On 6/3/20 08:59, Robert L Mathews via mailop wrote:

If large ESPs who consider themselves reputable added a consistent
header saying whether the subscription had been confirmed ("double opt
in"), that could be used in combination with DKIM source verification to
provide positive and negative filtering.

For example, if Mailchimp added "Subscription-Confirmed: Yes" or
"Subscription-Confirmed: No" to each DKIM-signed message they sent, and
I as an ISP believed they were doing that accurately, I could reward
confirmed mail from Mailchimp by whitelisting it. (And penalize
unconfirmed mail if the user wished it.)

One problem is that spammers lie. If someone comes to an ESP and claims that their list is 100% COI, the ESP either has to trust them or the ESP needs to reconfirm. Nobody wants to reconfirm as it will reduce the size of the list. Same issue if moving from one ESP to another.

If the ESP chooses to trust the customer and the customer is lying (or clueless, "There are two boxes for email address on the sign-up form and they need to match just like a password, so it's confirmed!") then that ESP's mail including spam gets tagged incorrectly as COI.

Then Mailchimp and their customers could decide for themselves whether
the improved inbox placement is worth the claimed hassle of double-opt-in.

A compromise could be in the first few messages sent from a new ESP to have two links: "Click here if this is a list that you want to continue receiving" and "Click here if you never subscribed to this list and want to be removed."

Those clicking the first get moved to the COI category. Those clicking the second get unsubscribed and if there is a critical percentage a reputable ESP will fire the customer or require 100% reconfirmation. Those who do neither stay subscribed without the COI tag.

--
Jay Hennigan - j...@west.net
Network Engineering - CCIE #7880
503 897-8550 - WB6RDV

_______________________________________________
mailop mailing list
mailop@mailop.org
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop

Reply via email to