English Español Deutsch Français Italiano Português (Brasil) Русский 中文 日本語

How to Do YouTube, Q&A with Tim Schmoyer




Sadie Sherran: Hi, hello everybody and welcome to this live stream from and brought to you by SEMrush. Now, SEMrush has got a great reputation for finding the very best speakers and they really have outdone themselves this time because we have today, the YouTube legend that is Mr. Tim Schmoyer. Welcome, Tim.

My name is Sadie Sherran and I am director at Falkon Digital, which is a UK digital marketing agency. My background is in technical SEO. More recently, we've been very more towards video marketing, YouTube SEO, video SERPs, and using video for brand exposure and conversion rate optimization.

Today, if you ask a child what they want to be, studies have shown that a third of children either in the UK or the US want to be YouTubers. Tim, you have been in the YouTube industry for... well, before it was cool. I mean, in 2006, you've been helping creatives build their channels. How did you get into that?

How Tim Schmoyer Started Out on YouTube

Tim Schmoyer: To answer your question Sadie, my very first video was on March 2nd, 2006. I was in graduate school down in Dallas, Texas. And I just wanted to know, could I get video footage off of this little FireWire device and onto this brand new platform that had just opened these doors a few months prior, called YouTube. 

I was dating this girl and I wanted a way of figuring out how do I introduce her to my family back home across the country? We would make little videos that today we would know as vlogs, but back then that wasn't a word, it was just being awkward in public with a camera. We'd make little videos going out to the park, going out to eat, go to movies and things like that.

I would publish these videos on YouTube, email the links back to my family, and I was like, "This is great." But other people started watching. I was getting a little bit nervous when I saw other people outside my family watching and commenting, I'd be like, "Who is catlicker72? And why do they keep commenting on my videos? What's going on here? Should I be concerned? 

I started reaching out to other people at that time, asking them, "How does this platform work? How are people finding me? Why do they keep coming back? Why are they still watching," and at that time in 2006, people are like, "We don't know Tim, but if you figure it out, let us know. We would love to know the answer to that as well."

I started diving into it. We kept making videos, everything from our engagement to our wedding, our honeymoon, our first house, first job. And at that time, we were doing about a million views a month on YouTube, which back then was unheard of, that was ridiculous. 

Other people start reaching out to me, they're like, "Tim, how are you doing this?" YouTube started contracting me to train their employees how to do their jobs, and I was writing a lot of the curriculum that YouTube was using to educate their creators and even their internal teams. 

I was there with YouTube from the very, very beginning. Then in 2013, I started this agency, Video Creators, where we are all about helping people grow their audience on YouTube so they can reach more people, change their lives.

We've done strategy for Disney, for Warner Brothers, for eBay, Budweiser, HBO, a lot of great brands, all the way down to just small business owners and hobby creators who are just getting started, organically helping all of them grow by over 17 billion views on YouTube and 71 million subscribers. It's been awesome. I actually ended up marrying that lady and we somehow managed to have seven kids in eight years.

Sadie Sherran: Wow. I did wonder if that was your wife. Yeah, that's brilliant. It's amazing.

Tim Schmoyer: We live here in Cincinnati, Ohio, homeschool our kids, and run a business online, which now is what everyone's doing. Again, we were doing that before it was cool as well, I guess so. Yeah, so that's all about my story.

How YouTube Discoverability Really Works

Sadie Sherran: I'd really be interested to hear about how it is you get started with your clients when you're working with creators or when you're working with brands?

Tim Schmoyer: I know one of the biggest questions that a lot of creators have is, how do I get discovered on YouTube, I want us to talk at a high level, how discoverability on YouTube works. The way we approach this is, if we understand what YouTube's goals are, then we can craft content that helps YouTube accomplish those goals. And if our content helps YouTube accomplish their goals, then they will be far more likely to surface our content and put it in front of people. 

If we understand what the YouTube algorithms are trying to do, and we can help them accomplish those things, then we will win. And the goal of the algorithm is to surface the right video to the right person at the right time. 

Okay, so all of these complex systems, how do they do that; deliver the right video to the right person at the right time? Well, they look at a couple of different things that are known as viewer signals, which you guys as SEMrush people know what those are.

We'll go through some of these pretty quickly here’ the first three because I think these are the ones that you guys are most familiar with. What videos start the viewing session for viewers? Or what videos bring people back? 

And when they come to YouTube, what do they want to see? You can see that in your analytics simply by looking at traffic that's coming from the homepage. Those are often the videos they're putting right in front of people there on the homepage. 

The second thing they want to know is, "Okay, we see what people are clicking on when they get here, but then how much time do they spend watching it? And that's measured in watch time, the longer someone spends watching a video, the more valuable YouTube deems that video to be for someone.

