We have 24 subscribers on YouTube, and I’m bullish that we can reach 1,000 subscribers and a million views by December 31. The numbers today do not put us on that path. What makes me want to try is how much we’re learning from making something, watching where it works, and fixing the next thing.
Last week, I wrote about spending two days on a 51-second animation. Two imaginary companies each gain 500 customers a month. One keeps more of those customers, and over time its little storefront grows into a headquarters while the other boards up its windows. That’s customer retention: people continuing to use or buy from a business instead of leaving. I wanted the difference to be visible even with the sound off.
Now that video is giving us a chance to test the same idea on ourselves. We got people to watch something we made. Can we give them a reason to come back?
When I checked on October 21, the finished video had been viewed 5,682 times on YouTube since we published it on October 15. Views count plays, so that doesn’t mean 5,682 different people. 8 subscribers were attributed to the video. People are choosing to follow the channel so they can find more of what we publish.
The more interesting numbers tell us what happened between seeing the video and deciding to follow:
73% stayed to watch rather than swiping away. Think of the stream of short videos on your phone: you can give the next one a chance or flick straight past it.
Average viewing time was 41 seconds. For a video about 51 seconds long, that’s roughly 4/5 of the explanation. It describes the average time watched among views where someone stayed beyond the initial seconds.
91% of views came from YouTube’s stream of short videos. Most plays came from people encountering it inside YouTube, rather than arriving through a link we shared elsewhere.
Through October 20, that one animation supplied about two-thirds of the channel’s 8,310 total views. Our next most-watched video “You Fixed Onboarding. They Left.” had 1,300 views, with people watching an average of 27 seconds, or about 59% of its length. The animation gave us a promising example of people staying with an explanation.
Our hypothesis is that showing the idea immediately, then giving people something meaningful to follow in each scene, helped. We changed the subject, visuals, length, and opening together, so we can’t point to one choice and say it caused the result.
Instagram made another difference hard to miss. The same animation had 2,575 views from 1,813 viewers, with the displayed figures showing all views coming from people who weren’t already following us. It earned 16 likes and 4 bookmarks.
Views through October 20, 2026; subscribers checked October 21. Required pace is a target, not a forecast.
Here is what the year-end ambition actually asks of us. Starting from the channel’s 8,309 views through October 20 and its 24 subscribers, we need another 991,691 views and 976 subscribers. Across the 72 days from October 21 through December 31, that means averaging about 13,773 views and 14 additional subscribers a day, after accounting for anyone who unsubscribes.
Over the previous 28 days, we averaged about 297 views a day and fewer than one additional subscriber. We need roughly 46 times that daily viewing pace and 16 times that subscriber pace. If nothing improved and we simply continued at those averages, we’d end the year around 29,675 views and 86 subscribers.
So my confidence rests on our ability to improve what happens next. The video brought in roughly 1.4 subscribers for every 1K views. If future videos produced that same result, about 693,000 additional views would bring in the 976 subscribers we need.
This is why the work feels connected to everything we make at Torta. A video needs an opening that makes someone stop, an explanation worth staying for, and a reason to watch us again. A product needs to help someone understand what it does, experience something useful, and want to use it again. Getting someone through the door only gives us the opportunity to earn the next visit.
We’re changing openings, trying different ways to tell the story, and testing subjects as fast as we can. The next useful experiment could keep the subject and visual style similar while changing the first few seconds. Then we can compare how often people stay, how long they watch, and whether they subscribe, giving each video the same number of days to find an audience. That won’t remove every difference between them, but it should teach us more than changing everything at once.
You can use that same approach without running a channel. If you’re making a website, showing someone your work, or explaining an idea, pick one moment where they have to decide to continue. Watch what happens there.
Do they understand what you’re offering?
Does the next step deliver what the opening promised?
Change something specific enough that the next attempt can teach you something.
I feel like we’re getting dangerously good at this. We have an opening that persuaded people to stay, an explanation they spent time with, and eight people who chose to follow after watching it. We also have a clear view of how much is still missing.
I’m bullish because we can take those observations into the next video, and keep doing that until we have more than one result to point to.
— Alan, Colin & the Torta Studios Team