About once a year, the blog fairy descends from the sky and grants you permission to bore your audience with data about your audience. With a graceful wave of their wand, an author is finally freed to do what he has wanted to the whole time: to explain, in exhausting detail, why they think they aren’t as big as they could be right now.

Then, in a touching end-of-film scene, the author learns the fairy wasn’t needed - the whining was inside of them, the entire time. They just needed to believe in themselves.

Without further ado and with no more adornment, I give you the 2022 statspost. May God have mercy on your souls.

Free Sub GrowthConventional blog theory on how to grow a blog goes something like this:

1. Write a bunch of really good articles, or at least the best articles you can.
2. Try to write stuff that a bunch of people want to read, which is sort of distinct from 1. if you think about it long enough.
3. Balance 1-2 in a way that makes sure you won’t lose respectability. Respectability is important! People don’t pay people they don’t respect, so don’t pander unless you can pander enough to make that your entire business model.
4. Once you’ve completed 1-3, try to get your stuff in front of tens of thousands of viewers. If you’ve done your job really really well, you will have a low-single-digit conversion to subscribers. Now you have some options:
5. Option 1: Withhold some amount of your output from people who don’t pay.
6. Option 2: Give people an option to pay, but don’t withhold any work and hope they pay for no particular reason - just from goodwill or something.
7. Option 3: Don’t give people an option to pay, but find some wealthy benefactor who pays for your work and who (hopefully) will not commit massive fraud and go to prison. Do not reflect on how often “will give tens of thousands of dollars to an OK writer on a whim” and “will commit massive fraud” overlap on Venn Diagrams.
8. Live the enviable life of an independent artist as portrayed in a Hallmark movie - buy a house in the woods, meet a person from a very different walk of life who teaches you the real meaning of creative output, etc.

This all implies, in a weird way, that free subscribers are the most important subscribers - they are upstream of everything else you want. Within their teeming masses hide paid subs, just waiting to make themselves known with sweet, sweet financial gains.

Sickening financial incentives aside, they are more important in a different way: If your work matters, they are the people it helps. The more people who read your stuff the more good you do, and free subs represent the vast bulk of words-read.

Thus I’m glad to say that this year has been good for subs:

At the end of the first year of writing, I had just under 1000 subs. Now I have 3296, which is, you know, nice. A math friend assures me this is technically exponential growth, which is the best kind of exponential growth. In addition to that, a certain amount of you dial in via RSS, which I can’t really track. Feedly says it’s at least 171 of you:

But there’s almost certainly more than that, given that Feedly isn’t the entire RSS market. So there’s about 3500 of you, or more! That’s good! I’m glad you are here and glad you put up with me.

Free Subscription SourcesSubstack is really, really bad at telling me where subs come from. I’m not sure this is their fault, since everyone seems bad at this - I have Google Analytics tracking, and it’s not better. I have other tools, and they aren’t better. I sometimes randomly email people who just signed up to ask how they found me, and they are mostly (and justifiably) creeped out by this and don’t answer. It’s just hard to know where people are finding your writing, which I’ve come to accept.

That said, I do have some probably-broadly-correct guesses. If you take a look at my overall subscriber growth in the graph above, you will note that things start picking up in April, and then really kick into gear around September. The first is when Substack introduced its recommendation feature, which lets me recommend various blogs to you (and vice versa). This was the first time Substack did anything substantial to leverage the platform into small-creator reader growth, and it worked well.

September-ish is when ACX started recommending me. This was a very nice thing to do for a couple of reasons: the first is that Scott is really big compared to most Substack folks, which means he can send a lot of traffic, but also means that a certain amount of people categorize him as “more than a man” in a way that makes them take his recommendations more seriously. The second is that Scott is really trusted. This lends even more legitimacy to the recommendation, which makes even more people come over. It was, as they say, a good get.

Substack claims that I have about 1377 subs from that recommend (it’s probably higher since the errors here seem to trend towards omission), and of those, 888 are from Scott. Other big contributors include Karlstack, David’s Overthinking Everything, Samstack, Tom’s Curiosity Shop, and Parrhesia’s Newsletter.

Other big sources exist, but are much, much harder to track. A lot of you found me from me having various arguments with people - mostly I don’t know that’s how I got you. Some of you found me from placing in Scott’s book review contest - I mostly don’t know that’s how you came here. Ditto Hackernews folks. Ditto everywhere else.

If you can, I’d love to get some comments on this, especially if you came to the blog some way I haven’t mentioned. I want to cater to you to the extent I can without subverting my morals! I really do. But It’s hard to do that without knowing who you are. I’m especially interested in folks who are here for “I’m a member of group X, and you are an attempt to be outside my bubble” and “I’m a religious person who reads a lot, and you are one of those things I read” reasons.

Paid SubsTalking about paid subs is icky - it’s important to me, not at all important to most of you, makes me seem money-motivated (which is especially normed against in tech/startup/programmer communities, which a lot of you live in), and tends to be boring. But I’m doing it anyway, because it’s important-ish to get the whole picture.

