Transcription
We hit 100,000 subscribers in just a little over two years, but here's the thing: we could have done it faster. In this video, I'm talking with my producer Connor about the nine biggest mistakes we made and how we corrected them so you can grow faster than we did.
Thank you to Artlist for supporting the channel.
I'll jump in and do the first one: packaging is everything. I did not understand this at all in the beginning. I didn't even think in terms of the word "packaging." Now, I think about packaging for every platform that I publish on.
But packaging is everything. Another big add-on to that, from our episode with Patty recently, is that packaging improvements are not linear improvements on the outcome of a video. Packaging improvements can be exponential. A title with a 10% improvement could have twice as many views, four times as many views, or even ten times as many views as a title that is 10% less effective.
Now you know why I fight so hard over certain words and titles. The mistake I often make is being a bit snooty about what our titles should be. Connor comes to me with a great title, and I say I don't like it because of some reason that I think will reflect poorly on me and my integrity. I don't like clickbait, blah blah blah. But Connor is always there to say that doesn't matter.
It's just a YouTube title; it's not that deep. When I think about it, you know, it's like our Jenny Hoo's video. I don't actually know the exact title of that video. I couldn't tell you verbatim what that title is. I know directionally what it is, but the point being, I am closer to that video than anybody else, except for you maybe. I don't know what the exact title is, but that title did so well that it drove three and a half million views.
So why would I worry so much about a word here or there if we think this is the best package to get the most views? The job of the title is to get people to click and get into a video that actually backs up that title. But to get the click and get people into the video.
One of my takeaways was that packaging is as important as everyone says. We hear that all the time, and you're like, "Yeah, but what if I just have a good video? Will that succeed?" Usually not. You have to get that click.
Something Patty mentioned in his last episode was hitting that core casual new fit. Even for a show like ours, where our new or casual viewers aren't pure entertainment watchers, trying to capture a view from someone who might not be a hardcore creator but is just generally interested in what's happening on the platform or with certain people is important.
Those small viewers can snowball into getting more viewers that will stay on the channel. It's so, so important. I was so bad at packaging early on, and I think you just get better with time. Now, I'll have 25 to 40 different title variations, usually around 7 to 10 different actual titles, and then word changes within those.
We take that super seriously now and have a whole process of narrowing down and hitting different click triggers, like what Jake Thomas talks about. The reason it's important, and I'm going to dovetail into the second one, is that even if you think this video isn't for some people, you don't know that YouTube is not showing that person the video on their home feed.
If YouTube is giving impressions to people who you think maybe this title doesn't appeal to, well, if it doesn't, they're not going to click it, and that's going to impact your CTR over time.
So if I were starting again, trying to solve for a couple of things—how do I get people to click on the video? That's solved by packaging. Then how do I make the video good enough for people to keep watching? That's average view duration.
I would focus on solving the CTR question first because if you can do good packaging and get people to the video, then you're going to get a lot of data, and you'll know whether that video is good enough or not.
Do you agree? Would you do it in that order? If I am fairly new, would you say focus on packaging a little bit more in the beginning than AVD?
I would say idea and then packaging, which I think those two kind of play into each other. You can have really good packaging, but if it's not a good idea, I don't think it's going to be clickable anyway.
But yeah, I don't think retention is something you'd worry about until you've started to get good at what you're doing. I don't think looking at retention graphs is necessarily what beginners or small channels should do. One, there's probably not that much data, and you're still getting better. There are more important things to improve on first than getting super nitpicky with retention graphs.
The idea is that you can package it in a way that people click. Look at your CTR, try to raise your CTR, and then you're going to get more data about whether people like the video. You'll get more people seeing the video, and that's going to tell you a lot more about the quality of that video and your retention down the road.
I see a lot of people post screenshots of their video stats from YouTube Studio on Twitter, and what I notice a lot of times when the views are low is that the CTR is very, very low—like sub 2% or sub 3%.
So my takeaway is, if someone asks what they should do, I would say focus on getting your CTR up because then you're going to get more impressions, more views, and learn what else you can improve from there. But this is preventing you from even getting more impressions and more data back.
And do that without clickbaiting as well.
Yeah, it has to pay off.
This actually leads me into my other point: pre-production is everything, which includes packaging. One mistake we made for probably our first year was worrying about the packaging and idea and how we were going to frame an episode after we had already recorded it.