If someone spends five minutes watching your video, and two minutes watching someone else's, all of the things being compared considered equal, which they never are, but let's assume that they are, then your five-minute video should outperform the two-minute video. 

But it's not as simple as just that because they're also measuring the overall viewing session, which is, well, maybe someone only spends two minutes watching this video, but they stay on YouTube for another 30 minutes after watching this video. Whereas even though they spend twice as much time on your video, they only watch for maybe another two minutes and then abandon YouTube. They also want to know, "What gets them here? What do they watch when they're here and then what keeps them here the longest?"

They started running into problems with this because they started figuring out, "Oh, we can make really long viewing sessions." But then they found that some of the content that keeps people here the longest was not content that really made people feel good about their viewing session.

In fact what happened is, a couple days later, a few weeks later, that person just doesn't return to YouTube because they only spent two hours the day before, it didn't end as a good viewing experience. And so now YouTube looks at this from a different perspective, they're saying, "Not all watch time is considered equal." 

How YouTube Measures Viewer Satisfaction

The biggest thing now, that is the thing that YouTube is really looking for is satisfaction. How satisfied is the viewer with the videos that they have here? YouTube doesn't only want to just surface videos that people watch, but they want to surface videos that people indicated are valuable to them based on these viewer signals, but also based on satisfaction.

They measure that very briefly, in terms of a lot of different signals, but the main one being the satisfaction surveys that maybe you guys have noticed showing up on the home feed with a video, they will ask a little survey before you watch the video that says, "Do you feel like this is a good video recommendation for you?" And then they also have a survey that could play out at the end of other content is like, "Hey, what did you think of this video?" It's just like a quick little emoji that you tap. A

They're collecting millions of those today, and they are getting very, very accurate at figuring out what type of person and what type of viewer will be satisfied with what type of content.

What's happening here is the way you get a video to rank number one and be positioned favorably on YouTube is to make content that people react favorably to...whenever they watch it. 

Optimizing for YouTube: Focus on People, Not Robots

Most people think it’s about the metadata, or it's about the keywords or it's about the tags. And YouTube is like, "No, we are not a platform that is trying to find viewers for videos. We are a platform that's trying to find videos for people." We can't take this mentality of what we bring to Google search and just assume that they're owned by the same company and just flop it over to YouTube search.

Typically, all I have to ask people when they're kind of in that mentality is, "Okay, well, how's that working out for you so far?" You'd be, "Well, I took a video from 10 views to 20 views." I'm like, "All right. I'd rather take a video from 10 views to a million views." 

There's really not as much of a thing anymore as like ranking number one for everyone because results are so personalized, they're based on so many different viewing signals. And so if you really want to dominate YouTube, it really comes down to optimizing for people, not for robots. 

That's the main creed that we drive home with our clients is, YouTube's recommendation system finds videos for viewers, not viewers for videos. If you really want to grow, you have to follow the audience and where they're at. It's not about hacking the algorithm, it's about creating content that really gets your audience to be like, "Oh, I want that. I want to click on that. I want to watch that. That's exactly what I've been looking for."

When we take clients from 2,000 subs a day to 39,000 subscribers per day in growth, 100% of that... No, no, it's more like 97% of that traffic comes from homepage and suggested traffic. We're not even looking at search anymore. We're going after like, "Let's let YouTube put our videos in front of the right person at the right time, even if they're not even looking for it."That's where explosive growth comes from. 

Keywords and Topics on YouTube

Sadie Sherran: Thank you so much for that. We've had a few people put some questions forward. “How to optimize videos for search if there's no statistical data on the queries and the volume is not known, is there a way to optimize for a topic so that it ranks for keywords around the topic?” I mean, I think this sort of follows on quite well to what you were saying about actually the keywords aren't what you should be looking at, it's what the user intent should be.

Tim Schmoyer: Exactly. Yeah. I'm going to sound like I'm contradicting myself a little bit, but I think there's a clear distinction here that we should probably point out, and that is that keywords still matter, but not from the perspective of YouTube's algorithm cares what you're putting in there. They really do not care at all, what you're putting in there. 

Keywords matter again, from a human perspective, not from a robot perspective. Let's say I'm typing in for best Nike shoes to buy or something, now I'm the viewer, am indicating that I am looking for something about best Nike shoes to buy, and so I, therefore, expect that the content I'm looking for should say something about Nike, shoes, buying, purchasing, those are the type of words that I'm looking for. 

It's not because I use those keywords that make the video now perform better, it's because those words are more closely aligned with the value the viewer wants to consume, and thus, that becomes a relevant video for someone making that query. 

