Report spam button broken: an ISP perspective

This press release has been discussed in a lot of groups and sites I read. One of my favorite comments comes from one of the filter developers at a large ISP. He was asked “does the overuse/misuse of the this-is-spam button significantly affect the ability to do your job?” His response, reposted with permission,

The customer is always right. In my opinion, there is no such thing as ‘overuse’ of the report spam button. The more feedback we get, the better. Our job is to keep the user’s inbox in the state they want it. The more they tell us what they do and don’t want, the clearer picture we get about who is sending unwanted mail. So I would say, yes, it does affect my ability to do my job in that it enables me to actually do my job.
It might cause my job to involve more detailed research into people’s preferences and what to do with mail that people disagree about, but I don’t see that as a problem.
Just because a marketer doesn’t like that we consider our users’ opinions to be more important than theirs is not really a problem either as far as I’m concerned. I’m here to serve my users, not them. They can either send mail that people don’t respond negatively to, or I can put their mail in the spamfolder. It’s not like they are going to make any money by repeatedly mailing people who think their mail is spam anyway.

If senders really want ISPs to change things, that is one of the people they are going to have to convince to make the change. It does not seem that the current methodology to effect change is being effective. Senders who want more cooperation from receivers need to start listening to him, and his peers in the industry, and start making misuse of the this-is-spam button important to them.
The ISPs are open to feedback. Just yesterday I posted the request from AOL to get feedback on how ISPs and ESPs are using the data AOL is generating. They are actively looking at how bounce rates are used in order to send clearer, more useful data back to senders (bulk senders and ISP senders). Cooperate with them here, help them improve their processes and maybe they will be more open to listening to senders in the future.


  1. DanS says

    We get FBL reports, mostly from AOL. So I get to see just what it is people are hitting the “this is spam” button for. Now mind you we’re a general hosting company that deals with small businesses, and we have several folks who do send out broadcast emails, but there’s a ton of one-to-one emails too.
    You know which emails get fed back to us via the FBL? It’s the ones where someone sent individual mail to someone else. Sometimes, it’s the “let’s meet at 4PM for that conference call.” Or it’s the stupid joke one person got, and sent on to one friend. This accounts for nearly all of the reports. This is the AOL user deciding there’s too much crap in their mailbox, selecting it all, and hitting the “this is spam” button. This isn’t useful.
    Now the broadcast emails? When our clients send out newsletters, we get anywhere from zero to two responses back from AOL’s FBL. Usually zero. Most commonly, the client will actually know the person who hit the button, and when asked, they’re all apologies and say they reported it as spam by accident, and please, please, PLEASE don’t delete them from the mailing list.
    Perhaps our experience is different because we’re not a mailing house and have relatively low volume by comparison, but that’s the experience we see. People mostly are reporting friends in error, or just getting frustrated with there being lots of junk and reporting everything they see in front of them.
    Ultimately, it wastes our staff time, and does little else.

  2. MarkB says

    No, it’s not unique to you. We send roughly 25 million emails per week and we see very similar results. A large portion of our customer care’s time is spent handling angry calls from customers who have been unsubscribed due to clicking the This is Spam button.
    Like you, most of the stuff that is complained about is transactional or user-to-user email. The other stuff gets through with few complaints, with one exception. We also allow the user to get ‘search results’ in their email. For some reason, a small portion of our user population (probably half the complaints) clicks TIS when they just don’t like any of the results.
    There is clearly a problem with understanding the use of this button. I think users attribute far more magic to it than is warranted.

  3. J.D. says

    So, make your systems smarter. Blaming your customers seems like an ill-considered reaction, and yelling at ISPs has never gotten the email marketing industry anywhere.

  4. REC says

    JD, who is the “make your systems smarter” comment directed towards? It seems that it could accurately be directed towards all parties in the equation. ISPs give a TIS button and treat it in a relatively binary fashion, when the reality is it gets pushed for a myriad of reasons that may or may not have to do with the spamminess of the mail. If the button were labeled “Make This Go Away” and it fed into a whole set of “smarter systems” at the ISP and the mailer, it could be result in a win-win-win (for ISP, sender, and consumer). But in many cases today, the TIS is a hammer under which every email “problem” becomes a nail. Yay for smarter systems, but they’ll only help if all players (except, of course the customers who are always right — and often quite dumb) in the chain are being smart about how they’re used.