We were left with no idea of how the episode would go before we actually interviewed the person, which makes it really hard to create a good frame. In podcast terms, you hear everybody talk about the idea for a YouTube video. In our show terms, that comes down to the topic and the guest that play off each other.
It's not like a typical idea you think of as a YouTube video. We moved from just having guests on and trying to frame the episode after the fact to prepping exactly what the episode is going to be before Jay even interviews.
All of our research is usually based around a certain topic. I'll go through and research what topics our audience wants to hear, whether it's comments or what positions YouTubers are hiring for. Those are generally specialist things that can be offloaded.
If there's someone who owns that area, it'd be good to have them on to teach a generalist. Then we have an entire episode that we know is going to play well with our audience before Jay even interviews, which has been a huge help with packaging and making sure we're delivering on the promise of the channel.
This is an essential flip, especially for podcasters trying to move into a video world. A lot of times, this mindset helps with even just an audio-only podcast. But if you are interviewing people right now, you're probably booking them based on their name, and you just want to talk to that person.
I can't count the number of interviews that start with, "This is a wide-ranging conversation with this person." That might be interesting and fun for you as someone who's very curious, but a wide-ranging conversation is extremely hard to package.
It was a fundamental shift when we realized we should guide the interview to the package rather than trying to fit the interview into some package after the fact. It was never as effective; it was never as retentive.
I would invariably try to cover too much ground, which made it difficult to package around an idea rather than that person's name. It stopped us from trying to be everything for everyone, where now we can be much more differentiated.
Our interview with Patty Galloway is going to be different than the interview Patty did with the Editing Podcast or the interview he did with John Usai. That's how we approach all our guests. We want the experience of watching our show to be different than just watching any other podcast with these guests.
It gives them a reason to come back or just watch for the first time because it's going to be new. Maybe not all new information, but at the very least, a new angle or a unique angle that fits our audience.
Next mistake: giving people vegetables instead of candy. Yes, even if you want to feed people vegetables, they want candy. You've got to make them think they're eating candy in order to take their vegetables.
We actually call this "stuffing candy into your vegetables." The idea here goes back to packaging again, which underlines the point that packaging is everything. People don't want to click on something that seems boring or like work.
You've got to give somebody a package that makes them feel like this is going to be fun, easy, and will get them the big outcomes they're looking for. That's how you appeal to a broader range of people, in my experience.
Especially early on in the video and in the intro. If you're an educational channel or podcast diving into scientific topics, of course, you're not going to be super surface-level the entire time.
But you definitely don't want to lead off with the thick of things. You've got to ease people into it. I think of it like syllabus week versus the actual final exam. You're just giving the overview in the beginning, keeping it light, drawing people in.
You're not just hitting them right out front and expecting them to care as much about this thing as you do right off the bat. You have to lead them in with some lighter information, which goes to making sure the readability of your script is good.
You don't want massive words or anything like that. I know we get a lot of comments on the Jenny video about the downfall of society being writing shorts or intros at a fifth-grade reading level. That's the standard for journalism; that's how it's always been.
It's probably what led people to click on that video in the first place—the fifth-grade reading level of the title. You can be mad, I guess, but no one's saying you have to make content for fifth graders.
It's just a level of ease in bringing people into your content that's important. You can dive into the deep stuff later. You don't start right out of the gate with that.
It's the difference between a title like "How to Do Better with Monetization" versus "How to Earn $10,000 per Month." People don't like big, hard words. They don't need academic-sounding ideas; they actually just want outcomes they can wrap their minds around.
Is a voiceover generator something you can wrap your mind around?
I don't know that I like your tone.
Artlist, is this better? Or maybe this?
Yeah, those are better. Why don't you go ahead and tell the audience about our new feature?
Artlist is the first creative platform to offer a voiceover generator created specifically for video creators, and it's so impressive. Now, when you go to artlist.io, you will see a new voiceover tab with a ton of exclusive voices to choose from, modeled from professional voice actors.
Show them how it works. First, we'll choose a voice, click on it to select, then go to the text box at the top and type in what we want, and then click generate.
Jay hit 100,000 subscribers in just a little over two years, but here's the thing: he could have done it faster. In this video, he and producer Connor talk about the 11 biggest mistakes they made and how they corrected them so you can grow faster than them.
