox. Here are the tips on how to fix this:
- Keep training the system for wanted and unwanted email patterns by using the Mark as (not) spam button or drag-and-drop emails from Inbox to Spam folder and vice versa. The more the system learns, the more accurately it works for your mailbox.
- If you consistently receive legitimate messages from certain contacts that are being marked as spam, create an Acceptlist rule via the Jellyfish interface. This will ensure that future messages from desired contacts bypass the spam filter and are delivered directly to your Inbox.
- If you get too many false positives due to incorrect system training, please contact our Support Team via Live Chat or submit a Support Ticket.
NOTE: We recommend that you use the 'Mark as spam' button in Private Email webmail only, to ensure that the Jellyfish system is trained. Clicking the 'Mark as spam' (or similar) button in your email client to mark certain messages as spam doesn't train Jellyfish due to the individual settings of different email clients. However, you can drag and drop messages to the Spam folder in your email client and this will train the Jellyfish system.
That's it!
