Work in progress: a newsletter move

Working on it!

Attention: As of January 2024, We have moved to counting-stuff.com. Subscribe there, not here on Substack, if you want to receive weekly posts.

This ~10 year old picture from China is still my best “moving things” picture

TL;DR — Counting Stuff will be moving to a new host in the near future after I finish solo-organizing a conference in January. There’ll be a heads up and probably some minor chaos/learning pains along the way but I expect it to be largely seamless for everyone.

Substack’s got, among a bunch of other similarly nasty things, a Nazi problem that their CEO actively refuses deal with. It’s also been increasingly naggy about… everything as a product since I started this newsletter 3 years ago. My weak-ass excuse for not moving earlier is just because between work, wrangling a 4 year old kid, and writing 1.5 long form posts a week without fail — inertia is really exhausting to fight when you’ve got unending deadlines. But it’s high time I got my butt off of the platform as soon as I’m able. So now I’m actively working on this problem and since I’m burning brain cycles over this, I’m gonna take you all along for the ride.

As for why I can’t act even sooner, it’s because I’m solo-organizing my tiny online gamedev conference in January again for the 8th time, and that’s eating up a huge amount of energy that I could’ve otherwise used to evaluate new hosts and/or infrastructure. In the nooks and crannies of downtime I have I’ll be toying with infrastructure and the mechanics trying to figure this move out.

Come on, this can’t be so hard

On the surface, the problem isn’t THAT difficult. Grab everyone’s email here (which Substack does allow exporting and migrating the payments) and import it into a new service, continue writing.

But see, the experience on Substack has been, from the writer’s perspective, quite nice. I have a browser logged in, I write from any browser that’s logged in including my phone and it’s continually synced. Then I tack on pictures and schedule the post. That’s it. I’m somewhat aware things have steadily gotten worse on the reader side but having a simple way to “get work done” is pretty important when I only have a couple of hours a week.

I obviously don’t have to maintain servers. Perhaps most importantly, I don’t have to worry about maintaining my email sending reputation to avoid getting spam boxed. I also get to freeload off the VC money subsidizing the costs of blasting out 1.5 emails a week to over 3200 people — do the math that’s almost 20k emails a month, plus incidental web traffic, CDN, and server costs.

Things get much more complicated once I started looking into going somewhere else or taking up that workload. Up front, Substack lets me export the email list so I can theoretically start blasting your mailboxes from another service easily. But what’s the plan for migrating old posts? How will I deal with URLs and links? How do I make sure the people who paid money to support this newsletter get their status and subscriptions transferred correctly? And how do I not get banished into the spam box?

Sending lots of email costs money =O

The most prominent direct competitors in this space that I hear about seem to be Ghost and Buttondown. The former is an open source CMS that added newsletter sending features, the latter is a dedicated newsletter platform owned by one guy. I’ve heard positive things about both and have seen people migrate to them. So naturally I looked at them first.

The way both of these services work is they bill based on how subscribers you have, which at the time of this writing floats slightly above 3200. So Buttondown would cost ~$30/mo, Ghost’s hosted service would be $65/mo. Neither value is gonna completely break the bank but it will be a notable rise in cost (which probably averages out to $18/mo for me in fees right now, the power of aforementioned subsidies). I’m lucky in that some readers are willing to pay to support this newsletter, so either way it can fund itself. There’s a tiny worry that since I don’t push for subscriptions much, cost of readers might one day outstrip those willing to pay but that’s a bridge to cross if we ever get to it.

Putting aside raw costs for now. More important is whether I can trust these services to do the very important jobs of:

1. Letting me just sit down and write efficiently in peace

2. Get the finished work into your mailbox reliably

I’ve witnessed Buttondown have some occasional issues with #2 (and I don’t blame them, email reputation management is VERY HARD, constant work, random stuff is bound to happen). They also support markdown, which some people have strong feelings about but I’m ambivalent.

Meanwhile Ghost is a pretty slick, fully featured CMS that lets me put most of the blame for the ugliness of my newsletter squarely on my own shoulders. I’m not sure how their spam performance is yet so it’s worth testing. There’s a bit more unknowns here but appear to have most of the usability features I want.

