Newsletter Crew August 4, 2019

Newsletter Thoughts

Will Hankinson @SimianLogic

I'm still working on getting my newsletter house in order. I sent out 4 variants this month but got throttled a bit at 1500 per bucket (6k emails in a short span). Instead of thinking "1500 per variant," I'll tweak that to "I want to send X per batch, so divide that among however many variants there are."

I'm currently doing a lot of list creation and management manually within Mailgun, which is time consuming. Moving thousands of users from one list to another takes a lot of time (there's no batch delete API, so you have to make an API call for every user). I'm leaning towards just blowing away all my lists each month and recreating them from the master list, which would be way faster and can be automated a lot more.

Someone asked me why I'm even bothering versus using an off the shelf service, and I think my answer is a bit different than most folks. The site is profitable and I can afford to pay the ~$400/month it would take to use Mailchimp or Drip or whatever. But if sites are worth around 36x net revenue (broadly ballparking), I think of that as taking $14k of value off the business. Can I build my own for less than $14k? If so, I've "preserved" $14k of value that would otherwise be wiped out. I can certainly build the features I need for less than that. If you're a million dollar company, who cares about $14k. If you're a side project worth ~$100k, that's a huge chunk of the business.

If email were an important source of revenue, the math would probably be different. If you're getting good money back, just spend and move on. This is exploratory for IntroCave -- I'm not sure that this will lead to any increase in revenue even though I have a huge list. I'm brute forcing things now, which means a single newsletter takes about a day to send out... but over time I'm figuring out little tricks to speed this up. If I keep chipping away at the problem I expect to be able to have exactly what I want for a fraction of the price... and I'll be able to take that with me in any future businesses.

(And maybe write all this up so others can do the same.)

  1. 1

    How many newsletters do you create a month? If you send one a week and it takes you a day to do each one, then that ~$400/mo you're saving is costing you 4.3 days' time, which is <$100 day. Is your time worth more than that?

    1. 2

      Just one a month. It's slow going because I'm doing a LOT by hand right now. My list has around 45k emails and you can't send that many at once. I've been subdividing that by hand into smaller lists and sending those one at a time (I sent 18 yesterday, but 4 of those were an A/B test). Once I get a good handle on how many I can send per hour without getting throttled, it's just of matter of automating all those things I'm doing by hand right now.