My company sends transactional email from a real address as well, but there are some problems with this approach. As the post mentions, we get a ton of email intended for other people (like if we send "John Doe assigned you a task" and the person responds to us thinking John Doe will get it) and sometimes there's very sensitive or embarrassing information in the replies. I'm never quite sure how to deal with this because if I forward it on to the intended recipient, then they know I just read their personal email.
This is mostly just a problem of perception. I know that it's not my fault that I read an email which was sent to me, and I know that I can be trusted with the sensitive information. The problem is that my customers might not trust me as much as I trust myself, so by forwarding this mail on, I raise unjustified concerns about our security and trustworthiness.
On the other hand, one out of every hundred replies actually are meant for us, and our customers love the fact that it's so easy to reach us. That's why we don't use no-reply addresses, but it's not as easy of a decision as some people seem to think.
We get some of that as well. It's not enough that it's a problem yet, so we just handle it manually, but I've imagined routing it automatically. E.g., if 99/100 of replies to email type X are to John Doe, I'd probably just automatically forward all of it to John Doe. Or make John the Reply-To address, like LinkedIn does with invitation emails.
For now, you could just fake the auto-forwarder and see if people like it. E.g., forward the message on with the kind of From line and automatic header that you'd use on an auto-forwarder.
It's probably dependent on the type of site, then. If you're doing CRM, then that's very different than, say, updates on an order I just placed at an online merchant.
Yeah, this is only a problem if you're sending a notification on behalf of another user, like if tasks/events/contacts/leads are assigned by one user to another. I'm considering fixing this by auto-forwarding any replies, but then that has the same problem as a no-reply address in the event that a user actually does want to reach us.
There's also another problem which is that tons of people have auto-replies for every single email they receive (not just when they're out of the office). I filter them all out, but I'm always concerned that there could be a false positive.
That makes sense. If my app was dealing with, say, medical records info, we would assuredly not want to forward mis-replies... in that case it would make more sense to use an auto responder that might say something like "To protect the security of your personal info ..."
Because in our app, if there are bounce-backs or deliverability issues we want them to come to us, not to our user who is trying to send an email blast.
Correct me if I'm wrong but aren't bounces sent back to the "FROM" of the envelope? The mail server does not care much about the "From:"-email-header and much less about "Reply-To".
This is mostly just a problem of perception. I know that it's not my fault that I read an email which was sent to me, and I know that I can be trusted with the sensitive information. The problem is that my customers might not trust me as much as I trust myself, so by forwarding this mail on, I raise unjustified concerns about our security and trustworthiness.
On the other hand, one out of every hundred replies actually are meant for us, and our customers love the fact that it's so easy to reach us. That's why we don't use no-reply addresses, but it's not as easy of a decision as some people seem to think.