The way paid subs work on Substack is that they come in two forms: lump sum and monthly. Lump sum people are supposed to be paying for the whole year and getting a discount, but my prices are already as low as they can be, so they are more often people who are trying to give me a one-time gift of X dollars.1 For the most part, a year’s subscription obtained either way costs you about $60 and I get about $50 of that after Substack and Stripe take their (substantial) cuts.

As mentioned above, I have three main ways I can work within that model are to rope off some content for paid subscribers, to leave everything open (which is better for free sub growth and for maximizing total readers) and hope people pay anyway, or to not allow people to pay at all and hope some rich person funds me for some unknown reasons (or just be OK with not having a paid option). I have always done the “let people pay if they want, like a tip jar” model, which results in numbers like this:

That’s about a 2% conversion rate, which is fine (It’s always been surprising to me that people choose to pay at all) but also pretty low compared to some - I’ve seen 10% conversion rates out in the wild. I think several things are driving this:

1. I don’t do a great job of giving people reasons to pay. I’m pretty open about my financial straits being much better lately. If you look at last November compared to today, I’ve had 50% revenue growth during a time my readership tripled. But last November most people knew me from an article that was explicitly about me being poor - I think that made a lot of people feel bad for me and give (which, by the way, thank you for being nice in that way).
2. Substack is now pretty old, as such things go. A lot of people were excited about the “Here is a platform that works against free speech restrictions from group X” aspect of Substack at first and subscribed to a lot of stuff - you see less of that now. A lot of people are already paying for everything they want to pay for.
3. I’m not being particularly consistent in both the quality and quantity of my work. If I was better at getting articles up weekly, this number would probably be higher. If I was full-time and really managing the community/making more kinds of posts, even better.

So this is high enough in some senses - I’m happy about it - but could be higher if I did some of the stuff I’m supposed to do. Note that this isn’t a call for more of you to pay; I’m glad you are here, and it’s free on purpose. But hopefully, I’ll get a few things under control and be able to write a little more this year, which would make it make more sense for more people in a position to pay to choose to do so instead of understandably not relying on me as a consistent content production source.

TrafficTraffic is distinct but connected to subs in a weird, highly visible way. Last year, my average “real” article (excluding weird stuff and stuff I didn’t mail out, but posted anyway) had 6469.05 views, a number that’s at least somewhat inflated by Substack double-counting viewers a bit. The number is further inflated by articles that did well elsewhere and brought in a ton of hits; last year’s top performer had 42922 hits, which juices the stats a bit.

If you take the five best-performing articles out of the running, that number drops to 3589. That’s y’all - subscribers who regularly tune in, plus direct shares to friends. And in 2022, those numbers grew somewhat; my last ten articles (omitting my very most recent, which is still gaining numbers) averaged 4812.6.

That view count didn’t keep up with subscriber growth, but I consider that natural - I’m losing and gaining subscribers all the time, and it makes sense that people who have subscribed to me longer think of me more as a core must-read and open emails as a result.

Traffic is impossible to track well. It just is. Sorry. If you want the broad strokes, it’s probably a lot like my subscriber sources summary above - most people find me one of a few ways.

Progress and GoalsOverall it’s been a good growth year. I haven’t shrunk, which is always a fear, and I’ve become much bigger judging by some-but-not-all metrics, which is good.

Earlier I mentioned that a friend had called the growth exponential, which is true-ish but also a pattern I don’t expect to continue unless something changes. Most of my growth this year has been based on recommendations; that’s great, but at best I’ll maintain linear growth if so.

My current goals, such as they are, are these:

1. Write more consistently. This is always a goal, and it’s made hard by having to fit a life around it. I had 37 “real” articles last year, meaning you missed out on 15 articles I should have written in addition to some extra content you got in lieu of that.
2. Find new ways to share my work. Recommendations have been big, and Hackernews has paid out a few times, but I continue to be bad at Twitter/Reddit and don’t have any other great ways to grow my audience outside of those three things.
3. Be better at engagement. I’m good at responding to comments, but terrible at making opportunities for people to chat/make friends/build community.

If you have good goals for me not listed here - let me know. I’m a good writer but I’m also fundamentally dumb in a lot of ways; advice is always appreciated.

I think my long-term goal is still to grow enough that I can do this full-time; I’m nowhere near that right now. To replace my current income, I’d have to do something like 25x my audience (or else increase paid conversion rate while still experiencing high-but-lesser growth) so I’m not expecting that to happen very soon.

You are helping just by being here and reading - that’s already more than I deserve a lot of weeks, and I’m thankful for it. The blog will remain free-with-tipjar for the foreseeable future, so don’t worry about that; I don’t expect the average person to pay, and I suspect there’s a lot of you for which paying would undercut some more important responsibility (family, friends, basic needs, etc.) and who I’d actively discourage from doing so if I knew.

If you feel inspired to help me along, I always appreciate shares to any audience that might like my stuff, especially if it’s an audience you suspect is outside of my normal reach. Otherwise, I’m glad you are here; I hope you continue to enjoy the blog, and happy new year!

---

Share

Subscribe now

1Sometimes X dollars is some significantly high multiplier of X, which reinforces my guesses about “one time gift in support” payers.