Learnings from having Mailchimp ask me to kindly leave.


You know that little button beside each email? The little one that says “Mark as spam”.

Clicking that button feels like one little victory against spammers. I love clicking that button.

And so did people on my mailing list. It grew to 3,000 subscribers before it reached a tipping point and had to be abandoned.

The rules to live or die by

Mailchimp monitor “marked as spam” rates for mailing lists.

If more than 0.1% of subscribers mark an email as spam, this rings alarm bells on Mailchimp. Three warnings, and you’re out.

Did you get one spam report from a 1,000-receiver email? You’re already smack bang on the limit. 2 people for an 800-receiver email, now shit is hitting the fan!

My story

Rob Walling makes the highly valuable point that people generally don’t spend upon their first visit to your site. I have top respect for Rob’s approach to online marketing.

My aim was to build up traffic for Bitesize Irish Gaelic, an online collection of Irish language lessons. Subscribers to my mailing list got an ebook on how to learn Irish online.

I had already collected emails on a landing page before Bitesize officially launched in 2010. I emailed the launch announcement to the list of 179 subscribers. 2 abuse complaints later, and I’m at a 1.1% abuse rate, well above the 0.1% allowed.


After some more emails, I emailed out an announcement about a small price increase. 459 recipients, and one marked as spam. Ugh, I’m still treading above Mailchimp’s rate.

Fast forward a bit, and I send an invite to a segment of the list to register for a free live webinar to learn some Irish. The webinar was fun to do, but of 1,479 recipients, 6 marked it as spam. 0.4% abuse rate.


Mailchimp asked me to delete the list.

Things to learn from the failure

There’s a lot of factors at play here that I haven’t been able to get in to. For example, subscribers were receiving an autoresponder series while also getting once-off mails. Some people may have received an autoresponder and a mailing in one single day.

The mailing list needs to be split into two: prospects and customers. Thankfully I had already implemented this early on.

The autoresponder series may not have been interesting enough, although in my opinion it was highly-targetted. It may have been too often, or not often enough.

I didn’t collect first names on the subscription page. Perhaps this may have helped to make the emails more personal.

Despite the double opt-in nature of mailing list subscription, it may not have been clear enough that ebook == mailing list.

Last but not least, I certainly can’t skip the fact that the emails themselves may just have been a bad read. I tried to make them personal, and in my opinion were targetted to a very specific niche. Yet people may not have been impressed.

Be warned here, and then go for it!

Despite resulting in a dead list, this has been a great learning experience in the world of Internet marketing.

A major downfall to email marketing is that testing is difficult! Stray on the wrong side of abuse rates, and you have a real chance of causing actual trouble. In other words, it’s hard to learn.

Email marketing is something that needs to be mastered. But it’s too valuable to ignore.