It's that easy. The voiceover generator is included in an Artlist Max subscription, which we use along with great music, sound effects, cinematic stock footage, and motion graphics templates too, which is an insane deal.
If you just want the new feature, there is a standalone voiceover plan as well. Get two free months on any annual plan using our link in the description. Thank you to Artlist for sponsoring the channel.
This relates back to what we talked about earlier regarding pre-production. The idea is that guests do not impact growth in the way you think. A lot of times, you'll hear people saying, "We thought this when we first started: to grow a podcast, you just need to have big guests."
The problem with that is if you just have a big guest and rely on their name, no one's going to stick around. You might have an influx of views, maybe even an influx of subscribers, but if you don't give them a reason to stick around and watch the next video, they're not going to.
That's why we made the shift to packaging or focusing more on topic than the actual guest name. The only guests who will reliably and consistently grow your audience are people who have the same or similar audience as you.
For us, that would be people like Patty, Ed, John Usai—those types of people who have an audience looking for the same thing as our audience. Of course, those episodes are going to do well and consistently help the channel because it's a similar audience.
If we had Mr. Beast on, I'm sure we could get a ton of views, but most of those views aren't really going to do anything for us because they're not the audience we want. In fact, that episode could end up hurting the channel because if it gets millions of views and those millions of views aren't our audience, then the next video we post is going to be shown to all those people who aren't going to click on it.
Then YouTube thinks people don't like this video when it's just being shown to the wrong audience because we attracted the wrong people. When you package a video around an individual, there do seem to be more potential risks than potential upsides.
I was listening to Colin and Samir's most recent audio episode, and they were talking about their episode with Scott Galloway. Scott Galloway is a well-known person to a certain audience, which turns out is not huge in overlap with Colin and Samir's core audience.
When you package a video around a person who is well-known but not to your audience, that is not going to perform when shown to your audience. They're not going to know why they should listen to this person if they don't recognize him.
If it does get out to the audience that likes Scott and brings them to your channel, if that's not your core audience, it's going to at best water down the stats and the understanding the algorithm has of who likes your videos, or at worst, not attract people at all.
It's a risky proposition to try to package around the identity of a person, and it can happen accidentally too. I think we had that conversation after the success of the Jenny Hoyos video. We could easily double down on making shorts content and get tons and tons of views right after that.
That's not something we wanted to do. Even at VidSummit, I was talking to a friend, Trent Hara, about the hangover of a viral video. Even in the best-case scenario, if you hit a really good viral video and it still hits a lot of your target audience, you can still have that hangover for the next couple of videos.
People can be one-and-done with your channel, and it can take a little bit of time to get back on track and get back to being shown to the audience that you want. We did see that on our channel.
I was just looking into our stats because we did see that hangover. I think the episode right after was Dodford, which is a great episode, or it was the thumbnail video, which is also a really good video. You should definitely check that out.
Then Dodford, the YouTuber's guide to hiring, is a more niche video. We knew that wouldn't do too well, even though it's probably one of the most informative videos on our channel. Justin's brand deal video is another one that's incredibly valuable but wasn't watched as much as we thought it would be.
Then we started to get back on track with Dr. K, and I think that was a good one because Dr. K, as a psychiatrist, kind of applies to everybody. Mental health is a really good topic to talk about.
We framed it around getting views, which I think is really interesting—like a psychiatrist helping people get views has that performance coach vibe. Obviously, his name draws a lot of people, and I think the way we packaged it was a nice overlap between his audience and people who would stick around to watch our videos.
His name actually isn't in the title or the thumbnail; it's just his face. This speaks to one of the lessons I had: treating your guests as teachers and not celebrities.
Instead of packaging around their identity, package around whatever unique thing they can teach. You're basically creating workshops through a person rather than profiles of a person. Profiles are inspirational; workshops are useful.
I think workshops have a much longer shelf life and are much more evergreen content than a snapshot of someone's life. Especially because if they do interview rounds, there are going to be 10 interviews just like your profile.
You might be a really good interview, and it might be a great profile, but there are still 10 others. If they're giving a workshop, it holds its value as long as the information is pertinent.
A lot of what we do is on the audio side. I've interviewed a bunch of authors who were on their book tour—not a bunch, but a handful. I've noticed that over time, those interviews have performed a little bit worse.
