> I don't send mail that gets reported as spam in the first place.
I ran a newsletter where people had to opt in to receiving it. It was announce news for a video game. You only ended up on this list if you entered your email, clicked join list, and then clicked the link in the email we sent to you to confirm subscription. We had a big unsubscribe button at the very top of the email. We still regularly got people who hit report spam on us, presumably as a way of saying g they didn’t want the email anymore.
They're probably expecting their email provider to take that info and use it somewhere upstream of their own individual account. Which, as you've pointed out, does happen.
Maybe they don't believe that it happens often enough or something, but the thresholds do need to be reasonably high since, as you pointed out, some people hit the button whether it's justified or not. If the threshold for email provider action was too low, you'd end up not being able to send to anyone with Gmail because one guy forgot he signed up to a list (or signed up and immediately reported it as spam to spite the sender).
The person you replied to also sounds like they may be using an offline or third-party email client, though. There's a difference between a "Report Spam" button somewhere your email provider controls, and a "Mark as Spam" button in your third-party email client. I'd assume there's some kind of protocol that could potentially allow third-party clients to report it back to the email provider, but would also assume it may not be as reliable as first-party interfaces.
> They're probably expecting their email provider to take that info and use it somewhere upstream of their own individual account.
Report spam, as a generic feature? It's an okay starting point "as-is" but useless for preventing malicious use and it hasn't meaningfully improved since launch.
Specifically for google: allow users to block whole domains; I can already do that on my own mailserver, why can't I do that on Google's? Then, block mail from foreign countries -- or at least countries that I don't care about; I can block whole ASNs on my mailserver, why can't I on Google's? That then leaves only mail that I can bring legal action to.
Another iteration: when you "unsubscribe", then keep a record of it, and also show the history of emails that you've received from them on a confirmation dialog. Show me anything interesting like purchases, warranties, appointments, etc. When confirmed, keep a record of it. Show me a list of _all_ of the things I've unsubscribed from. If email is still received, automatic report spam and block the domain. Oh, that means that mailing lists must come from the same domain that sales are made on.
Another iteration: a subscription should require a confirmation. Let the email server recognize the confirmation, and block emails whose unsubscribe links aren't in the list of confirmations. That means an unsubscription link should go to the same domain that a subscription was confirmed on.
That's just a few spitballed ideas. Spam reporting functionality is clearly iterable, but it hasn't meaningfully changed for decades. It's still primarily done through opaque "reputation" scores and little else.
I don't want "report spam" which doesn't give me feedback and continues to let spam onto the wire to my client, and isn't powerful enough to use to block bad actors from trivially getting to my inbox. I don't want to be expected to (and trained to) click on unverified links which take me to somewhere I don't recognize, and could take me somewhere malicious. I expect more from the largest email provider(s) in the world.
You can either search first, then create a filter, or create the filter and then set your specific search terms/scope. The only downside of filters is that they only apply to incoming mail; you can't apply them to already existing mails, but you can use the same search terms as a kind of mask, then use the select all checkbox, then click the link to select all mail that matches the search instead of just selecting all mail on current page/view.
I’m not accusing you of this, but I will mark things as spam even if I signed myself up if what they’re delivering is just garbage.
It’s usually not newsletters for me, but small niche companies who sell very specific things and feel a weird urge to have a weekly newsletter. It’s like all they sell is 2 models of guitar capo, but they still feel the need to send me weekly updates on I don’t even know what.
The kind of things where I not only don’t want the emails, but I want to register that I feel I was misled when I signed up.
> I don't send mail that gets reported as spam in the first place.
I ran a newsletter where people had to opt in to receiving it. It was announce news for a video game. You only ended up on this list if you entered your email, clicked join list, and then clicked the link in the email we sent to you to confirm subscription. We had a big unsubscribe button at the very top of the email. We still regularly got people who hit report spam on us, presumably as a way of saying g they didn’t want the email anymore.