Building a Scalable YouTube Strategy to Accelerate Amazon Performance

Paul Sonneveld
I'm your host, Paul Sonneveld, and today we're diving into a powerful but often underutilised growth strategy for Amazon, driving traffic from YouTube. With me today is Brett Curry, the CEO of OMG Commerce. Brett and his team are masters at building YouTube campaigns that don't just get views, they drive high-intent traffic and convert shoppers into customers on Amazon. In this session, we'll explore why YouTube matters more than ever for Amazon growth and how agencies can make the case to their clients, and of course, how to get started in the right way. Brett, welcome to the show. It's an absolute pleasure to have you here.

Brett Curry
Hey Paul, excited to be here. Love this topic, love what you're doing, and we play in both the Amazon and YouTube spaces as an agency, and so really excited to dive in.

Paul Sonneveld
I'm personally very excited. I mean, I think you're one of the, we work with a lot of agencies. I don't hear the word YouTube a lot. So very, I think I met you at the Prosper Show. Very interesting to talk, go deeper in this topic. I also know personally for our business, we do very little on YouTube. And I know it's actually, it's definitely an opportunity we're missing out on.

So I'm looking to definitely take a couple of notes for myself here on the side as well. All right, well, let's jump straight in. Why do you think, what would be your pitch, right? Why do you think YouTube is such a powerful but underutilised or underused lever for Amazon growth?

Brett Curry
Yeah, I think it's difficult because it's not well understood and it's not that easy to translate success from other platforms directly to YouTube. So as an example, brands are successful growing on Meta or successful growing on TikTok. You can't take that creative strategy exactly and port it over to YouTube without making changes and expect success. So that's part of it. It's harder to get the creatives from other channels to work on YouTube. It's also harder to measure. It's just a difficult platform to measure. We'll get into that. 

But why is it a big opportunity? It's a big opportunity because that's where users are and that's where users are spending a lot of time. Some data shows that 42 minutes is the average session duration on YouTube during the prime time hours of 7 to 10 p.m. There are more people that watch. YouTube than any cable network combined. YouTube is the second most popular search engine. It's not a search engine, it's a video sharing site, but more people search on YouTube than any platform not named Google. 

So a lot of people search for how-to information, review information, things like that on YouTube. And then it's also the most streamed platform on connected TV. So if you look at Netflix, Hulu, Peacock, Disney+, all of those, YouTube, more people stream YouTube than any of those platforms. And Netflix is a not super distant second, but it's a decent distant second. But then if you take like the next five streaming platforms and combine them, YouTube is bigger in terms of the number of people streaming it on a regular basis. And so just huge opportunities. It's got the right users. Users are spending time there. There's really no reason that it shouldn't work to drive sales and to build a brand and to build a business, even on Amazon. It's just not as straightforward as other platforms.

Paul Sonneveld
That makes sense. There's some great stats there. You certainly convinced me. I think the other thing is, as I was thinking about YouTube, clearly there's a lot of people there just for entertainment, right? But there is, there seems to be some higher intent type activity going on beyond just a, Hey, it's Friday night. I'm just consuming content cause I want to chill. 

There are people, as you're saying, it's a search engine. You know, I, I don't know. I was looking up how to my heating is broken, right? So I was, I was on YouTube yesterday trying to work through, okay, well, what are some sort of, you know, self-taught mechanic type tips here. So yeah, very high on its end.

Brett Curry
Yeah, and there's a lot of growth over the last five years or so of unboxing videos or product review videos or things like best travel pillow, best toothbrush, best fill in the blank, best shoes for first time runners, things like that. The number of queries related to that has really exploded in recent years. Because for some things people would like to see an answer in video form i want to see this thing in action before i purchase it and so there's some high commercial intent related to physical products as well on YouTube. 

And then, of course, there are people that just, you know, they've got their favorite creators that they watch on YouTube, whether that's Mr Beast or someone that's, you know, niche to their, you know, their area of interest or, you know, I bought a truck a couple years ago and then looked at accessories and I found all kinds of, you know, 4x4 and off-road video channels that I really started to consume and had a lot of fun doing it and took notice of some of the products they were recommending.