  5. J.D. says

    My comment was directed towards DanS, but you’re right: ISPs should (and do) look at more data than merely the complaint. They look at relative complaint rates per user, and across time. They pick apart various elements of the message, and apply the complaint to each — or exempt some.
    All you receive is the raw complaint, but that doesn’t mean there isn’t any other processing going on. And if ISPs can apply smarter processing, so can you — because you know way more about how that subscriber has interacted with your offerings than anyone else.

  6. MarkB says

    Actually, that’s wonderful in theory, but it is tricky in practice. When you sign up to these lists you agree to certain things, like not repeating a send to someone who has complained w/out their direct intervention. So you’re violating the terms of the agreement if you opt not to remove them from the list or ask them to confirm or any of those things.
    Second, these complaint numbers are fed to other organizations, and are used to determine how “good” of a sender you are. So if you just let the customers keep hitting that button as a means to make the email go away, then it impacts your reputation as a sender.
    That doesn’t mean there aren’t things we can do – we’ve opted to simply refuse to send more email to customers before, and told them to lump it to avoid their constant complaints. But that’s not a GOOD solution.
    What would be good would be if ISPs would be willing to admit that this is not a perfect system (in real terms) and start looking at how they can better handle this not just in now they use the data internally, but how they provide it externally.

  7. J.D. says

    “When you sign up to these lists you agree to certain things, like not repeating a send to someone who has complained w/out their direct intervention.” Really? I’ve worked on a lot of feedback loop systems, and haven’t seen that. Maybe a few small ones?
    Right now I work for Return Path, directly on Sender Score, and I can tell you that we also know complaints aren’t a perfect stand-alone measure. Why do you keep assuming that nobody who works on anti-spam technology knows this?
    What would you like to see changed in the external information? Try to consider it from the user’s and ISP’s point of view, not merely what would make life easier for marketers.

  8. MarkB says

    “Why do you keep assuming that nobody who works on anti-spam technology knows this?”
    Because most people who works on anti-spam don’t – or at least they act like they don’t.
    BTW, my concern really isn’t about making life easier for marketers. My concern is the end-user. I’m not a marketer, I’m a sysadmin, and I want email to be useful. Sadly, too many anti-spam people are so deep in the war that they’ve utterly forgotten the reason they’re fighting in the first place.
    Secondarily I work for a given company, and that company wants their email to be delivered. More importantly, their customers want their email delivered. I like the idea of the TIS button, but its implementation leaves a lot to be desired. Unfortunately, there seems to be an unwillingness on the part of most anti-spammers to even admit this, let alone engage in discussions on how we make it better.

  9. J.D. says

    Well, you’re not going to see ISP staff saying “it’s okay to ignore our users when they complain.” They work for their users, not for marketers. But that doesn’t mean they aren’t going to ignore their users themselves, sometimes.
    All that aside, this is your chance to talk about how the button might be improved. People who run feedback loops read this blog.
    The suggestion I hear most often is “give the user more choices,” yet I’ve seen studies showing that if a user has to click 2-3 times to report spam or once to hit delete, they’ll just hit delete — and then nobody gets any useful feedback.
    Do you have any new ideas, or just a vague dislike of the current paradigm?

  10. Andrew Barrett says

    The comment Laura highlights is the best refutation to date of Al DiGuido’s rant on why senders should “Kill off the Report Spam button”, published eons ago in Internet time:

  11. Reputation as measured by the ISPs at Word to the Wise says

    […] The ISPs put more weight on what their users tell them than on what the senders tell them. The customer is always right. In my opinion, there is no such thing as ‘overuse’ of the report spam button. The more […]

  12. “Do Not Reply” Address? Don’t Bother. - Inbox Ideas: Email Marketing Tips by AWeber says

    […] … don’t get that the main thing that affects your email deliverability is basically whether you make subscribers mad. Make them mad, they click the “spam” button. Enough of them do it, and you have a delivery problem. […]

  13. Do-Not-Reply emails « Jack Of All Trades says

    […] is basically whether you make subscribers mad. Make them mad, they click the “spam” button. Enough of them do it, and you have a delivery problem.To subscribers, spam isn’t just unsolicited bulk email, it’s any email they don’t want. And […]


Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.