I've wanted to launch one of my side projects for a very long time. I'm the kind of guy who loves starting things, but never finishes them. Well, this week I finally finished one of them, by launching The Joy of Svelte, my first online video course!
Finding inspiration in an old five-year plan
Back in December, 2019 (a year ago, but feels like a decade), I got a new notebook for Christmas, because I had filled up my old one that I use for meeting notes, To Do lists, and stuff like that. I started re-reading my old one, and saw that near the start, I had a five-year plan from December, 2015. I had a goal to expand beyond my freelancing business and launch my first video course in 2016, with the goal of continuing to create courses, apps and other products over the coming years. By the year 2020, I wanted to have a whole catalog of courses and products under my belt.
Well, when I reading that, four years had passed, and I still hadn't launched anything. Not a single app or course, other than the dozens of free videos I'd recorded for YouTube, and, of course, all the client work I'd done as a freelance web developer. But I still wanted to do something for myself, something of my own creation.
The disappointment and shame I felt while reading this was the push I needed to finally commit to this lifelong plan and stick with one of my side projects long enough to actually see it through to launch. So, in January, 2020, I decided that I would focus on the things I was most excited about: teaching, recording videos, and Svelte.
Trying to stay motivated by committing publically
I had it all figured out. I put up a landing page that said "Coming in the Spring of 2020", and publically announced that I was working on a new course. I committed myself to my newsletter subscribers and Twitter followers, I put out an announcement on YouTube, and then hoped that all the public accountability would force me to follow through and finally launch something.
Well, that was all back in February. Spring came and went, and I was still stuck planning and trying to decide on the course contents. In June, I had to update the landing back and change it to "Coming Summer 2020..."
I decided early on to self-publish The Joy of Svelte by developing my own video course platform. I'm a web developer, after all, and it's way too easy to feel like I'm being productive when I'm writing code. So in a way, it was a form of procrastination, because I could sit down to integrate Stripe, or create a video player interface, or write code to deal with emailing out access links, and feel like I was making progress. In reality, I could've just used one of the many video course platforms available and saved myself a lot of time and effort.
Having now built all that out, I'm happy I did, because now I can self-publish more courses in the future. But I realise now that I could have launched a lot sooner if I had focused on recording videos and spent less time writing code.
Pivoting to focus on learning objectives
I started to record some videos, with the idea of making an SVG drawing app using Svelte. I recorded three videos showing how to do this, until I got to a point where it was starting to be more about SVG particulars and less about Svelte.
Eventually, I came across some very useful advice about creating course content: identify what it is specifically you want people to learn, then go and teach those things. I know that seems super obvious, but somehow I'd lost track of that and was instead accidently trying to make a course teaching people to make an SVG drawing app, but I don't think many web developers have the need to make SVG drawing web apps.
I looked at the landing page I'd originally made, and saw that I'd already outlined some key topics that I was planning to include:
You will learn about:
- getting started with Svelte
- using templating syntax to render data
- data fetching strategies for components
- using components to simplify complex web user interfaces
- building custom stores for state management
- integrating Svelte into an existing web application
- ...and more!
I decided to make six new videos, each one focused on one of these learning objectives. It was a simple, straightforward approach that ended up working very well, because it kept me focused on what it is I wanted people to learn, and less on what cool thing I wanted to build as a code example.
Back to the drawing board
So I abandoned the SVG drawing app videos, and started from scratch. I looked for some simple free web APIs and found one for Nobel Prizes, and decided that I'd use that to show people how to fetch data from an API. It needed very little explanation, didn't introduce any new, unrelated concepts, and more closely resembled the kind of API that I'd often used to build web interfaces for my clients. It might not be super fancy or flashy, but it allowed me to focus on Svelte instead, which is what mattered.
