1 Reply Latest reply on Nov 21, 2016 9:24 PM by Sanford Whiteman

    SPAM Filter & Email Threading - is there any relationship?

    Yoav Guttman

      This came up in conversation with our SDR team, who wanted to know if there was any impact on triggering a SPAM filter based on whether an email is part of a thread on the email client or is a different email chain. I do not see anything online that suggests there is any relationship between the two and logic would tell me there is not, since one is domain based and the other is more controlled by the email client (gmail, outlook, apple mail, etc).

       

      Does anyone have any insights on if there is any relationship between these two issues? Or have you noticed any connection between the two?

       

      Thanks

        • Re: SPAM Filter & Email Threading - is there any relationship?
          Sanford Whiteman

          (Q should be moved to Products methinks.)

           

          The strictest answer is that there is no difference in presumed spamminess between an original email and one that claims (via the References: header or the "poor man's threading" of a common subject line) to be part of an ongoing conversation. Such characteristics are easily forgeable and so cannot be used to confer legitimacy.

           

          However, the wider answer is that if, by virtue of being part of an actual conversation, an incoming email is a response to a earlier outgoing email to the same address, then an intermediate mail scanner may have auto-whitelisted (or auto-counterweighted) the sender based on previous outbound communications. Determining whether said outbound was legitimate is a little tricky since you have to exclude automated responses.  But it is certainly done. A scarily long time ago I wrote a utility called AUTOWHITE that did exactly this, for the then-popular IMail SMTP server.

           

          So the overall answer is that being part of a real convo can be helpful for transmitting mail. Not so much because of "threading" but because of prior behavior.