You never know when an old fishing boat will inspire you to change your mind about something.

Products as Fishing Boats

I started off thinking I'd write about rising tides. I'm still super bullish on YouTube, and could probably spin a few hundred words about how hanging around and making the product better will allow me to ride that rising tide. The more I kept staring at that old fishing boat, the more a different metaphor started to form. When it comes to old fishing boats, success is inversely proportional with smelling nice. That boat looks like it stinks, which is to say it looks like they've caught a lot of fish. Is it possible to catch a lot of fish without smelling like it?

When I make changes to the site, I'm generally trying to make changes that are good for users. This is a perhaps-too-optimistic belief, but I believe a better/faster/more useful (choose your scoring mechanism) website should also be more successful as a business. I'm a bit of a podcast junkie, and one thing I hear over and over that resonates with me is the idea of being useful on the internet. I feel like most of my development efforts so far have gone into fixing leaky buckets and building missing features, but as the site stabilizes I want to put some thought into how an intro maker can be more useful to more people. How do I stay out of the fishing business and stick to something more win-win? Whale watching?

Longer Storage for Paid Videos

Paid intro videos are now stored on my delivery servers for 14 days instead of only 3.

I actually rolled this code out about a month ago, but I wanted to keep an eye on it and make sure it worked fine before updating all the copy sprinkled around the site. While most people hit the site looking for a gaming intro maker, IntroCave's intro maker is surprisingly (to me, anyway) popular with small gyms. The original "I only keep your video for 3 days" policy was entirely cost-driven. Because the preview renders and finished renders used the same bucket on S3 (Amazon's hosting), both types of videos got the same 3-day expiration window. This makes perfect sense for the preview videos! If you haven't purchased an intro within 3 days of creating an intro video, the stats say you're likely not going to purchase—last time I pulled stats, 95% of purchases were happening within 2 hours of getting a preview back from the render server.

Slight tangent: one of the reasons 95% of purchases happen within an hour or two of making a preview is that it's super hard to go back to an existing preview. I keep them around for three days, but there are no links to them in your dashboard. I'm trying out some new email sequences when logged-in users create new previews, and hopefully they're more on the useful side than the annoying side.

What do gyms have to do with 3-day expiration windows? Most gyms are small businesses with a lot going on, and making videos to post on YouTube or Facebook is just one of the many many things they have to get done in a day. I was getting a ton of support emails from customers ordering a video on a Friday and then... having a weekend! Come Monday afternoon, their video would be gone already. It's easy to point to the order emails where it clearly says they were kept for 3 days, but the honest truth in product development is that most people don't read anything they don't feel like they have to. Issuing a coupon for someone to re-do their video isn't the end of the world, but it's a pain in the ass for everyone involved (and that customer has to wait for it to render again).

So how much will it cost me to switch over to 14 days of hosting? Less than $1/month for the foreseeable future.

I think this is a big win for the business and the customers.

No Watermarks / Free Renders

I ended the recent watermark experiment, turning watermarks back on for now. Ultimately, this was a really crappy experiment from a design perspective. My traffic mix has been changing lately (more email, more social, less search) and I believe that recent dips in average revenue per visit have more to do with the traffic mix than the disabling of watermarks. I want to run this experiment again when I have time to do it properly—actually having a control and an experimental group that I can directly measure against each other instead of hoping that my numbers are flat enough month-to-month to just do the lazy thing and toggle the watermarks on/off.

Besides watermarks, I'd love to figure out a way to make the intro maker free for some users. My render servers sit idle most of the time, and it would be really fun to figure out a way to put them to work. Maybe that's some sort of queue or timed waiting period. "Gamification" makes me a bit ill as a game dev, but maybe there are activities other than giving me a direct payment that move the business forward. If I can figure out a way to measure those and tie them back to a user profile, maybe I can figure out a way for users to earn a free render. Maybe that's overthinking it! It would probably be a lot simpler to build a referral system that directly rewards people for giving me a shout out—a win-win for another day.

Happy intro making!