But it's not because it contains the keywords in it; it's because the viewer is saying this is the type of value I want to consume. As long as we use words that portray value, we're expecting to see those things that indicate this is the value you want to consume. It's still a viewer-driven system, it's not an algorithm-driven system, does that makes sense?

Sadie Sherran: Yeah, I think it does. We use keywords, we use tools like SEMrush of course, and we use keywords for our video marketing, but we don't get the video and then use the keywords. We do the keyword research before we've even thought about storyboarding, filming, anything.

There's certain words that will be more popular than others. You know that that's a more popular way to go, direction to go, not that it's necessarily going to be in your title or anything else, it's more about how you then storyboard the video and create a video that is going to give people what they're actually searching for and actually looking for.

Tim Schmoyer: Yeah, exactly. We don't use the word keywords anymore internally in my team, because keywords has this connotation of, "I got to put this word in this spot in order for this video to perform well." We think about it more in terms of... we call them topics or ideas.

What we have found is, there's a lot of valuable keyword tools out there, but when we do the actual query on YouTube in an incognito window, it's still going to be personalized, but even then, we'll see that, "Oh, half the videos that are ranking number one for this thing, don't even use this keyword in it." 

Revitalizing Old Videos on Your YouTube Channel

Sadie Sherran: A question from Angie Simpson now, if you've got any recommendations for revitalizing old videos, so old videos that have been live for six months or a year or longer, are they lost in the YouTube graveyard?

Tim Schmoyer: A lot of the time, it's just that content just wasn't very good. I think some sort of level of humility and self-awareness (is needed). If people were engaging well with it, then it would be getting more views, but if it's not, it's not dead. The first thing I'm saying is, just kind of do a gut check and like, "Is this really the best video for someone on this topic or around this story or whatever."

The second thing and this is probably more practical, what you're actually asking is, we have certainly resurrected a lot of content, we're like, "No, this is the best content, this is really solid. People ask this question all the time, why are they not finding this video?" And we will go back and update titles and thumbnails on that. Maybe about half the time, if the new title and thumbnail is better, then we will see that it performs better.

Updating metadata, tags, descriptions, and captions don't matter either, by the way, I know a lot of people kind of think that if they put keywords in the captions that'll make a difference, it doesn't at all. Updating title and thumbnail can if the content is solid, but if the content is not, then updating title and thumbnails is just kind of a waste.

YouTube Thumbnails, File Names, and Video Descriptions

Sadie Sherran: Another SEO question from Will Marcy, do you label the file for the thumbnail with the description? And how many words can you use?

Tim Schmoyer: File name doesn't count. I was actually in the room when this myth started at VidCon about four years ago now, five years ago, it was a panel and there was some MCM with multi-channel networks, which aren't as much of a thing as they used to be, but one of their managers was saying, "Oh, yeah, we do this and all these videos did way better because we think YouTube when they see the keywords in the file name, that's the first thing they know about that video." 

YouTube has come out multiple times themselves and said, "That doesn't matter, we don't look at, it doesn't make a difference." It also makes sense when you understand what YouTube really wants is not to surface videos. This is legitimately straight from YouTube, they say that anything that is classified, or anything that the uploader has to do that the viewer doesn't care about is classified as a bug.

Anything the uploader has to do that the viewer doesn't care about is classified as a bug. That means the tags, they don't look at tags, and the only reason tags are still there is because sometimes they do use that for misspellings that people have. But for the most part, tags, the viewer doesn't care about your tags. They don't care about your captions.

Topic Research Tips for YouTube Videos

Sadie Sherran: Jonathan Devore has asked, can you describe some of your topic research? What is it you do to research your topics?

Tim Schmoyer: The quick version is that we start first with the target audience, it always starts with who are we trying to reach? Who are we going after here? When we do our consultations and one on ones, we always start with the client, describe your most ideal viewer, if we're going to grow this to a million subscribers, who is that person going to be? 

We figure out who we're trying to reach? And if it's an educational channel, what's the problem this person is trying to solve? If it's an entertainment channel, it's still the same question, but they're trying to solve different problems with entertainment than they are with education.

Once we have a really solid idea who it is that we're going after on YouTube, then we start brainstorming. We don't actually go after search because we see much faster and explosive growth always from homepage and suggested traffic. 

In fact, if you look at all the traffic sources on YouTube, search is one of the smallest traffic sources of the overall volume that's available on YouTube. Everyone goes to the homepage before they even make a query, but not everyone who goes to the homepage is actually making a query. Some are just following the rabbit trails of videos, right? 