What happens is a lot of times when you interview one of these people on a tour, their team will ask if you can air this episode during a specific time period—usually about a week—because they're trying to do it right as the first week of book sales is coming out.
You literally, as a listener, get a deluge of, "In my feed, I have this person on these four podcasts I listen to." I'm going to listen to one, maybe two of those. If you're not in the top two shows or channels for some viewer who is now seeing this person everywhere, you're just not going to get that view with that person.
I heard Tim Ferriss say the same thing on his most recent episode with Kevin Rose: he's not going to do book tour interviews anymore. You really have to think about how to frame this differently than everybody else.
Even if you do, listeners don't necessarily know that you're going to be different than other shows or that the way you're being different is what they want. It's a challenging thing to try to capture the personality when they're coming up from their hole of doing work and saying, "I'm ready to be interviewed right now."
I think it adds a unique angle to your show when you do that too because a lot of the draw of a typical podcast or interview show is the audience feeling like they're a fly on the wall of this conversation.
We very much want to take the approach of you and the guest talking to the audience, which gives a unique feel. I think that's one of the reasons why our show is successful: there's a better connection from guest to audience where you're not just listening in; you're being taught and talked to.
What do you have next?
You need an intro for anything, especially a podcast. Because you're making a long-form video, you have to let people know what you're going to talk about. It's so, so important.
Early on, we would do typical podcast intros, like what you used to do with audio, where you do three minutes of catch-up and talking, which is great in a medium like audio where people are just kind of listening and doing whatever.
But people who don't already know you just have no patience or tolerance for that. It's a huge difference between audio and video podcasting. YouTube's reach is so big; you're getting so many people who have never seen you and just don't care about you.
That's not even bad; they just don't know you. You have to earn their time to watch the video, and the way you do that is with a good intro.
With that, you need to pay off on the packaging right away. All the time, our intro usually has the hook clip or the hook sentence—if I don't use a clip—somehow related to the packaging, either the text on the thumbnail or just the title in general.
I want to pay off on that package right away and let the viewer know that this video is exactly what it said it would be. It affirms the reason why they clicked because they want to learn that thing.
Then, setting up who your guest is, which I think is important. If you're interviewing, never, never, never ask someone to explain who they are or what they do.
Please, no. People don't know how to talk about themselves unless they're a trained speaker. That's when you're always going to get the "ums" and things like that, and people are going to stumble over it because it's kind of uncomfortable.
It's just a waste of your time, the guest's time, and the viewer's time. They're going to be super humble or really off-putting; there's no in-between.
If you do it yourself, then you control the length of that as well. You do the bare minimum of setting up who this person is and why the audience should listen to them, then go into exactly what the episode is going to spell out.
That's the format for all of our intros: hook, which is a sucker punch clip. Your hook needs to have zero context, no buildup. I don't need any explanation; that's pure candy.
We're not even trying to hide vegetables in it or anything like that. It's pure candy because that's the hook. It has to be fast to get people's attention.
Then we're going to set up who the guest is, why they should listen to them, and explain what's going to happen throughout the interview, generally with some more hooky clips or explanations as well.
Once we started doing that, I believe the first Jake Thomas episode was the first one we started to do something similar to that. It didn't really hone in on what it is now until, let me look, it was probably around Jay or Jenny when I started doing the more refined format of hook, setup, explain, that kind of thing.
It's worked. That's when our channel started to blow up. There are a lot of different factors for why it's not just the intros, but that certainly helped a lot.
You need that for people who don't know who you are. You need to set up why they should be there. There's this classic speaking advice that is the format of "tell them what you're going to tell them, tell them, and then tell them what you told them."
This is a three-part way to get people to grasp this and be bought in. I had this note framed a little differently, but I noted that basically the highest leverage way to increase AVD over the long length of the video is to increase AVD in the first 30 seconds.
If you're able to do that, it has a meaningful impact across the entire video. This is something I remember Patty talking to us about in the accelerator way back in the day. He looked at our dashboard and said if you can increase retention in the first 30 to 60 seconds, then you're going to have a massive increase across the length of the video.
So when you do get to that point of focusing on retention, the highest leverage place to do it is right away because that's where you see the most people drop off. If you can turn off that spigot a little bit, that's going to be really great for your whole video.
