Stat alert!
Looking Back at Stats and Themes for 2020
I've dialed back a bit on posting stats and financials here on the IntroCave blog, but I think it's okay to let my inner stat nerd out once a year. If you want more of this stuff, you should follow along on Twitter!
I went through my commit history and Trello boards for 2020 and pulled out a few themes.
Site Hardening
Starting towards the end of last year, I had a bit of a problem with botnets (mostly coming out of Chinese IP addresses). It's unlikely that this was some sort of automated crawler. The first few times it happened, I just manually blocked the IP address of the attacker. It kept happening over and over again, though (could be a competitor, but my money's on a pissed of script kiddie somewhere). I banned a huge range of Chinese IP addresses (I see maybe one sale from that IP range per year, so not much risk of harming legitimate customers) and that made the problem go away for a little while. I've added 3 or 4 more anti-spam measures since then without full-on switching to annoying CAPTCHAs, but it was something I spent quite a bit of time on in 2020 that I'd never really thought about before.
Site Speed
Besides hardening the site against malicious actors, working on performance bottlenecks is also just good for business. Google claims that site speed is a ranking factor now, so I put a TON of work into page render speed and LCP optimizations (least contentful paint). I set up Memcached and started doing some fragment caching across the site. I added webp image generation to my asset pipeline and started using TinyPNG to compress all my assets.
Content Creation
In the first year and a half of running IntroCave, I didn't add a single new video to the site. Without new content, that made writing blog posts and newsletters a major pain in the ass. Starting in February, I released new content every month.
I got into a pretty good groove this year with releasing 6-10 videos every month. All in all, I added 80 VIDEO TEMPLATES in 2020. A typical month looks something like this:
Those new videos are great fuel for all the other places I want to show activity: newsletter, blog post, tweets. In June, I started posting a monthly intro reel on YouTube as well. It takes a LOT of time to put together all those assets each month, so I've put a bit of work into automating as much of it as I can.
A recent addition was writing a photoshop script to export all those PSDs into the right folders and formats (webp + jpg + mobile size).
The "wall of videos" that goes in the newsletter and blog posts has a consistent format that can be generated by machine. The copy and links that go in my scheduled tweets can be written by a machine (using the copy/metadata that I've already added to the templates as part of getting them on the site).
The cleverest bit of automation, though, is pre-writing a shot script for each of the YouTube videos I post. I did this by hand the first few times and it was super time-consuming. Given a list of videos, I can spit out the YouTube description BEFORE actually editing the video. This gives me timeline markers to hit. Instead of editing and then piecing the time markers together after the fact, I can use the video metadata I already have (titles and durations) to build the description first and then just edit to that pattern.
2020 Actuals vs 2019 Goals
In my 2019 Roundup, I laid out a few goals for 2020. How did I do on those goals?
Goal: Finish the new render servers (and update all the templates to use the new system).
Well, shit. This one was a bit of a miss, but I did make some progress. Sticking with "blame Covid" on this one.
I made good progress on it through the summer months but shut this down entirely when remote school started up in August. My wife was busy shipping Rogue Company, so I was basically full-time daddy daycare + IntroCave + regular job. I've been building the new videos for BOTH render pipelines, which adds a bit of extra work each month but will prevent me from getting too far behind on the new videos. I'm not optimistic that this will happen in 2021 unless school goes back to in-person school pretty early in the year.
Start adding new video intros to the IntroCave.
Good job me!
Build at least one free intro tutorial (either for IntroCave or IntroMaker).
IntroMaker went into deep freeze and I never got around to this for IntroCave. I was starting to assemble some video recording gear before covid hit and everything sold out, which proved pretty fortunate. I've got a pretty good Zoom/streaming setup now, but I didn't take the time in 2020 to actually record any tutorials. I'll talk a bit more about YouTube down below, but this is something I should do more of in 2021.
Continue using HARO to grow backlinks.
I responded to about one pitch a month, which seems low compared to how much space this takes up in my brain. There are probably better ways to do this through Twitter or YouTube. I did get on one podcast and do a couple of email interviews, though.
