I Studied 5,400 YT Shorts to Crack the Algorithm

I analyzed over 5 500 YouTube shorts across 34 different niches to try and decode the algorithm and figure out the best way to go viral and I want to show you what I found so here's how it all happened so the other day I was scrolling through Twitter procrastinating as usual and I came across this tweet from Paddy Galley now Patty is a YouTube expert who has generated billions of views with a b and worked with huge creators including lord Jimmy and basically Paddy's tweet broke down a massive study that he and his partner did to decode the YouTube shorts algorithm and despite the fact that most YouTube analytics studies make most people feel a little bit like an Edward Munch painting but the findings in this study are actually so profound that it got a huge amount of views and even Elon Musk did A celebratory live stream in 144p just in its honor all jokes aside though this study is actually a real thing so today I want to Deep dive and expand on it so by the end of this video you will have everything you need to know to start making viral shorts that get you you more subscribers make you more money and get you lots of girlfriends so a quick note though even though this video is probably titled or introd something like I analyze 3. 3 Billion YouTube shorts to crack the algorithm I want to make it very clear that all the credit and insights for this study are 100 with Patty and his team I didn't have anything to do with it I just needed a clickable YouTube title so please don't sue me Patty anyway with that out of the way let's get into this because this is fascinating stuff now this study had five massive findings that will help you get more views with your shorts but before we can get into those you need to actually understand a little bit about how the YouTube shorts algorithm actually works and the best way to understand this is from the perspective of a viewer so if you're a viewer opening up YouTube shorts for the first time ever what YouTube's gonna do because it has no data on you it's gonna send you random General popular shorts that's going to measure which ones you engage with so which ones you watch which ones you maybe like and comment on and it's going to gradually start to build up a profile on what type of content you'd like then once it has an idea of what kind of content you like is going to explore that Niche for the best shorts in it and we'll talk more about what best means in a second and it's going to try and feed you those shorts because those shorts are going to be more likely to keep you on the YouTube platform for as long as possible binging your life away which makes YouTube money so let's say for example whenever YouTube happens to show you a cricket show you really engage with that shot you watch it all the way to the end you like it you comment on it the algorithm is going to take notice of that and eventually it's going to be like hey let's search our YouTube library for other videos of Sports and send them to this person and see if they also engage with them so you'll find yourself getting more Cricket shorts you might get some badminton in there maybe some baseball you know the real like lowest of the low but as we mentioned previously there are probably a lot of cricket shorts and baseball shorts and badminton shorts so how is YouTube know which ones to serve to you so it's going to find the best shorts in those categories and give those ones to you first because those ones are going to be the most likely to keep you on the platform and this is where we can start to get into what best actually means because if we take a step back from this example and put ourselves back in the shoes of the Creator your job is to have your shorts be perceived by the algorithm as the best because that means that your shorts are going to be the ones that are most likely to be pushed to viewers first which means you'll get the Lion's Share of the views subscribers and all that good stuff but instead of me just telling you what the best is I actually want to show you some data on this so here is a spreadsheet full of YouTube shorts data from real YouTube shorts and if some of you are watching this right now I wonder how you can get a spreadsheet of die like this as you come to your channel and you go to analytics and in analytics you find that the analytics that you actually want the data of let's say I just wanted the overview I come to overview I'd click see more it's going to open up a more advanced analytics panel and then you come up here to this download button and you can pick which stats you want to download into a spreadsheet and which ones you don't by clicking on this plus button here and then adding say different columns to your report and then once you're happy with your report you can't appear and you click on this download button it'll export your data so back to as you're about to see this data can give us some really interesting insights so in this shade each one of these roads is an individual short so what I'm going to do is come across to average view duration here which is on average how long did a viewer watch particularly short for and what I'm going to do is I'm going to sort from z2a which is going to order this sheet in a way that the shorts with the highest average view duration are going to come at the top and then the shorts the lowest average view duration will be sorted to the bottom and once we've done that grab the views here so from here I'm going to highlight just down this is a really basic analysis and click down here and grab the average what you can see is that on average the top 26 shorts from this Channel with the highest average view duration have about 3. 2 million views on average now if I change this and sort this to A to Z this is going to give us the lowest average view duration shorts with the lowest one having just nine seconds and then I do the exact same thing I grab our top 26 videos and look at the average has plummeted the average amount number of views here is now about seven hundred thousand this channel has some pretty good shots on it so even it's worst short to do pretty well but 700 000 is nowhere near 3.