Paul Sonneveld
Yeah, absolutely. So let's put this more in an Amazon-specific context, right? Are there certain products or categories that you feel work better with YouTube, you know, clearly four-wheel drive accessories and parts, but Are there certain products and categories that work well with YouTube, and are there maybe others where you would encourage people to stay away from YouTube?

Brett Curry
Yeah, I think the products where you stay away from YouTube for the most part would be things that are commodity-type products. I think the quintessential example in the Amazon world back in the day was like Garlic Press or Pizza Cutter or something like that. If you're just kind of competing on price and you got a quality Pizza Cutter, quality Garlic Press, things like that, just lean into search. That's always going to be more query based traffic driven. YouTube is going to be tricky there unless there's something really unique about your product. And I would have a hard time understanding what would be super unique about a pizza cutter or a garlic press. So the commodity products, probably not ideal for YouTube. 

The areas that are great for YouTube are either higher consideration products. So I'm willing to do some comparison shopping here. I'm looking for a new pillow. So is it the Tempur-Pedic pillow that I want? Is it some other kind of material that I'm looking for? Is it the shredded memory foam or the cooling gel memory foam or whatever? So there's a high consideration there. I'm researching. Is it either a higher order value, so I'm going to be spending some money here, that usually means it's higher consideration set, or is this something I'm going to purchase again and again? 

And so we lean a lot at our agency into supplement brands, food brands, CPG brands, like a native deodorant, and we've got sort of coffee brands and things like that where unique story, unique angle to this product, and we sell someone one time, we've got a pretty high degree of confidence they're going to buy X amount over the next three months, six months, that sort of thing. 

And part of that is because YouTube is not the same as sponsored product ads. It's not the same as Google shopping ads. It's not the same as meta retargeting ads. You're generally trying to reach a net new customer. They weren't searching for you. You're showing them how awesome your product is. And now you've just got to convince enough of them to go and buy it. 

And so this is where I think a lot of Amazon, especially Amazon first brands, get into trouble is like, OK, I'm used to getting a 300% return on ad spend on my sponsor products. I'm two weeks into YouTube. Why is it not doing the same thing? Well, it's like it's a different beast. It's more like TV ads than it is like sponsored product ads. 

Paul Sonneveld
Yeah. 

Brett Curry
And so, higher AOV, higher consideration, ideally a consumable or a product or business given to repeat purchases, all those things line up. And then it's also gonna be something that's visual, like can I demonstrate this or at least tell the story visually? 

Now, it's great if it's something that I can demonstrate and show in action, right? But it can also work like we got this great brand, it's a healthy coffee creamer, but one of the founders is Phil Mickelson, the pro golfer. So there's a great story there because it was kind of formulated for Phil as he was going through some health issues and now it's like his go-to and I drink it every day, I love it. 

So it's still just a creamer, that's not super interesting to film, but the story you can tell, and that's powerful. And you still show someone making an artful cup of coffee and stuff like that, and putting this for Wellness is the name of the brand, or the good stuff, putting that in there. So it's gotta be given to visual storytelling as well, because that's what the medium is, it's video storytelling. And so yeah, so those are some of the components that can help you to decide is this a good product for YouTube or not.

Paul Sonneveld
Yep, got it, got it. Okay, so let's talk about some of the nuts and bolts here. What does a typical YouTube to Amazon funnel look like? Walk us through it.

Brett Curry
Yeah, so this is where things get a little bit trickier, a little bit trickier for YouTube. So YouTube is not inherently a click-based platform, meaning A lot of people that watch YouTube are, or most people that watch YouTube are there for a specific purpose. Either I want to watch this creator or I'm listening to this podcast. 

It's also the number one podcast platform in the world right now. So I'm here watching this content. I'm not as likely to click off as I would be if I was on Meta, right? Meta, I'm just kind of going to hang out and see what's going on and doom scroll or whatever, just waste a little bit of time. Much more likely to click and add if it's interesting inside of Meta than I am in YouTube. 