This suggests an homepage becomes far more powerful for discoverability than search. That's what we're targeting. We'll start with title and thumbnail. Usually a team of us will kind of brainstorm and go back and forth and then kind of vote on one or two.

After we have the title and thumbnail already figured out, then we'll go ahead and we'll be like, "Okay, how do we craft this content? What is the expectation that the title and thumbnail is setting? 

Then what do the opening seconds need to be of this video that affirms for that viewer, yes, you're in the right place, what you clicked expecting to get is in this video; holds their attention. Then we will go ahead and just make that video. 

If we're targeting search traffic specifically, we'll deviate a little bit and instead, we'll figure out who Tom is and what his things are. And then we're like, "What would Tom be searching for?" And then we will just go in an incognito window on YouTube and just literally make those searches and those queries.

We're looking for what videos are showing up on the first page there? And we're paying very careful attention to the titles because usually, the search query that's made, there'll be some videos in there that match that pretty closely, but there's usually just about the same number of videos that don't match that as closely. 

We're paying attention to what you indicated, the user intent, is what YouTube thinks that is when they're making this query. We pay attention to the titles and the thumbnails and how those two things work together to create some tension in the brain for the viewer that will cause them to click, we look at the opening seconds of that. We're also looking for which one of these videos looks like they have a lot of views compared to the amount of subscribers on the channel.

The next thing we do is then we want to know, well, is this video still gaining momentum or is it dead? Did it go big on Reddit once and then just kind of died and that's why it got a million views? We'll use vidIQ, Chrome extensions and other tools to kind of look at the historical data of the video to see if it's still getting momentum or not.

Then we'll look at the suggested videos to that video, what's the next video people naturally watch on their viewing session? Because sometimes, this video is going to be really hard to outrank, they get on the first page of search results, but we could be the next video they click after.

It's a combination of doing some research on YouTube, paying attention to titles and thumbnails, following the rabbit trail a little bit yourself of what videos is YouTube suggesting next based on this video and get an idea of what YouTube thinks the viewing session will look like for a viewer who's searching for this.

We very rarely end up making a video that the title is the exact search query, we end up making something that can stand out from all the videos who are using the exact query. 

Sadie Sherran: We do something quite similar, but we tend to plan out a few different thumbnails before we've even started filming and usually photograph them first and we look at what's going to stand out. We might use a complete opposite color than what everybody else has done. 

Long Audio Files on YouTube (Podcasts, Interviews, Etc)

Ryan Phelan has sent in a question, have you got any advice on best practices for videos that are an hour long? Podcast interviews, and he's got quite a lot that needs to be released and he wants to think about the best strategy.

Tim Schmoyer: YouTube is going to be doing some things that will make, how should I say this, not podcasts, but videos that are primarily audio in nature, they're adding a thing that will make that content more consumable on YouTube. And that will be coming probably later this month if it's not out already.

Sadie Sherran: That sounds really exciting actually.

Tim Schmoyer: Yeah. And so they're looking at a few different signals. Again, it's all viewer signals, there's nothing you do to indicate to YouTube that this needs to be indexed this certain way. They're looking at signals like, "Oh, we found that most people watch this in a background tab, or most people minimize this." 

Or if they have YouTube premium, they learned that most people just listen to this on their phone, minimize and they're not watching it. That doesn't have to do with content length or anything. 

The answer to your question, I think you're going to have an easier time with that content going forward here pretty soon than you would have in the past. In terms of optimizing it, it's all changing. Until it rolls out, it's hard to talk about what it is.

Let me just say, optimize the way you would normally optimize a video for a viewer and just let YouTube figure out the rest. That's probably the best answer right now, and they will and they're super smart, the algorithms are ridiculous. That means they more benefit-driven titles instead of descriptive titles that create more intrigue. And yes, the only other thing I would say on YouTube is to make sure in the description, you're adding time codes to it...so people can easily skip ahead and get the different sections of your video.

Should You Have a Certain Audience Size Before Live Streaming?

Sadie Sherran: Okay, so Team Hazard writes again, would you recommend having a certain size audience before live streaming? Does it help if you have a single subject for your live stream?

If you're going to do a live stream such as this, I mean, obviously, SEMrush, they've got a quite a big following, they've got globally loads of people, loads of fans. It's quite a safe live stream that people are going to be watching and people are going to be commenting. We've had loads of questions, which has been absolutely amazing, thank you guys. If you did have a smaller channel, would you recommend live streaming?

Tim Schmoyer: I would for a few reasons, because one is you learn a lot, and you make a lot of mistakes when you live stream. And it's better to make those mistakes and learn those lessons with a smaller audience than with a larger one. 