2 million so what this is telling us is that if you have a higher average view duration not just see your short but actually watch it for a long amount of time YouTube algorithm sees that as a positive signal that that sure is a good shot and makes it more likely that that shot's going to be pushed out to more viewers I know this is really basic and obvious stuff but often we're always looking for the hacks and I think we need to start on a foundation of what are the fundamentals that really matter and clearly seeing the data here the average view duration is one of those and Paddy's study also backs this up for example this is a graph that encompasses 33 different YouTube channels across different niches and you can see the averages it looks pretty similar right with shorts between an average view duration of 50 and 60 seconds on average got 4. 1 million views think about that so in essence if your short can maintain an average view duration of between 50 and 6 60 seconds especially as it scales chances are on average you're going to get about 4 million views but that's a lot easier said than done so how can you actually do that well our study has some insights in this as well I want to come back to this channel here and show you something else very interesting so if we sort out views again Z to A which is going to give us the top viewed shorts on this channel and if we come across to the duration of this shot so this is an average of iteration this is how long each shot is let's have a look through some of these numbers so we've got 45 seconds 59 seconds 52 seconds 54 seconds we keep going down this list we'll see there's only about three shorts in this entire list that are in the 30 seconds or below and even those shorts two of them are like 37 seconds so they're like close to 40 seconds anyway however if we come into views and we sort this the other way around to the shorts with the lowest amount of views I want you to have a look at this column again look at how these numbers have changed right we've got a number of shorts in the 30 seconds and below I can just see 10 from eyeballing this but in general if you just look over this you can can see that on average the shorts that performed worse tend to be shorter no pun intended than the shorts that are longer and interestingly this is also backed up by Paddy's study as well so you can see on this graph the shorts that Paddy analyze most of them were between 30 and 40 seconds you can see shorts between 50 and 60 seconds performed far better than their shorter counterparts now why is this well again with the caveat that people are actually watching your shots because if you just post a long shot no one watches it then it doesn't matter but a longer short has the opportunity to get more of the Thing YouTube cares about most which you know is high watch time for that shot if you're posting a 30 second short chances are your most engaged viewer will probably watch that short all the way to the end so you'll get about 30 seconds of watch time now some really engaged viewers if you're a little bit sneaky and loop your short might watch your short one and a half times maybe even two or three times but that's the vast minority right on the other hand if you have a short that's double the length so not 30 but 60 seconds most of your most engaged viewers will watch that shot all the way to the end the difference is is an Engaged viewer watching your show all the way to the end on a longer shot gives you one minute of watch time as opposed to just 30 seconds so a way to think about this is that the longer short has the opportunity to get more watch time and a higher average view duration than a shorter shot just inherently it does due to length now obviously there are outliers you can see it in my data for example this shot is one of the worst performing shorts on this channel and it is 51 seconds a reason for that could be as you can see the average view duration for this short isn't actually so great but regardless in general if your content is really high quality you're going to have better results if you create a longer short between 50 and 60 seconds versus a shorter short but we're just getting started here because the study found something really interesting with a stat you might not have ever heard of called vvsa AKA viewed versus swiped away essentially if you go to say one of your shorts through content and then come down to the analytics of it by clicking on say this button and then come across to the reach tab what you'll see is this little box box here how many choose to view which is your vvsa think about it for those of you who know click-through rate for long form videos it's basically like your click-through rate but for shorts because on shorts most people don't really click on them they find them in the shorts feed so one of the big colors for shorts is not how many clicks it can get but rather whether or not people will actually watch it versus swipe away from it so essentially what this measures is the percentage of people who actually watched your short when they came across it in the feed versus the people who'd kind of just landed on it and swiped away from it because they weren't that interested so this is a decent metric to look at if you want to gauge the quality of your shorts hooks whether or not you're able to actually really grab someone's attention the first couple of seconds because if you can't do that you're not going to watch the remainder of the short and they're going to swipe away and that's obviously bad but let's get some real numbers on this now if we come back to our data and we have a look at this column here which is a viewed versus swiped away we just get a quick gauge of it you can see that on average for these lower performing shots the viewed versus swiped away is about 70 so on average 70 of people watch the a short rather than swiping away from it and that's decent which explains why even the lowest performing shorts on this channel still manage to get a large amount of Impressions and Views however if we change this and again we sort by the highest performing shorts what we can see is the viewed versus swiped away stat changes a little bit so now we can say it's gone up 10 and on average it's 80 and that might not seem like a big change but when you're competing against so many shorts and you're trying to be the Pinnacle you're trying to be the best in your Niche so you get served first a 10 difference can be a big difference and the study found some interesting insights on this if we have a look at this graph this is kind of laying out all the different videos and shorts that this study analyzed based on vvsa what you can see down here in this corner if we move across this line you can see that if you have a vvsa less than 40 like chances are your short is not going to get more than 10 000 views on the other hand if you're having a vvsa between 70 and 90 your shorts could be getting hundreds of thousands of views as you can see in this kind of green section here now does just having a high vvsa guarantee that your shot is going to get hundreds of thousands of views not necessarily for example if we come to the channel you're watching right now which I'm going to out myself here but um and we come down to one of the few shorts that I've posted click on it we come across to reach you can see this short has almost 90 percent vvsa and yet this short only has 10 000 views so it's not a guarantee that you're gonna get hundreds of thousands of views but based on the data we can see I would still think it's useful to look at this stat and kind of treat it as a you must check this box in order to have the chance to go viral in other words you can see if you're getting less than 40 vvsa like the chances of you really blowing up are pretty much nil however if you're getting more than that that doesn't guarantee you're gonna blow up but at least the door is open and there's a potential assuming you do some of the things we talk about later in this video and also the things we talked about earlier but with all this being said obviously we can see that the competition for shorts is getting tougher and tougher it's no longer the wild west so a short even worth doing well if we have a look at more data from Paddy's study what we can see here the number of Subs gained per 10 000 views for YouTube shorts versus long form videos and what you can see is really interesting is that audio overall across all of the channels that Paddy studied on average long form videos actually got more subscribers per view than short so per 10 000 views long form videos got 22 subscribers if a short got 10 000 views it was only going to get about 17 subscribers however something that might be important to note here is that on average it's probably going to take you a lot more time to create a long form video than a short so that's kind of something that you'll want to take into account looking at these numbers figure out how long it takes you to create a YouTube show on average how long it takes you to create a long-term video on average and use your own Common Sense assuming you want to do shorts that is as to like whether or not it's actually worthwhile you're doing that assuming your goal is to get subscribers and what about making money we've all seen those videos about how to make ten thousand dollars a day with YouTube shots and you actually go and post a YouTube shot for those of you who are monetized you'll probably have found that your YouTube shorts generate more like 10 cents per day but recently shorts were properly monetized by YouTube and what we can see here in Paddy's graphs are the before and after so before shorts were probably monetized they were earning on average per thousand views about 0. 27 cents now however after monetization they only got 0.