4 ms·
The question not being asked is how your audience changes A vs B in other ways than size. You can easily A/B test yourself into mediocrity by constantly optimi
by colechristensen 5y ago
The question not being asked is how your audience changes A vs B in other ways than size.
You can easily A/B test yourself into mediocrity by constantly optimizing for a larger, dumber audience. They stop liking your old content (despite your dumb faces, etc) so you make new content they like more, keep that cycle up for a while.
And you end up making emotionally charged rage-porn political videos and have become the next Facebook/cable news/etc.
This is also why the second or third album for a band is usually the best; they get popular and start letting their audience design their music and become repetitive and mildly awful.
- ASalazarMX 5y agoCreators are optimizing for money, YouTube's algorithm is the one optimizing for audience size.
- goohle 5y agoHalf of general audience has IQ below average.
- marginalia_nu 5y agoThat's not how averages work.
- goohle 5y agoThis is how IQ is defined.
- marginalia_nu 5y agoNo, it is not. 68.2% of the population is in the average span of 85-115 IQ. 15.9% is in the below-average span of 0-85.
- goohle 5y ago> No, it is not. 68.2% of the population is in the average span of 85-115 IQ. Yep, 50% of them are below 100 IQ, 50% are above 100 IQ. Official definition of IQ: measure average IQ of a population, then label the point where 50% of population is below average as 100 IQ points.
- readflaggedcomm 5y agoIt is. Specifically, IQ is assumed or asserted to follow the normal distribution, where mean equals median and mode. By definition, half are below the median. The normal distribution's median is equal to the mean, so half are below mean.
- WastingMyTime89 5y ago> The question not being asked is how your audience changes A vs B in other ways than size. That's not how it works. The most important factor to get views on YouTube is the recommendation algorithm. It makes a huge difference. YouTube will not even push your videos to people who are subscribed to you if the recommendation algorithm doesn't favor them. As the click-through rate is an important part of how the algorithm decides what to recommend, you need a catchy thumbnail and a somewhat clickbaity title or you will be reduced to oblivion by people who use one regardless of the actual quality of your content. The story is the same with the length of video. Veritasium made a couple of very good videos on the issue if you are interested.
- geerlingguy 5y agoThis. And the Veritasium video is linked later in this thread.
- yessirwhatever 5y agoWhat the parent comment said still applies. Algorithm or not, you can become a top pop singer either by having a really good voice and being talented or by being singling about how you enjoy something controversial just to create outrage. The only difference is your audience.
- BoorishBears 5y agoThe algorithm accounts for that. If you sound like a dying cat, your click through rate will be high, but your average watch length will be low. I agree there's a problem with how the algorithm drives creators to this stuff, but it's also not that unsophisticated that clickbait = success
- yessirwhatever 5y agoYet, all this does not justify clickbait.
- WastingMyTime89 5y ago
- qwertox 5y agoAlso, I believe Jeff would have even more viewers if he held a bra instead of a Pi in his hand, with his facial expression being the same. Jeff, could you do a test and report the results? ;) Maybe dealing with Raspberry Pis isn't as lucrative as you thought. Other than that (I also get triggered by the thumbnail thing), your channel is one of the very best on YouTube.
- jgalt212 5y agoA/B testing attracts the bots in my experience. Peoples' tastes change quicker than the bots' decision models do. Given that's somewhat of a bold statement, then try this thought experiment: If A/B tests actually work, then wouldn't you expect there to be zero films and/or TV shows released that are flops?