Continue growing my IntroCave Twitter following.
I bailed on this strategy when Covid hit for entirely silly reasons. Before things went into lockdown in March, I was playing a couple of hours of Apex Legends every night with a few buddies. I built a bot to help me filter for people tweeting about gaming videos on Twitter, and I would use the queue time in between games to browse and like a bunch of those tweets. That was super effective for adding followers, but it wasn't super effective for getting any engaged followers.
When schools shut down and I had to start being "get the kids to class in the morning" dad, I had to stop staying up until 1am playing video games every night. If this had been an effective strategy, I would've found a way to keep doing it... but it was always more of a vanity metric than something that was driving sales or traffic.
Test paid ads at low volumes on Twitter and Reddit.
Never got to this.
BIG STAT ENERGY
IntroCave crossed over 10 MILLION page views in 2020. This was from over a million unique visitors. Not bad for a $25/month Digital Ocean box.
I rendered over 750,000 PREVIEWS in 2020, which amounts to more than 2,000 HOURS of intro videos.
I sent 880,000 EMAILS in 2020. Across all email types, I got an 18% open rate and a 5% click-through rate.
Dollar Bills
Traffic and usage exploded when things went into lockdown. It turns out making YouTube vidoes is something people stuck at home with a lot free time are really in to.
IntroCave made over $50k gross in 2020, but expenses were way up as well. I spent a good chunk of change updating my home office, bought a new related domain, and hired a logo designer to design a logo for Intro Maker. I'm spending a couple hundred dollars a month on the (currently inactive) new render pipeline. It's fantastic that the site is doing well enough that I don't have pinch pennies. If 2021 goes well, I should clear the original debt and finally start taking a profit from the site.
EMAIL THOUGHTS
I've basically given up on mailing all of the IntroCave user base. I spent most of 2019 building out my own newsletter and sequence tech, but I was never able to get great open rates past around 40,000 newsletter sends in a single month. I've got around 200k emails, but I've pruned and decreased that from around 40k/month at the beginning of the year to more like 10-15k/month right now (basically just to previous purchasers at this point). The sequence emails perform a LOT better with new users than the newsletters, anyway, so one of my goals for 2021 will be to expand my welcome sequences and continue sending the newsletter to just a tiny fraction of users.
Top Searches
I have a search field in my header, and it's useful to review what people are looking for every once in a while. The top searches for 2020 were:
- games/gaming (lots of new content here)
- news (added this as a category, lots of new videos)
- fire (TODO: more fire videos)
- PUBG (can't do official videos. TODO: tutorials)
- glitch (several new videos)
- Fortnite (can't do official videos. TODO: tutorials)
- Minecraft (can't do official videos. TODO: tutorials)
- horror (a handful of new videos)
All in all, I give myself a pretty passing grade here. I need to add more fire intro videos to the site, but the other search terms are getting pretty decent representation among the new videos I've been adding. For the game searches, I need to either create or commission some tutorials on how to use non-branded templates to create intros for those games.
YouTube Performance
I started posting a monthly video reel to YouTube in June. I mostly wanted it as something I could embed in the blog posts and newsletter ("watch this one thing" is a lot more convenient than "click on these 8 things to watch"). This wasn't a ton of effort on my part, but it's been surprisingly effective. I wasn't able to get much engagement out of Twitter, but I added 155 YouTube subscribers through the back half of the year (up to 256). My videos are only a minute or two in length, so that doesn't actually amount to much watch time (27 hours for the year).
You need 1,000 subscribers and 4,000 watch hours to start monetizing...
Both of those goals seem achievable.
It seems obvious in retrospect that I'd have better luck on YouTube than Twitter. Just because I hang out on Twitter all day doesn't mean that's where my users are hanging out.
I'm not making big goals for 2021. While the kids are in remote school, I'm pretty happy to just keep the content pipeline running, the customer support queue empty, and the servers running. The site made pretty good money in 2020, and that's enough for right now!