With iOS 14 and privacy updates throwing advertisers for a loop and increased ad costs making reaching your ad goals harder than ever, scale on Facebook may seem illusive.
On today’s episode, I wanted to hear from an up and coming Facebook ad manager and co-host of the Rich Ad, Poor Ad podcast Dyland Carpenter. Dylan specializes in eComm media buying and is fanatical about testing, experimenting and finding ways to scale ad accounts. Here’s a quick look at what we cover in today’s episode:
- Common mistakes keeping you from scaling with Facebook ads
- Why creative is everything and how to approach creative testing
- Finding the winning campaign formula - learn how Dylan uses controls and tests to always be leveling up.
- Thoughts on iOS 14
- Budgeting for scale and how to be aggressive without breaking the algorithm
- Plus more!
Dylan Carpenter
Mentioned in this episode:
Episode Transcript:
Brett:
Well, hello and welcome to another edition of the eCommerce Evolution Podcast. I'm your host, Brett Curry, CEO of OMG Commerce. Today we're talking about scaling on Facebook ads and how things are shifting and changing and how you should be scaling right now with your e-commerce business. I've got a guest today that I think you're going to absolutely be blown away by, both in terms of his podcast skills and in terms of his Facebook ads skills.
Brett:
I'm delighted to welcome to the show, he's the co-host of the Rich Ad Poor Ad Podcast. Love that title. One of the greatest podcast names out there right now, in my opinion.
Brett:
He's also the owner of Koality, and that's like the koala the bear and TY, Koality Media. Does a ton on Facebook. With that, Mr. Dylan Carpenter. What's up, Dylan. Welcome to the show and really stoked to be chatting with you, man.
Dylan:
Yeah. Brett, man. Thanks for having me on. I'm getting sweaty listening to that intro. I'm like, "Man, hype."
Brett:
I love throwing out a good intro. You've earned it. You've got a great business. You've got a great podcast. So, yeah, I'm excited about this chat. But you and I, we get to record a podcast. I'm going to be on your podcast either in the future or in the past, depending on when someone's listening to this. But we had a blast with, "Hey, let's flip this. Let's get you on the eCommerce Evolution Podcast and talk Facebook ads." So here we are.
Brett:
But tell me, Dylan, just a quick background, how did you become a podcast host and how did you become a Facebook ad specialist in e-commerce? And so, just the quick story, if you can.
Dylan:
Yeah. So for the podcast, we just fell into it. Zach and I were looking for something to do, to be honest, and he had a huge audience already. So we..
Brett:
Yeah. Shout out to your business partner, Zach, and a quick plug on Zach.
Dylan:
Zach Johnson, FunnelDash. But yeah, we had one called The Benchmark first. It was really metric-oriented. We wanted to know your CPA, your CAC, your RoAS, break even points. But it was a little too personal for businesses to share. So Rich Ad Poor Ad came out. It was a lot harder than I thought it was going to be to launch a podcast. But luckily, we have a good team behind us. We've got three people who do the video editing, the transcription, the website, the blog posts, the copy for all of us. It's pretty -
Brett:
It's pretty labor-intensive, man. It really is a lot more work than I thought. When we first started, we were like, "Oh, you just hit record and then post it and then see." He's like, "No. If you do it right, there's a lot that goes into it."
Dylan:
Yeah. So we fell into it. Luckily, is we've got 115, 116 episodes recorded right now. We've launched about 60, 65-ish. It's getting good traction. We've had Billie Jean on, Neil Patel. So a lot of big names. For Zach, it's a good funnel for FunnelDash. But I mean, on my side, I just love to talk to badasses like you and network myself further and get more out there to build my authority, essentially.
Brett:
Yeah.
Dylan:
But yeah, it's about 80 to 100 bucks an episode to cost us right now. It's slowly racking up, and we're fixing to launch some ads for the podcast actually too.
Brett:
Nice.