I’m also vaguely aware there are other competitors in this crowded space, and if readers happen to have a recommendation I’m all ears. I think WordPress has a hosted newsletter function, then there’s beehiiv, and more…

And there are similar alternative services…

Obviously, there are many competitors in the even broader “email marketing” space where newsletters is a tiny subset. These places think of emailing people in the true “marketing” sense. These are the “e-commerce store you bought socks from once sending you promotional content every week for eternity” providers. There’s advanced segmenting, tracking, upselling/cross-selling/whatever-selling functionality that I’d never use but then have to pay for. There’s automation, triggering on events, integration with other service providers. I’m sure all that stuff is useful to many professionals and businesses out there, but they’re not made for my needs and not priced for my budget.

As an example, Mailchimp is a big name in this space and would cost close to $75-100/mo to support this a newsletter of this size while having a writing experience that’s much more complex and annoying because there’s so much customizability available. There’s even services that layer ON TOP of Mailchimp that simplify the experience to what I’d want, but they cost extra.

Or just give up and go solo…

There’s also the option of self-hosting a platform myself. If I’m tired of having to re-evaluate and move platforms ‘coz the current (or future) boss likes to hang out with Nazis and grifters, I should just make myself the boss and avoid the mess. I already run tiny servers here and there, adding another to maintain isn’t out of the question (albeit inefficient).

The aforementioned Ghost offers this as an option. Looking into self-hosting an instance of Ghost somewhere, it seems that the bare minimum cost is somewhere in the range of $50/mo to get a virtual server to run everything and a mail sending service like Mailgun. The nice part is that I can comfortably just keep going with it until the server overloads from software bloat.

I have a test instance running on my desktop right now that I’ve half set up. It’s decently nice and I need to spend some time trying to use it for its actual purpose.

There’s also old granddaddy WordPress to consider. I’m not interested in going back to straight-up blogging since RSS use is mostly nonexistent these days, but there is most definitely a bunch of plugin options that might be worth looking at to see if the features available cover what I want. If only their feature lists were clear about whether they have the features I need. The costs seem similar to other self-hosted methods so it’s a matter of user experience.

And if I were extra brave/bored/dissatisfied, there are plenty of other newsletter-sending tools out there. After all, plenty of software engineers have thought” how hard can it be to have a system that blasts emails to thousands of people while collecting money”? The features and usability vary massively, the community support and future existence of said tools are also a mystery.

So yeah, there are too many options, all with complex implications. Evaluating and eliminating even just one or two of these options is a pain in the butt because it means spinning up test machines and getting software to install just right. None of this is hard work, but it’s the sort of work where you start one evening and suddenly 8 hours have passed.

Personally, I’m leaning towards self-hosting things so that my next platform migration will on my own terms and timeline instead of due to techbros doing techbro things. Let’s just work these trust issues out in solitude while dusting off those rarely-used data engineering skills. Plus, what do you expect out of someone who’s running a conference because no one else would do it right.

So yeah, I’m working on this. It’ll take me a while because there’s everyone promises a bunch of features using conflicting terminology and there’s no way to compare the experiences and feature completely other than trying them out. All done while I’m making regular posts and and organizing a conference.

In theory the transition should be fairly seamless from a reader’s perspective, even the subscribers paying to help support this newsletter. I shall work out the kinks as I come across them. But one day quite soon, hopefully before February rolls around, you’ll find me with a shiny new domain and email design — hopefully in your inbox and not your spam box. Thanks for your continued support in the future!

---

Standing offer: If you created something and would like me to review or share it w/ the data community — just email me by replying to the newsletter emails.

Guest posts: If you’re interested in writing something a data-related post to either show off work, share an experience, or need help coming up with a topic, please contact me. You don’t need any special credentials or credibility to do so.

---

About this newsletter

I’m Randy Au, Quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. Counting Stuff is a weekly newsletter about the less-than-sexy aspects of data science, UX research and tech. With some excursions into other fun topics.

All photos/drawings used are taken/created by Randy unless otherwise credited.

randyau.com — Curated archive of evergreen posts.
Approaching Significance Discord —where data folk hang out and can talk a bit about data, and a bit about everything else. Randy moderates the discord. We keep a chill vibe.

Support the newsletter:

This newsletter is free and will continue to stay that way every Tuesday, share it with your friends without guilt! But if you like the content and want to send some love, here’s some options:

• Share posts with other people

• Consider a paid Substack subscription or a small one-time Ko-fi donation

• Get merch! If shirts and stickers are more your style — There’s a survivorship bias shirt!