So YouTube has, half the click-through rate of meta on average. So YouTube is more of a view-based platform. It has a lot more similarities to TV than it does to other social platforms. Now, there is certainly a social component. You got YouTube Shorts. A lot of people watch it on their mobile device, so it feels a little more social, but people aren't clicking YouTube ads as much. And so what that does is that kind of wrecks tracking and attribution, right? It makes it really difficult then to piece together, okay, I saw this YouTube ad and that led me to search on Google or search on Amazon. But how do we as advertisers connect that back? So I know we'll get into more into tracking in a minute, but I want to kind of set that up. 

Generally speaking though, we like to run YouTube ads, send someone to a lander. Ideally, if it's someone selling on Shopify, we're sending them to a specific lander on Shopify, we're pushing the sales there, because what inevitably happens is if someone clicks on that Shopify store, maybe they read up on the product, pretty likely that if I'm an Amazon first customer, I'm going to go check out, can I get it for the same price, maybe even a tiny bit more on Amazon, I'd be okay with that because I get free shipping and get it fast and Amazon's got all my details so it's easy there. Generally speaking, we're sending to a separate landing page and not directly to Amazon. It's also nice if you send someone to your own page, maybe you can capture them on your email list. You can pixel them and remarket to them. So generally speaking, we like to send them to a Shopify or other lander before they go to Amazon. So that's generally how we like to start.

Paul Sonneveld
Yeah, that makes sense. I mean, obviously you're making the investment off Amazon. um, you know, why sort of send them into the big shopping mall at Amazon? Let's just get them into your own little dedicated shop first. And yeah, as you say, maybe you can grab some contact details or whatever that looks like.

Brett Curry
Exactly.

Paul Sonneveld
Talk to us a little bit more about the creative elements, right? I think you made a really interesting point around, you know, you can't just take whatever's working on TikTok or Meta and just go, Oh, let's just upload it on YouTube and rinse and repeat. You hinted really clearly at the start of this podcast that it's different and that's where a lot of people trip up. So talk to us more about, what are the top creative elements that make a real difference on YouTube?

Brett Curry
Yeah, so a couple of high-level things to consider is that, you know, when you're running, say, a video add-on meta, the video is there, maybe the video is the thumb stopper, the scroll stopper, but then you've got that block of text above it, call-to-action buttons around it, things like that. Well, on YouTube, the video is it. The video stands alone. So the video's got to do everything. It's got to hook somebody and demonstrate the product and overcome objections and provide social proof and get someone to either click now or to come and search for you later. So really, the video's got to do all of it. It's got to do all the heavy lifting. 

So, because of that, in general, we find that slightly longer videos do better. If we're trying to measure a direct lift in sales from YouTube, then generally speaking, 60-second to 90-second ads work better. And I think that makes sense if you can keep someone engaged for a full minute or a minute and a half, and we even have a lot of ads that work that are three minutes long. Be a little more on CPMs that way, but if they stick around for a minute to three minutes, and you can keep their interest and keep them engaged, there's gonna be much more likely to buy than if they just saw a few seconds of your ad. 

So slightly longer is better on YouTube, so that's important. You wanna tell the full story on YouTube, or at least you don't have a block of text around it to lean on. Provide all those elements I talked about, you know, the hook, the product demo, the social proof, the overcoming objections, the offer, the CTA, call to action, all that's gotta be in the video. And then, you know, a few things to think about is like, hey, it needs to feel a little more like a TV ad. It can be, it can feel raw. We love having user-generated content or, you know, raw type content in there, but it's gonna probably feel a little more like TV than a Meta ad or than a TikTok ad would as an example. 

One kind of tricky thing on YouTube is that ideally you want to have multiple dimensions for each ad because as a YouTube user, I may watch my favorite podcast creator on connected TV, I watch shorts potentially on my phone, and then maybe there's some sports influencers, Colin Cowher, I really like that guy, I watch him on my mobile device. And so really each of those formats, you kind of need a different aspect ratio of your video, right? 