It's very similar to the exponential tip, just like packaging is retention as well. People are way more fickle in the first five minutes of watching a video regarding whether they're going to keep watching or not versus 20 minutes in.
If you're 20 minutes in, it would take a lot for the actual video to get you to click off. But in the first minute, where you're still evaluating things, you're much more likely to click off.
So spending way more time and effort—whether it's editing, writing, or in our case, reformatting the episode to make sure the stuff at the beginning is related to the packaging and is the really sweet stuff—goes a long way.
We're just kind of taking a bet that if they like this stuff earlier on in the episode, they'll stick around until the end. If not, that's fine. If they only came for the first thing, that's just how it works.
But really focusing on the intro and the first couple of minutes of your video is crucial. The interesting thing about this is there are two shows that have very different takes on this.
One is Diary of a CEO, which dials it up to an 11 or 12, where their packaging is kind of rage-bait-y. If we're talking about candy, this is the highest fructose corn syrup out there. The beginning of these clips is always super no-context-needed sucker punches with big-time sound design that heightens tension, getting you into who this guest is and why you should care.
Then you have Chris Williamson, who has been on the show. In a lot of his episodes, there's no intro at all. He literally just jumps right into the conversation. You actually start to question if you missed the intro; it happened so fast.
But now you're in the flow really quickly. I don't know which one is better; it's an interesting test to run at some point, but they're very different and both finding success.
My guess would be that it is harder to grow from zero with the Chris Williamson strategy. I still believe certain things get easier as you get bigger.
As we've seen, I think there is a lot of social proof factor on YouTube. If your video has a bunch of views, it's more likely to get more views than if it doesn't. I think that's similar to certain things like retention tactics or how you do your intros.
You're more established, and if your video already has more views, I think people are less likely to click off if maybe the intro isn't as good because they think, "Oh, all of these other people watched this," or something like that.
The next one I have on my list is that fewer, better videos are better than more, worse videos. In the beginning, we were trying to mirror the audio feed by keeping up with a weekly publishing cadence on the video channel.
That means you have a week to produce the episode, and there are limitations depending on your capacity for how much you can do in a certain period of time. What we found was we got better results when we put more time into making a smaller number of videos better that we thought had high potential, rather than just trying to keep up with the publishing schedule that was a legacy artifact of the audio feed.
I think that also ties into us talking about pre-production too because now different guests or video ideas get vetted way more rigorously. Before, it was just like, "Oh, we have this guest recording; let's make it a video."
Now, there are tons of guests on the audio show that don't necessarily make a video—not meaning they're not good episodes, but it's more about audience fit.
Let's play a little game on the spot. Name one of your favorite YouTube channels that you watch frequently.
Does the niche matter?
Nope. Gift Gash.
Okay, when does Gift Gash upload?
Once a month, maybe. Some of his videos get taken down, but yeah, once a month, twice a month at most. But it's not a consistent schedule; he uploads when he uploads.
I would bet if we did that again for multiple channels, the same is true for me. I can't name a day of the week that I expect somebody to upload or a specific schedule on YouTube. It doesn't really matter to me.
In audio, I do still have some rituals where I will listen to somebody on a weekly basis, but I still probably wouldn't be able to name the day of the week that that person uploads.
I listen to this person once a week, but I think the schedule is far less important now than it used to be.
The counter to that could be if you are a daily show, something like Good Mythical Morning or whatever. That channel, like Rhett and Link, is such a part of their audience's lives.
They've been making videos for over a decade, daily for the most part. So of course, every morning or lunchtime, you know, "Oh, now I'm going to watch Good Mythical Morning." That's completely different than making a full-scale video.
There's the contrast of daily—if that's your thing, commit to that and be a part of people's routines. But generally, it's probably better to focus on quality.
I don't know; short-term quantity probably would net you more views, but I think long-term, business and audience-building-wise, taking fewer videos of higher quality would probably win out.
It's really hard to say, but I would actually assume the opposite in most cases. Depending on the size of your channel and past video success, you're going to get some baseline of impressions, which will turn into some baseline of views per video on average.
That's what a baseline is. My guess is if I posted one video a week and my baseline was that those videos were getting 1,000 views, maybe 2,000 views, then in a month, that would be 8,000 views from those four videos.
