A reflection on The Palindrome's journey upon passing 15k subscribers
work it harder, make it better//do it faster, makes us stronger
I'm not a fan of milestones and celebrations. The Palindrome recently passed 15.000 subscribers, and we were almost halfway to 16.000 before I noticed it.
There is an upside to this reclusive nature, though; it's why I can spend three hours perfecting a single illustration. Crafting words and images that unravel intricate concepts is what I love. I've been doing it since around 2020 when writing online became my passion and eventually my livelihood.
Somewhere along the way, I've lost something.
In the beginning, things were simple. No pressure to build a brand. To grow. To make money. It was just you, me, and a ton of geeking out on topics like how the concept of expected value arises from simple games of chance, or what the seemingly magical formula of entropy has to do with guessing a number.
Words and formulas were flowing out from me, and you enjoyed every bit of it.
Fast-forward to the present day, and the vibe is different. This is my main job now. I craft CTAs, convert free subscribers to paid ones, grow audiences, build marketing funnels, sell my book, optimize my workload, and do a whole bunch of stuff that makes me money but doesn't bring me joy.
Reflecting on the state of The Palindrome, I realized that I sacrificed quality. Rushed some of my illustrations, overdid the CTAs, and sometimes released with bland writing. I focused too much on making a living.
It's time to go back to the beginnings. From now on, I'll focus on only quality and quantity.
In the future,
I'll remove almost all CTA-s, email headers/footers, and other conversion-increasing tools. There will be, at most, one per post, which cannot interrupt the flow. You’ll upgrade to paid when you feel like I provide enough value to justify the change; I won’t bother you with this anymore.
I'll make significantly more free posts and stop sending out previews of paid posts. Previews are great for getting more paid subscribers, but now I feel that they are annoying for the recipients. Even though I attempted to place the paywall break to get a free mini-post, it still feels cheesy, so I won't send them anymore.
I'll properly render all math formulas into images instead of using Substack's built-in LaTeX blocks. Although Substack's solution is much faster, it's also unbelievably ugly. Played around with it, used in a few posts, but in its current form, it's far below the standards I want to reach.
I'll put Substack first. Let me explain. For instance, I used to turn some of my Twitter threads into newsletter posts. The quality was there (I hope), but the flow of a thread is significantly different; the 280 characters per tweet significantly impacts the writing style. When I write a post in Substack's editor, I always feel the result is more colorful and entertaining. Thus, I'll do what's best in terms of quality and provide you with a better experience.
I'll also significantly upgrade the free and paid benefits in line with these improvements.
So far, the biggest challenge for me is balancing the free/paid posts: keeping most of the content free while also providing a worthy premium offering. I tried many different approaches, and I wasn't happy with either of them. I spent quite some time revamping the current system. This is what I came up with.
From now on, free subscribers will receive almost all new posts (~75%) via email and the Substack app. This is a threefold increase from the previous 25% free - 75% paid ratio.
Thus, free subscribers will receive
about three high-quality posts per month,
and access to the subscriber chat.
In turn, paid subscribers will receive
access to the archives (that is, posts older than a month),
about one paid-only post per month,
the new chapters of the Mathematics of Machine Learning book,
and monthly behind-the-scenes posts. No engagement bait BS here; I aim for depth. For example, one of the firsts will be about my inspirations, like Andrej Karpathy or the famous death metal band Cannibal Corpse. (No joke here: Cannibal Corpse is a stellar example of extremely high quality yet prolific output throughout decades. I'll tell you all about it.)
I am confident that this'll be a good system: it's more dedicated to the democratization of learning (which is one of my core beliefs) and also brings special value to the premium subs.
Let me know in the comments what you think! I always appreciate any feedback from you.
After all, your support is what keeps me going.
Congratulations Tivadar Danka for going back to quality.
In a society where there is too much content (regardless if it is free or paid) we are faced with the sad reality that most of the content is click bait meant to grow subscribers and/or sell something.
I believe writers like who realise the quality is more important that number of subscribers, are the only ones that will end up having loyal readers that will support you and keep you up and running for years or decades to come.
I myself will always support a writer that makes me think, teaches me something and does not believe that I am stupid enough not to understand how online writing works and the fact that at the end of the day the writer needs to pay his own bills.
Again I congratulate you for realising that quality is all that you need to keep subscribers engaged and loyal.
I can really appreciate your learning, honesty and sincerity here. For a reader, her originality in your highest quality is what matters to us. If you're not feeling the flow joy and immersion, we might notice it in your work.
Congrats for reaching a very decent amount of readers.