You need a 16 by 9 for anything that's connected to TV or if I'm watching in landscape on my phone. I need a 9 by 16, so more of a portrait view if I'm watching on my phone either a 9x16 or maybe a 1x1 or a 4x5, similar to the meta aspect ratios. And then I'd like to have that minute to 90 second video, maybe up to 3 minutes. But I'd also like a slightly shorter version to run on shorts, because on YouTube shorts you need something that's under a minute. Those are some of the elements to consider. 

The first five seconds is absolutely critical. You want to hook someone, but you want to hook the right person, so your ideal customer. And in general, you want to hook them in a way that ties into your product. This coffee thing is fresh on my mind, but one of the hooks they led with was stop putting crap in your coffee, this healthier creamer with all the garbage in your coffee. So that gets me hooked a little bit, right? Sometimes you can lead in with something kind of funny. 

So True Earth, long-time client. We kind of helped them launch on YouTube back in the day. Laundry detergent strips. So instead of having a big jug of laundry detergent that you pour into your washer, these are like little strips of paper you put in your washing machine. And the opening was a witch saying like, what should you never mix with water, right? And so you like, you know, mix, you know, water with witches. And then it talked about laundry detergent, which was kind of like a, what? Why would I not mix, you know, water with laundry detergent? 

Anyway, so you want to hook somebody, but hook them related to your products. That's super, super important. So those are a few of the elements to consider. And hopefully, as I laid that out, one, it makes sense, but two, it also lines up with, okay, yeah, this is why it's maybe not quite as straightforward as, you know, meta ads.

Paul Sonneveld
Yeah. And in terms of producing this video content, what's your advice? You said actually user generated content's okay. Obviously you need a good, good storyline and you need a hook. Um, but what about like, you know, I'd love to see how you guys tackle this, you know, when you do this in house, whether you have, uh, you know, uh, videographers or specialists who really help you create this sort of, you know, how do you trade off? Do I do this myself? Uh, how polished does it need to be? You know, how do you think about kind of, practically doing things like this, right? Because I mean, a lot of agencies that are watching this podcast, they don't have whole video content teams. So if they want to sort of. get an experiment going with a couple of clients? What is a pragmatic way to create some of this content?

Brett Curry
Yeah, I think you don't have to be overly polished in your production quality. YouTube can handle a higher production value. Again, it has some elements that are a lot like TV, so if you had a high production value video, like as an example, our friends with and partnered with the folks at Raindrop in San Diego. They did Dr. Squatch, and we partnered together on Native, and they've done things like Shady Rays, and Manscaped, and a number of others. These are really high-quality, TV-worthy. They've done Super Bowl commercials. 

So YouTube can work like that, whereas I think sometimes on meta, the real high production value doesn't work as well. But you don't have to start there. We're big believers in MVP type videos, minimum viable products. So let's prove out a hook. Let's prove out an offer. Let's prove out an angle that we're going to run here as an ad. 

And so I think one way we look at it is if you get top-performing video ads on Meta or on TikTok, those can become your hook for your YouTube videos. So maybe that's the beginning of your video. Then we do need a product demo of some type. We need some social proof, whether that's just showing reviews and get 10,000 five-star reviews, or here's some actual reviews from customers, or here's some user-generated content that we can show. Here's our offer. So it's a coupon, it's a discount, it's a something to get someone to try. And then we're actually telling people to click now and try this out. 

So often we're repurposing some of the elements of other winning videos from other platforms, but we're needing to tie it together with a narrative and likely add to it. So maybe you got that 15-second video that's crushing on meta. I need to pull in other assets b-roll and other footage to make that you know ideally a 60 second, but maybe you start with 45 seconds and and kind of iterate from there. And so we recommend to people is like hey try out different hooks and angles and offers and As you really start to dial that in and you're starting to get some scale. That's when you maybe go big on production value or you can then really go to the moon. 

Paul Sonneveld
Yeah, great to see that those social channels can still play a role in just maybe just defining that hook. Actually, that's a great tip. So let's talk about tracking and metrics and particularly demonstrating that things are working and money is being well spent. So how do you tackle that conversation?