If we think about the leverage you get in packaging, it's better if you put more time into a video. I think it would be very likely that you could move your views on a video from 2,000 to 8,000 views with a slightly better package.
It's dangerous; I recognize this is kind of dangerous advice to give because there's a world where you take this advice and become so perfectionist that you don't publish enough at all.
In general, I think if publishing is not a problem, if you have the muscle to publish something, you can probably listen to your intuition and say, "I would do better and make a better video if I gave myself a little bit of extra time to produce it," and that would probably have an exponential increase in views.
But again, it's dangerous advice to give because you can't not publish at all.
The general rule of thumb is that you have to be self-aware about that and where you're at in your YouTube journey.
I also think one of my humbling takeaways is that you never know as much as you think you do. When we started the channel, I was like, "Oh yeah, I know all this good stuff; we're going to be great."
Looking at how the channel is now and how I approach things now versus when we started is such a night and day difference. It's like, man, I had no idea what I was doing.
I know you probably look back at this and think, "Man, I had no idea what I was doing." It's that constant growth where you're never quite at where you think you are.
There are also videos that we expect to do really well that don't, and videos that we think are going to be average that overperform. A lot of times, I feel really good about my ability to judge the trend of metrics in the first 24 hours, but recently that has been shot a little bit.
It is always surprising, sometimes in a good way.
YouTube's hard, man. I feel like anybody who's telling you it's easy probably doesn't know what they're talking about. Even the top people in their field are like, "Yeah, it never gets super easy."
There's always that random variability in things that you can't account for because there are so many factors that go into a video's success. It's a lot of trial and error and just figuring out and growing along with it.
What do you have next?
I think this might be the biggest one I have: end screens are super important. When we started, I did not think an end screen or an end-of-video CTA was important at all.
Early on, we did normal podcast outros where you would recap the episode. Throughout the accelerator, we were talking to Patty and some other creators about a dip in retention at the end of a video and how much that can impact overall channel views.
Viewers binging your content and keeping a viewer on your channel or on the platform in general is really good for your videos because YouTube wants to maximize the time people are spending there.
If you can get people to watch another one of your videos, that is a great signal for that video and the channel as a whole because people are moving down your funnel from new to casual or from casual to core.
If they're going to continue to watch more videos, the other thing we've noticed with a couple of videos is that a successful end screen can put a video back into recommended.
We've noticed with a couple of flatline videos— I shouldn't say dead, but flatline videos where they weren't getting much impressions—we would throw them on an end screen with a CTA.
It's not just throwing the video up and hoping people will click; we do a "this is why you should watch this video" that usually relates to the content they just watched. It's like a bridge: "Hey, this is video A that you've been watching for 40 minutes. If you enjoyed this, you should watch video B next."
It's literally me at the end of the video saying that, and that bridge kind of hits you in the face unexpectedly. The momentum of the interview carries right into it, so you don't anticipate the video ending.
As soon as people feel like the ending is near, they'll probably click off. We're not surprising them, but we want to be very direct with how we're asking them to take action afterward.
We're very clear about that, and we're not building up to the ask the entire interview. The ask is the reason we're asking, and we feel like we could ask them to watch another video because they just spent the last 30 to 60 minutes watching this thing.
We've seen all of these flatline videos that were doing well after being put on the end card or end screen that started to do well on the homepage and recommended.
Increased CTR, increased AVD—all of these things—because it's now being served to either a new audience or just the right audience that it wasn't getting at the time when we posted it.
A lot of these are older videos that boost the entire channel up, and if that end screen is good, maybe it brings up another video. It's this whole exponential thing where worrying about that dip in retention at the end of a video could be a huge factor in channel growth or older videos being put back into the algorithm.
It's like resuscitating.
It's kind of fun because we can take some of the videos that we feel really good about that didn't get a fair shake because their CTR just tanked too early or we just didn't have the audience.
It's fun to try to bridge those and give those a second life. There's nothing more pleasing than an asset you have already created suddenly doing more for you with no extra work on your part.
That's great. The genesis of our channel growth was the first Patty episode, which then led to a bunch of views for the Hayden episode. I forget what was on the end screen of that one; maybe it was Jake Thomas.