Off screen, I sat down and built a UI for browsing, search and filtering Nobel Prizes, to see if that would work well for the videos, and it turned out to be perfect. It gave me lots of different opportunities to demonstrate various Svelte features, and plenty of ways to show off what makes Svelte a joy to work with, and all the different strategies for making clean, reusable web components using Svelte. None of it felt contrived, all of it was applicable to real world web applications. I was ready to start recording.
Early access pre-launch and a final push to finish
Summer 2020 was coming to an end, and I did not want to change the release date on the landing page again. So, in one day, I sat down and recorded three of the six videos. I uploaded them to YouTube as unlisted videos, and on the very last day of summer, I sent out an email to my newsletter subscribers announcing that Early Access was now available.
It felt so good when I had my first sale ever! And then another one came! And then, while I was sleeping, another one! People were actually willing to pay me for my videos! This was a huge milestone in my life and career, and really validated all the work I'd put into it.
Still, I had three more videos to record to finish it up.
Benefiting from my own misfortune
Then something horrible happened. I recorded two more videos in one day, but when I finished, it turned out that OBS had used the wrong microphone, and so the audio was total garbage. I had to painfully decide to throw those videos out and re-record them.
Actually, that turned out to be beneficial, because I wasn't totally happy with some of my examples, and ended up coming with better examples that demonstrated the strengths and weaknesses of the different types of Svelte stores before I re-recorded the videos.
Launch day, and being too early
Three months after my Early Access launch, I had finally finished all the videos, and was ready to put the finishing touches on my web site, so that people could get a private link to watch videos directly on joyofsvelte.com instead of on YouTube. Using unlisted YouTube videos had felt a bit unprofessional, although I don't think anybody would have complained if I had stuck with using them.
Finally, on Monday, December 14th, 2020, I launched my first video course ever! I created a promotional video, and posted it with an emoji-filled tweet to Twitter.
On launch day I had two sales, and woke up the next morning to a third sale. I had tempered my expectations so that I wouldn't be disappointed, and so I was actually pleasantly surprised to make any sales that day. I'd figured most people who were excited about the course would have bought it during Early Access, and that turned out to be mostly true.
I also have come to realise that I'm probably way too early to be launching a course about Svelte. I chose Svelte because I'm so excited about it, and am happy to talk about Svelte endlessly, but the fact is, Svelte is not yet widely adopted amongst web developers, so there really isn't a huge audience there yet. It's still somewhat a niche topic. And that's okay, but it means that there was no way I was going to have a ton of sales on the first day.
There just aren't that many people learning Svelte right now. I think this will change over the coming months and years, and I'm glad to have put this course out into the wild to help people looking to learn Svelte. I hope it helps people to see what it is about Svelte that I find exciting, and why it has changed the way I approach web development altogether.
Lessons learned for the next course
This won't be my last course, it's just the beginning. Here are some lessons I've learned from building this course, that will change the way I approach building my next video courses.
I'll focus on learning objectives from the start. I'll make a short list of what I think people will want to learn about, and make videos focused on those points. The code examples I use will be chosen for how well they can demonstrate those key learning objectives.
I'll avoid perfectionism, and limit how much time I spend planning the course up front. Planning is a trap that I fell into, because you can keep planning the same thing forever. At some point you have to say "good enough" and start doing the actual work. Chances are, when you actually start recording the videos, you'll find out the best way to do things.
I ended up re-recording a lot of The Joy of Svelte by accident, and that benefitted me by allowing me to improve the content before recording the final videos. I will do this on purpose next time, maybe live streaming the content on Twitch, or possibly running a workshop beforehand, so hopefully I can get some useful feedback first as well. (And I'll try to remember to double check my microphone before recording the final videos!)
On to the next side project
One of the best things about launching The Joy of Svelte, is that I can now start working on all the other side projects and ideas I came up with this year, but wouldn't allow myself to work on until the course launched. If you're interested in following along, you can sign up for my newsletter.
And, of course, if you're interested in learning Svelte, check out The Joy of Svelte!
Published on December 16th, 2020. © Jesse Skinner