It just even comes down to feeling comfortable on camera and knowing how to manage everything that's happening behind the scenes and whether or not you got someone cool like Kate, who's behind the scenes here, kind of bringing questions up and stuff. 

But not everyone has that, so you got to learn to manage all that stuff on your own. Yes, I would (live stream) just for the practice and the skill, learning the skill, the second reason I would is because...it's a different experience when you're live, even if let's say you have two people show up, just two, you can make those two people feel like the most amazing two people in the world.

If you make them feel like you're talking directly to them, you're going to probably go away with that live stream with only two viewers and the two most loyal viewers you could probably have going forward, right? It's an amazing tool for community and audience development as well.

Sadie Sherran: I think that goes with anything on YouTube, really. If you look after your audience, they're going to look after you. We work with some very small brands and we build that channel up, so small businesses that nobody has heard of, and it's a case of instead of how am I going to get loads and loads of subs, don't worry about subs, what's the goal? It's brand awareness, it's to turn them into loyal customers, it's to have that loyalty, you have to give them something. 

And if you can give them something that is worth something, whether it's one person, 100,000, million, then that's good content, and you're going to be successful. And it's just scalable, really.

What to Focus on When Optimizing a YouTube Channel

This is a good question from Covideo, they want to optimize their YouTube channel, can you tell us the top three things that they should focus on?

Tim Schmoyer: The first thing I would ask is optimize for what? Optimize for views? For subscribers? For sales? For revenue? It's a couple of different optimization things that you optimize for. But let's assume, usually, when people ask this, they're probably thinking views and subscribers. 

The first thing is that it needs to be drop-dead stupid, simple, clear for that first time viewer, they're asking the two questions, kind of is this for me and do I care about it? And so how do you answer both of those questions in literally a half-second? A finger snap? How quickly can they pick up on that? 

Because if they did a lot of homework, and maybe watched a bunch of your videos, maybe they could figure it out, but they're not going to do that. It's like, what's the first thing they need to see in that channel art, and then it's the channel trailer? They're very rarely taking time to actually read, and that's why it needs to make sense really quickly.

What does your target audience need to see to immediately feel like, "This is for me. Where have you been my whole life? This is exactly what I've been looking for." Then that value that they're going to consume and get from you, how obvious is that? 

When you're optimizing your content for growth, you really have to optimize this for that first time viewer who knows nothing about you, never heard of you before, and frankly, just doesn't even care. They're just there because that title and thumbnail pitch the value that they want to consume and they're there to get more of that. 

Using YouTube to Increase Sales 

Sadie Sherran: You did also mention about sales as well as views and subs. Using YouTube for sales, how would you recommend doing that? Because you're going to be taking somebody away from YouTube to make a sale, but does YouTube take into that satisfaction consideration, that although they're leaving on this occasion, they're going to come back at a later date or?

Tim Schmoyer: We want every video to get lots of views, to get lots of subscribers, get lots of comments, get lots of likes, we want to generate a lot of these, get a lot of sales, we want to rank number one, we want to go big on Reddit, we want to be highly shareable. You can't make one piece of content do all these things. We're like, "You're going to pick one goal and you're going to craft this piece of content to accomplish only this one thing."

There's three things that we craft content for. Each of them is their own bucket. The first bucket is discoverable content. These are videos that we're crafting intentionally for our first-time viewer, never heard of us before, really high-value production, and the only call to action there is to get someone to click and watch this next video, right? No, "Hope this was helpful, see you later." 

The second bucket is community content. That's when these videos are not intended to go big, they're just intended to grow the know, like, and trust factors with our existing audience. And the only call to action here is to ask for some sort of engagement, typically, it's comments or likes.

The last bucket is the sales bucket, and that's where it's still good, valuable content, it's not a commercial, it's good, valuable content, but it's creating an itch that then you're going to present a sale at the end.

Sadie Sherran: Yeah, solving a problem.

Tim Schmoyer: Yeah, it could be any time you're ending the viewing session. So it'll be ‘buy, download, register, sign up, check out whatever, click the link down below’. It's above the fold and screen card. Take a solid one or two minutes to pitch that thing. And get people to take that action and go off-site to that video. 

If people actually take you up on that offer, you should then no longer really be performing well long-term, but that's okay because you're going to have the discoverable videos that are constantly bringing in new people, you're going to be building the relationship with the community and then generating the sale.

Sadie Sherran: We are going to have to wrap this up. It has been absolutely brilliant to talk to you, Tim, some fantastic insights into YouTube. And thank you so much for joining us. 

Tim Schmoyer: Thanks guys.

Sadie Sherran: Until next time.

All levels

Check out other webinars from this series