In September, I finished up with the overall design for how the new preview rendering backend would work. In October, I was able to start building it out. The bulk of the hard work is complete! The new data format and file structure is going to make it much easier to add new intro templates to the site and my early tests are very promising for render speed. I've still got a few loose ends to tie up and a lot of testing to do. It's a pretty drastic departure to how things worked before, so once I roll the system out and gain some confidence that it's an improvement, I'll post a more technical write-up of the changes.
October was a good month productivity-wise, but not as great if you look at the metrics. Traffic was up ever-so-slightly (about 3%), but conversion rate and revenue were down for the month. I reduced the number of render processes in September to trade a little bit of rendering speed for some much-needed stability. Average preview render time was slightly better in October (4:37 vs September's 4:40), but that doesn't paint the whole picture. I had a couple of outages in early September that precipitated the throttling, and those couple of rough spots drove up the average render time for the whole month. I got the stability I wanted in October, but at the cost of shaving points off the overall conversion rate.
It's too soon to say whether the SEO improvements I mentioned in my last post will offset the dip in conversion rate, but it feels good to be working on mulitple steps in the funnel. I'm about half way done rewriting all the intro keywords and descriptions. I keep waffling on whether the content improvements are more pressing or the improved render speeds, but ultimately I think it comes down to matching the right task to the right amount of time I have to tinker on it.
When I only have a bit of time in a day to work on IntroCave, it makes sense to keep pushing forward on rewriting the intro keywords and descriptions. A couple of hours is not quite long enough to bite off one of the meatier programming tasks that are still needed on the backend. When I do have large blocks of time free (as was the case last week), it's definitely worth pushing the server tech forward as much as I can. I find myself more and more dividing up my ongoing tasks into these two buckets.
The first bucket is for incremental work that can be done whenever I find an hour or two and want to push things forward. The second bucket is for tasks that require deep work (or FLOW). I've heard the opposite of deep work described as shallow work, but I don't know if I quite agree with that description. I feel like the short tasks and the long tasks both move the business forward, so it's mostly a matter of doing the most with the time I have available.
With a lot of progress being made on some of my shorter-term goals, I think it's time to start talking about the long term vision for IntroCave as well. I've acquired a new domain: https://intromaker.com. IntroCave is not going anywhere, but in the process of rebuilding the render servers for the IntroCave intro video templates, I've got some ideas about how I could make a more dynamic system with a LOT more customization options. Once IntroCave is humming along on the new system, I want to turn my attention to building out the new customization engine for IntroMaker. I'll continue to add custom templates to IntroCave, but the focus of IntroMaker will be more of a design tool. It should be a fun engineering challenge!
Top Intro Videos in October
After briefly losing the top spot to Quick Break Title last month (okay, technically it tied for first place), the Streaks - Logo title as back in first place this month (Quick Break Title dropped all the way to 8th!).