This is a special post for quick takes by ChristianKl. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
One of the major problem with getting marketing emails is that we lack good feedback mechanisms to incentivize companies whom we do give our emails because we do want to get some information not to spam us with other information that we don't want to receive. At the moment we have two options to punish companies who abuse the relationship. We can click on “mark as spam" or we can unsubscribe. The first version is a punishment as it means that more emails of the company end up in spam folders. Unfortunately, the company usually doesn't know the specific email for which it is punished and thus can't effectively improve their behavior. Unsubscription does work as a specific punishment but we can only use it we we want to stop getting all emails from the company.
We could have a better system:
A plugin that let's us rate the emails we are getting on a 5 point scale.
Once we rate a few emails we can have a machine learning algorithm that predicts our rating and allows us to filter out emails with predicted scores that are under a specified threshold
The company that provides the plugin for free can sell access to the scoring data to email marketers who care about whether customers welcome their messages.
In the end they do care about the fact that people buy, but the fact that marketers care about metrics like open rates suggest that it's useful for them to have more information.
A lot of emails are send out as a form of content marketing where the goal of the company is to create a trusted relationship which can be later monetized. In those cases it's not easy to measure the effects of an email on sales months down the road.
The fact that the marketing platforms have a spam score doesn't mean that the spam score accurately captures the spamminess when it comes to how annoying the email is to customers.