7 ms·

People hate that as well, as can be clearly seen in the comments on this story. It's basically impossible to do right for everyone.

• mcv 6y ago
  Honesty is always the best solution in my opinion. Send a verification email when someone subscribes. In the verification email, put a "yeah, I want this" link, and explicit "unsubscribe" link. If you also want to consider loading remote images to be valid verification, then say so explicitly in the verification email, and explain they may get unsubscribed if they don't either click the link or load the images. Explicitly unsubscribing will always override the other options. Unspoken assumptions are always going to be wrong for some people.
  • relix 6y ago
    For subscribing, you just described double opt-in and that already happens. When unsubscribing, users are very sensitive to any subsequent emails (just read the other comments in this thread). Sending an email to confirm their unsubscription might annoy a lot of the users, or will make them doubt that you're above board ("legit" newsletters boast about not sending any emails after unsubscribing, so this is a little red flag) - they will then flag that email as spam, and then we're back at square one. They also get annoyed when you ask them to confirm that they still want to read the emails. And Gmail will flag you if you keep sending the emails when they're not opening the emails. I think there really is no good solution here.
    • zaudo 6y ago
      There is no perfect solution but I think annoying some users by sending them an additional email to confirm what you are doing is preferable to annoying some other users by taking actions without informing them. Honesty really is the best policy.
      • relix 6y ago
        The problem is that doesn't stay contained for those few users. If a certain number of users mark your confirmation email as spam, your reputation suffers and there's a higher likelihood that suddenly your emails get sent to spam for all other users, because of that reputation hit.
        • mcv 6y ago
          If sender reputation is tracked that way, maybe reporter reputation should be tracked the same way. If someone reports a legitimate confirmation mail that they signed up for, as spam, then it's the spam report that shouldn't be taken seriously. Overeagerly classifying legitimate email that someone subscribed to as spam is no better than classifying spam as legitimate. Especially on a system like GMail where one person's incorrect classification would lead to other people not receiving their subscriptions in their normal mailbox.
          • relix 6y ago
            Certainly. But that is not how it currently works.
        • hedora 6y ago
          I’m continuously surprised to learn about all the little ways Google makes my internet experience worse, even though I don’t use their services. Is there a reliable way to tell if an account is forwarding to Gmail? That way, you could reserve your spammer tools (not judging you for using them) for gmail users, and treat the rest of your subscribers with the respect you clearly intended.
          • relix 6y ago
            Unfortunately I don't believe this is limited to Google. For example Yahoo lists the same best practices on their bulk sending best practices: https://postmaster.verizonmedia.com/best-practices https://postmaster.verizonmedia.com/best-practices
  • hamlsandwich 6y ago
    Part of the problem is that so many of the 'actions' people take with email are invisible to the sender: If the recipient marks you as spam, filters you to junk, blocks you as a sender or their client/provider auto-files you away somewhere, you as the sender get no indication of that. People won't click links in emails they've never opened, let alone read. 'Verification' tends to decay over time - even if I clicked that "yes, I definitely want this" link 6 months ago, it doesn't mean I haven't junked you since. The reason "You haven't loaded images for n months" is used as a signal is that there's a cost in sending unwanted email to people, and there's often no other way to know if you're wanted or not.
    • achairapart 6y ago
      What I do with the email is invisible to the sender and it's ok. It's actually the way it should be. I clearly expressed interest in your newsletter by confirming the double opt-in. It's also ok if I open your newsletter only once in a while, I may be busy, or I only want to read your content when its title click something in me. The most successful newsletters respect this and did so for so many years. They never messed up with my subscription. I will never mark them as spam because I trust them. And there is a clear unsubscribe link in every one of them that I can click if I change my mind. This is nothing new, Permission Marketing from Seth Godin is what, more than 20 years old now? Instead, automation looks always fun and clever, until your growth-hack goes wrong.
      • FlownScepter 6y ago
        > Instead, automation looks always fun and clever, until your growth-hack goes wrong. "We found we can increase the number of people subbing for the newsletter by skipping double opt in" "Why are we getting so many spam flags?" Because the increase you gained was all the people who didn't want it, cosmic brain. The people who wanted it already had it.
      • tomlagier 6y ago
        This is a bit of confirmation bias, I think - successful newsletters have enough reputation that the reputation loss from low open rate or higher spam flag rate is offset. Newer newsletters don't have that reputation, so they are more heavily penalized for low open rate or higher spam flag rate. Thus, they need to react more quickly to users invisible actions or risk damaging their reputation. I don't think this is a growth hack, I think this is a necessary action as a new entrant in a very unforgiving space - see some stories upthread from legitimate newsletters. While your actions are invisible to the sender, they are _not_ invisible to your email provider, and ultimately they are the ones who make reputation decisions that can destroy a newsletter.