Brett Curry
Yeah, this is the hardest part. And I'll just warn everyone. There's no easy answers here, right? But it can work. It does work when run properly. And so you just got to be up for the challenge of measurement. And again, I'll reference, it's a little bit like TV, right? Some great brands have been built with TV and the proof should definitely be in the sales volume and seeing sales numbers is just not always directly attributable. 

Let me talk about ways you can track. First thing I'll mention is there was the largest incrementality study ever done on YouTube was recently released by Haus Analytics, H-A-U-S. They did 190 incrementality tests with YouTube really quickly for those that don't know. Incrementality tests are usually done based on geo holdouts. So I pick one DMA or one city, I run YouTube, I pick another city that's similar demographics, similar size, similar current sales. I don't run YouTube, I compare the results. You need to do those over multiple cities to kind of see what's working. 

The idea behind incrementality is what's the real impact of this ad, right? So yes, we can see in-platform ROAS, but as we all know, that can kind of lie, right? If I spend a ton of money on branded search, ROAS is through the roof, but how incremental is that? Would I have gotten those sales anyway? Yeah, a lot of them I probably would have. 

So what this showed was, what this study by Haus Analytics showed is that for every, whatever you're seeing in platform, in YouTube, so in the Google Ads platform, your real incremental results are probably three and a half times better than that. So I'm seeing a one ROAS in platform, I'm probably seeing a three and a half X in reality. And they've done this over now, 74 brands, 190 tests, like these are big spenders, like AG1 and Oura Ring. and a ridge and things like that. 

And so lots of data behind what they're saying. And so then it's like, well, why is YouTube so bad at this? It goes back to what I said before, right? People aren't clicking on TV. Over 50% of views now are on connected TVs. When it comes to YouTube, I'm not as likely to click even if it's on my phone. I'm probably logged in on multiple devices in multiple ways. So like with Meta, I've got one Meta login, one Meta account. I may have 10 gmails and I'm logging the different things like on my work computer versus my mobile device or whatever. It's just really hard to tie things back together. 

So here's how we look at it. We do look at in platform numbers. Most of that we're looking at like click through rate because some people still do click, but click through rate, look at a view rate. Um, anything north of 20% on in stream is pretty good. But as you're approaching 40, 50, 60, you know, I heard this, this video, people are watching it. They're choosing not to skip this ad. That's good. I'm looking at things like total watch time, so I'm looking at some of those things. 

Sometimes I'm sending it to that custom lander that we talked about. Maybe I've got a specific offer on that lander that I'm only offering to YouTube, so I'm tracking performance on that specific offer. I'll give you an example. We had this hair care brand that we scaled from zero on YouTube to about $600,000 or $700,000 a month in ad spend on YouTube profitably. And they had a defined three-month LTV they were trying to hit, so they knew they could also make up a number $100 CPA worked for them to use a round number. And so that's what we were aiming to get. 

What's interesting is as we got a little ways into the campaign, their data scientist team came back and said, Listen, we've been watching the Amazon numbers and we think for every one sale you're getting D2C, you're getting two additional sales on Amazon. If you go back and read that Haus analytics study, it would not be a surprise at all. They found the same thing. So then it's like, all right, we're looking at baseline on Amazon and what is our growth rate on Amazon? How is that accelerating? How is branded search accelerating on Amazon and how is branded search accelerating on Google? There are ways to measure both. You can do search list studies on Google. On Amazon, you just kind of have to track things on your own. 

I think there are a couple of tools to do it. But that's really the best way to do it is I'm looking at some in-platform numbers to tell am I reaching the right audience? Are they engaged? Is the creative good? Are some people clicking even though click-through rates are going to be worse than on meta? But then I'm tracking sales and how are sales trending up. We did a big test for Arctic coolers. They are a Yeti competitor and we did a test to see, can we lift Walmart sales using YouTube? We do like a 19-market test study. It's phenomenal. We won a Google agency excellence award for it, but the average was like 15 to 18% lift in-store sales after driving YouTube. 