Dylan:
That'll be cool. But when it comes to my bread and butter, Facebook ads, I actually randomly fell onto it on indeed.com. I graduated from Texas State here in Austin May of 2017. I did marketing with a sales concentration, and I almost moved to Orlando, Florida to be custom suits tailor at Tom James to pitch suits and sell suits. I played ice hockey. I loved suits. They looked fly.
Brett:
Which, by the way, for someone watching rather than just listening, two things. One, you're rocking a killer stache right now. So I've got to give respect to the stache. You need to watch it to see that. And then also, too, if you look in the background, you've got a picture of the Miracle on Ice, which you've seen the movie. Miracle writes a great one. You play hockey. Is that your favorite movie of all time? I mean, you've got the mural thing, so it's obviously important.
Dylan:
It's up there. My dad won that in a poker tournament, to be honest, and I just loved it. But it's signed by the whole team.
Brett:
Wow, that's awesome.
Dylan:
I couldn't sell it for a hundred K, but I'm probably never going to sell it. It's so sentimental at this point. I'm a hockey-
Brett:
1980? When was that?
Dylan:
Yeah.
Brett:
Okay.
Dylan:
yeah.
Brett:
That's amazing.
Brett:
You were going to move to Orlando, be as suit tailor. Then you thought, "No, I'm going to do Facebook ads instead."
Dylan:
Yeah. They flew me out. I didn't like it. My brother worked at Accenture. I looked on Indeed, Accenture sales job, a random sales job popped up. I applied, got a call. It was for Facebook. I went to interview there, got a job literally two hours after my interview. So I started as a marketing expert. Knew nothing about Facebook ads. I'd boosted some posts in college just to get people to come to our hockey games, but I didn't know what I was doing.
Dylan:
Yeah, after three months there, realizing what I was learning, I was like, "Oh, this is lucrative." I became a sponge, took like 10 pages of notes and journals and stuff, and really just asked a ton of questions. After being there for a year, I left. Started Koality Media, and do a lot of freelance work. It's been history ever since.
Brett:
Yeah. I love it. One of the reasons I want to have you on, one, you're an awesome podcast host, also an awesome guest. But I love getting different people's perspective on how to run Facebook ads. As you know, I'm a YouTube guy and a Google guy and also Amazon ads. I don't play in the Facebook ad space, but I understand it decently well. It's obviously a huge part of most of our clients' media mix.
Brett:
We will get into some of the new changes and iOS 14 scares. Depending on when you listen to this, it could be heating up or it could be all in the past. But definitely want to address that as well. But I love getting different people's perspectives. And so, as we look at scale, because that's what every e-commerce brand wants, whether they're looking to sell their business or hold it and just grow it, increase the value, you want to achieve scale. How do we achieve scale on Facebook?
Brett:
Like you and I were talking about, lot of people listen to the podcast, already running Facebook ads. They're just looking to take it to the next level, or maybe things have stalled out, or maybe they are dealing with iOS 14 issues. From a creative perspective, so the ads themselves, what do we need? What do we need to be thinking about so that we can really achieve scale?
Dylan:
Yes. Creative's everything. I know we're going to talk about ad account structure, campaign structure, which will tie into this at some point, but creative is everything. It's not about the audiences, the bids. It's all your creative, basically. What's really helped us scale, I think is, creative -
Brett:
Just one quick thing I think I'll add to that. Sorry to interrupt. Google put together study. They've got a group called Unskippable Labs and they study YouTube ads that people don't skip and stuff. But they came up with the statistic that 50 to 80% of your success on YouTube is driven by the ad itself, and then the rest is media bids, budgets, stuff like that. Obviously, it's hard to quantify. Some been made up, but I don't know.
Brett:
So you think that's probably somewhere on Facebook, like 50 to 80% of your success is the ad itself and then the rest is media? Your expert opinion, how would you frame that?
Dylan:
It's a really good question. Now, this is where it gets tricky. It depends on your product and your conversion cycle. If I have a quick product that's pretty cheap, under maybe 30 bucks, then I want my ad just to... It's terrible to say. We always optimize for purchase, but I want my ad to make someone click to go to the website to learn more and then buy the product.