Jake Thomas is the first one that really hockey-sticked out of nowhere. Off one video that did well, now we have two others blowing up that contribute financially to channel views and money, but also just keeping people on your channel.
The more people watch your content, the more likely they are to like you and move down the core-casual-new funnel to keep coming back.
The frustrating thing is when suddenly a really old video starts doing well, and you get comments about, "Why did you do this this way? This isn't very good."
It's like, well, it was one of our first videos, and we didn't know what we were doing.
Yeah, it wasn't very good in this specific way; you're right. But that's because this is two years old now.
If people are coming from a new video and then going to a two-year-old video, it's an interesting experience.
The first video we posted, the very first video episode, was a really good episode with Justin Moore, but the first five minutes are like a normal podcast intro where it's just talking about, "Hey, we're starting this new channel."
It did decent views at the time, but I think the AVD was like a minute because people would just come and comment, "Oh, this is so sick," and then just dip.
I don't know if we could breathe life back into that video because the beginning and editing are so much different than what we do now, even though it's a really good episode.
This wasn't on my list, but I want to throw it out there anyway because I believe it to be true, but again, it probably comes with some caveats.
In general, I think doing long-form video podcasts is a hard way to do YouTube, but I do find that having longer videos might actually give you some leeway in your retention editing style.
Here's what I mean: when you have a longer video, even if the percentage of AVD is a little bit lower than your shorter videos, the actual time someone is spending watching a video on your channel is usually higher.
I think YouTube generally is maximizing for time on the platform. I've seen screenshots of other channels and people who have two- and three-hour videos that have a really low CTR.
The AVD as a percentage isn't super high, but those people are watching 45 minutes sometimes, which is something YouTube really wants to encourage.
I remember at one point you and I had a conversation where we started more aggressively editing the length of videos down. We thought, "Let's just make it all really tight."
I think there's a threshold where you can actually go further than is good for your channel on that front. The shorter the video is, that's less time on average someone is watching a video on your channel unless the AVD percentage is super high.
We've played around with that. We've gone from like an hour or more sometimes all the way down to like 30 minutes at one point, and now we've landed somewhere in the middle more times than not.
What's your take on this?
I agree. My general philosophy when I'm cutting episodes is that I want to make sure it achieves the purpose of what we set out to do. If there are tangents during the interviews, those will get cut.
The time only factors in on iffy stuff. Sometimes I'll leave some banter or personality in because I think it's good, even though we could cut five minutes here, five minutes there.
I do think YouTube optimizes for total watch time, not just percentage. Obviously, that's a factor in overall viewer satisfaction on how YouTube is judging whether or not people like the video.
But I definitely think a 20% average view percentage that is 20 minutes is better than an 80% average view percentage at four minutes.
Totally agree. That is an opportunity for long-form videos. Podcasts obviously lend themselves well to long-form video. That is one plus of doing this style.
Something really interesting when I was diving into our stats the other day about average watch time—I don't know what to think about this, but I think it's interesting.
I was looking at the device type of our viewers, and half our viewers are mobile, which blew my mind when I first saw it. Just under 50% are mobile viewers, which is also why I started using big captions in our intros.
You can see them on autoplay on your phone, so you can clearly read captions on a silent autoplay. Those are big enough for people to read.
Also, the view duration between mobile and laptop and TV viewers is non-existent; they all watch the same amount of time. I would have thought that mobile viewers have a lower average watch time.
The one device that does have a higher watch time, I think it's by like a minute, is tablet viewers, which is weird. I don't know; maybe that's more of a "sit on your kitchen counter while you're making dinner" thing.
But I thought that was really interesting, and I don't know quite why. If mobile is half our views, do you know what the breakdown is for television?
I think mobile is like 48%, TV and laptop are both like 20%, and then tablet was like 10%.
Super interesting. I would have expected more TV, less mobile. I wonder if people listening on mobile are treating it as an audio podcast, where it's in their pocket.
If you're listening to this on mobile right now, leave a comment. Tell me how you listen or watch this show. I'm very interested to learn more.
In the last 90 days, mobile phones consisted of exactly 50% of views. Computers were 30%, and the view duration is only five seconds higher than mobile.
TV was 16.5%, and the AVD was 14 seconds higher than mobile. Tablet was 3.3%, and the AVD was three minutes higher than mobile.