So basically, we're triangulating the data. So we are looking at it in platform inside the YouTube ad campaigns. We're looking inside of Shopify and Amazon to see kind of the trends there, and then we're trying to layer on things like search list studies, brand list studies, things like that, to see what is the actual impact here. Because a few people will click after seeing a YouTube ad, but those that don't, but those that are interested, their next activity is likely going to be to search, to search on Google or search on Amazon, and we want to track that.

Paul Sonneveld 
Yeah, that's a very comprehensive answer there. Unfortunately, no easy answers.

Brett Curry
I wish there was like, plug in this tool, you're good. And that doesn't exist.

Paul Sonneveld
Yeah, sorry. Yeah, unfortunately. But you know, I think the way I summarise that it was really around like, look at the in-platform metrics around engagement, people really watching. I think, you know, the question really, are people skipping the ad or not?, is a really good one to ask. And then it's really around, you know, are you seeing uplift versus your baseline versus that control period, right? Really, really interesting. 

We've got just a few minutes left of our time together. I just want to ask maybe one or two final questions here. The other one is really around mistakes, right? What are the biggest mistakes that you see people make? Or maybe that you at OMG Commerce made when you first started out, like what are the things you encourage other folk listening to this podcast to avoid?

Brett Curry
Yeah. I mean, generally speaking, it comes under the wrong creative and that's what's so important. You know, Google says that 85% of the results you're going to get through YouTube campaigns are driven by the creative. So generally speaking, it's someone pulling a creative from another channel, not optimising it for YouTube and that's usually a recipe for disappointment. That's a big piece. 

The other piece is just relying on in-platform metrics. So I'm only looking at my ROAS inside of Google Ads or my cost per acquisition or CAC, my customer acquisition cost. I'm only looking at that in-platform and I'm not checking in other areas. I'm not looking at the total lift of YouTube. So those are the two biggest pieces. And then some of it is also not targeting the right audiences, so we're not reaching the right person. But I think if you go back to the marketing basics of, I've got the right message, I'm delivering it to the right person at the right time. And so generally speaking, if you're not succeeding, it's one of those things, right? It's not the right message, it's not landing, it's not to the right person, and then you're not delivering it in the right way. And so we do see some mistakes being made with audience creation. 

We like to lean into the uniqueness of Google, where Google's got all this behavioural search data. You can lean into that search data. You can build audiences of people based on what they're searching for on Google and on YouTube. We love that audience, especially to start with. And we do like to, to lean into conversion-focused campaigns. So campaigns where we're giving Google a target of, Hey, we want to get people to opt into this list or people to buy this product. Those tend to work a little bit better than just view-based campaigns. And so generally speaking, it's one of those mistakes.

Paul Sonneveld
Thank you so much, Brett. Unfortunately, we are, I'm watching the clock here. I know you have a hard stop as well. We're going to have to wrap it up of this episode here on Marketplace Masters. Brett, thank you so much for coming on, being super generous with all of your insights and really helping us to sort of peel back the layers on YouTube and really giving us some real practical insights and advice. Really, really appreciate it. You know, YouTube is clearly one of the most on-tap growth leaders for Amazon brands. It's really exciting how agencies like yours, like OMG Commerce, are really leading the way in this piece. So yeah, thank you so much. I look forward to maybe touching on some of the other social channels in the future as well.

Brett Curry
Let's do it, Paul. Thanks for having me. Really appreciate it. And yeah, don't sleep on YouTube.

Paul Sonneveld
Take care, Brett.

Brett Curry
Thank you.

Paul Sonneveld
All right, everyone. That is it for today. If you're watching this live or catching the replay and you want to learn more about OMG Commerce or Brett's approach, we'll include some of the links in the show notes. And don't forget, you can access a full library of Marketplace Masters episode at merchantspring.io. 

And of course, if you are an agency looking for better analytics and reporting for your Amazon clients, including performance on YouTube, we'd love to show you what we're building here at Merchantspring. Thank you so much again for tuning in. I'm Paul Sonneveld. See you next time.