Dylan:
Now, there are other brands we have that are a little bit more high ticket, to where we have to go more in depth on the ads themselves, to more or less... Not seldom in the ad, but just reel them in. Then of course we have other hefty ways of retargeting. But it varies, to be honest with you, because it really depends on that conversion cycle. What was the question again? I'm sorry.
Brett:
What percentage of your success is determined by the ad itself? You said the ad is everything. I was just giving a little quantification there from Google's perspective that they say 50 to 80% of your success is the ad and not the media and the campaign structure and stuff like that.
Dylan:
I would say the same thing.
Brett:
Yeah, yeah, yeah.
Dylan:
I have so many accounts that I even run ads for where the click-through rates were killer, the cost per clicks are killer, but hey, there's a product page issue. So I would say it's probably similar to Google there, to be honest.
Brett:
Yeah. Yeah. Okay, cool. Awesome. Sorry, I derailed you, but ad is everything. What do we need to be thinking about in terms of getting our ads to scale?
Dylan:
What kind of budget you have basically, because, I mean, for the account we have spending 70K a month, I'll dive into, versus somebody spending 10K a month. For the 70K a month, we test 10 new creatives every single week. That could be three or four different creatives, different thumbnails, and then mixing some copy there. But I want 10 different creative variations every single week being tested, basically.
Dylan:
If somebody's spending 10K a month, I would probably say two to four creatives a week to test out, because everything dies. It's like life. Your creative's going to die, so you always need to be testing. Depending on your budget, I always allocate 10% of our budget strictly for testing. That way, I'll have our winning ads, testing ads. I feed our testing ads into our winning ad campaigns, which will go more into the structure side. But it's just having a control versus a variable substance for the most part. So it's case by case there.
Dylan:
But what you need is not crap loads of content, but relevant content. I've gotten myself in trouble here before. It's where I was telling clients, "Send me as much content as possible. The more the merrier," when in reality, that wasn't really the case, to where I just had a crap load of content to really go through and it wasn't the most relevant. So I had to take a step back and rephrase everything I was pitching to clients more or less and be like, "Cool, we want a lot of constant, but it has to be relevant to the angle of the product and the audience themselves."
Dylan:
It can't just be, "Hey, we got 10 different images here. They're all the same. One's black and white, one's colored. One's got a red button. One's got a green button." I want different angles there in different actual creatives where it's not just a color change, and I want to see how it impacts there. That's probably the biggest piece of it, is actually properly testing.
Dylan:
I always do this down rabbit holes, but I love creative testing campaigns versus control campaigns. Our control campaigns, they have all of our winning creatives in there basically. Those take up probably 70, 80% of the budget because I have retargeting, creative testing as well. But I like to have two or three control campaigns, which I just scale up the budget with. I don't even touch the budget, basically.
Dylan:
Then for the other campaigns, it's simple there to where I'll have a campaign for each creative test. Maybe it's every single week, maybe it's every single day, to where we just do broad targeting tests, two different creatives per ad sets, maybe three, to gauge what works best there. After it's spent to statistically relevant amounts, it's gotten some sales, we'll then bring our creative testing ads into our control campaign, because the control campaign, it'll die off at some point, just because the ads get creative fatigue. So this will be swap it out there.
Dylan:
But it's very important to keep those separate, because there's no reason to... If it ain't broke, don't fix it. So I would say it's more of how you're structuring these creatives versus throwing an abundance out there and seeing what sticks, because another point of this is... And I had a call about this earlier, but with one of my clients, is we had two ads. We were testing them both. One was getting really good sales. It had like maybe 20 sales and a terrible click-through rate, a terrible cost per click, but it was profitable.
Dylan:
The other ad we had wasn't profitable, maybe only had three or four sales, but it had a way better click-through rate and cost per click. Now, in Facebook's eyes, it was pushing more budget to the crappy ad because it had better front end metrics when that wasn't the case. So you had to go micro -
Brett:
Because that's what Facebook looks at, right? They're measuring more of the engagement, depending on the goal you get of campaign. But-
Dylan:
Spot on. Sometimes Facebook's not always right, and I'm sure it's the same with Google and YouTube.