This is super interesting. It leads you to believe that people are treating the tablet as a viewing background thing a little bit more, which you would expect would be TV.
That's what we're hearing from all the bigger channels on YouTube that are doing mostly in-studio recordings that are two and three hours long. A lot of them are getting a lot more views on connected TVs right now.
That's interesting that we're a counterpoint. Everything we do is a little bit counter to what we're hearing from the industry for people who are doing in-studio shows.
Doing remote shows is just different. One of the takeaways I had too is that remote interviews on YouTube cannot feel like Zoom recordings.
It's got to be a compelling visual experience, which takes a lot more work in post-production. I think a lot of times, not only do remote interviews just feel like Zoom recordings, they try to capture a similar fly-on-the-wall feel to in-studio podcasts.
At least for more education-based things, I don't think that works. You can get away with it, I guess, if you're more like a comedy podcast because people treat it like they're listening in on a friend's Discord call or something like that.
But generally, it's a completely different vibe, and I think you have to lean into it and not try to pretend that you're not a remote interview.
You have to lean into it, almost like how I started to view it as a Twitch stream, where you're talking to an audience, not talking at somebody else.
I see a lot of people mess up even with camera positioning when they'll try to set the camera up like it's in a studio or filming them watching somebody else on a screen rather than just talking to a camera.
I think that's usually a mistake because that throws off the whole vibe. I can see you're still remote recording; why are we acting like you're filming in person?
I just want to see the camera feed.
It seems like a good time for me to share the workshop we're doing.
Yes! Connor and I are co-hosting a video podcasting workshop. Depending on when you're watching this, you can either join us live or get the workshop and work through it on your own time afterward.
There's a link in the description. This is going to be a three-hour, very interactive workshop where, after this, you should be able to launch or improve your remote recorded video podcasts so that they get more views.
That is the goal of this workshop: to launch or improve your remote recorded video podcast to get more views. We're going to cover three specific sections: pre-production, production, and post-production.
Those are the three aspects of producing one of these videos. That's right, and we will be covering everything I do from packaging, guest and topic selection. Jay's going to talk about how he reaches out to try and get guests.
I'm going to talk about how I write intros, how I structure the edit, and all of those different things that go behind the scenes that make our show what I think is unique.
We're just going to teach everything we've done to get to where we are now. If that sounds interesting to you, just visit the link in the description, and we'd love to see you there.
Connor, I think we've covered all of my takeaways. Have we covered all of yours?
I think so. Let me check my notes to see if there's anything. The one that I had, we kind of spoke to, but I guess I'll call it out explicitly. The last one on my list was to stay out of success jail.
You talked about this a little bit with our Jenny Hoo's video. If you have a video that is such an outlier outperformer on a topic, you're going to be faced with the decision of whether you want to make more videos like that to attract more viewers like that or if this is not attracting exactly the type of viewer that you want.
You need to actively go outside of that because it's really easy to fall into audience capture if you have a video that does well but it's not actually what you want to make. The more you lean into that, the harder the problem becomes to overcome.
I've always said that playing long-term games based on attention is really stupid because you end up becoming a caricature of what you started out to do. You're always chasing maximum views, especially if you're not building a business behind it.
That's a little bit different for us because YouTube is just the top funnel of our business. We have a goal as a channel, and that goal isn't just views.
Of course, we want to maximize views because that helps the entire business, but our goal is to teach. It's to have this back catalog of all these different topics creators can come to learn.
If we narrow down and focus on just shorts content, that wouldn't fulfill our goal as a channel. It might get us more views and more money in the short term because that's what the immediate new audience wanted to see, but that's antithetical to what we want to do as creators and a business.
That goes to the episode we talked about with Patty regarding views maximization: where's the line? We always want to maximize views, but we want to make sure they're the right views because not all viewers are created equal.
Not everyone is going to come back, and we want to make sure that it keeps us happy in what we're doing. If you're making content that you're not happy with just to chase views, that's not sustainable.
You're going to end up hating it because this is hard and takes a lot of work. If you don't like it, you're probably not going to last.
We are grateful to you for watching this and helping us get to 100,000 subscribers. These were our takeaways for how we got there and what we would do to get there faster.
If you want to learn from the Zen master of YouTube growth, watch this video with Patty Galloway, where he explains how he makes any niche go viral.