Brett:
Yep.
Dylan:
But it's good to have some spend thresholds and actual goals to say, "Hey, I'm not touching this until we get 10,000 impressions," to gauge if it's relevant or not. Eight to 10,000 impressions is our sweet spot to gauge, does this work or not, for the most part.
Brett:
You're running an ad almost regardless of performance in terms of generating sales to about eight to 10,000 impressions to really give it time to show what it's going to do.
Dylan:
Yeah. That can range anywhere from probably 250 bucks to 500 bucks per actual creative, to tests.
Brett:
Yeah. Yeah. Totally makes sense. Okay.
Dylan:
If it's a lower price product like for a cosmetic brand, AOV is 40, 50 bucks, that's perfect. Now, for the other brand I'm spilling right now, our CPA goal's $140, which is a lot more than hoping for a $20 CPA. For them, I'd simply give them a lot more time. I look at our creative differently versus impressions to gauge... Our goal is $140 CPA. I multiply that by three or four. And if we don't have a sale by that spend, I'll shut it off completely, because I know it's a little bit pricier. It's going to take a little bit longer. So I like to give it a little bit more lead time, I should say.
Brett:
Makes sense. We're talking about, and depending on budget, and it does make sense that the number of creatives you need is somewhat dependent on your budget because more impressions, the more you spend on an ad, that the quicker it's going to potentially burn out. How do you approach that though? How do you approach, man, that 10 new ideas a week, 10 brand new ideas, or is it 10 variations? When you work with a client, where do you start, and how do you set up this testing structure, and what are you thinking about in terms of the creative part of the ads?
Dylan:
Yeah. Luckily, I have a better team now to help me align this because it was getting messy with folders of hundreds of assets. I had my team helping me out to where he'll find five of his data creatives, he'll match those up. Then I have a copywriter who writes 10 pieces of copy every single day. They actually mix and match these together and said, "Hey, here's for this week, here's for this week." And that's been the biggest game changer because I hate saying it, but I mean, I didn't have accountability. So maybe a week, I didn't want to do it, unfortunately, but now I'm sticking to my guns there. So it helps us -
Brett:
you don't feel inspired or something, and so you just don't get the headlines written and... Yeah. Yeah. Makes-
Dylan:
Exactly. But you need to have some sort of framework. I don't keep track of tests in an Excel sheet, but that probably would be helpful. But all of our creatives are so different in different angles where I can gauge. But if I were to hand over all of our work to another agency, they would have no idea. But I'd be able to give them a layout of what's working right now. If I get a new brand or account, it depends if they're established or not.
Dylan:
If they're established, they'll typically have a creative team in-house where I can work with them hand in hand. Also, in them, I have a drive, about 500 ads that I've made with other brands and I'll be like, "Cool, here are the best converting ads. We can take any of these, replicate them, use our product, et cetera, et cetera." That's one good way, while also...
Brett:
Love that.
Dylan:
Oh yeah. If they're already doing good, I'll reuse their ads. I mean, there's no reason to reinvent the wheel. I'll get traction with their ads, better structure, get better profitability. And then I'd incorporate my ideas to where I'll be scrolling inspiration ideas. I'm never like, "Hey, we need three more ads that convert." I'll send them actual examples like, "Yo, I saw this one. Let's totally test this one out."
Dylan:
They really don't come up with too much, to where I usually provide the inspiration there to an extent. But they usually have some assets to get the ball rolling.
Brett:
Makes sense.
Dylan:
Now, if they're brand spanking new, that's a whole nother ball game. All refer them to Promo or Billo. It's where we can send a product, get a free influencer video for 60 bucks for something. I kind of-
Brett:
What were those resources you mentioned, and are those good for a brand that has a little traction too? Or are they just mainly good for startups?
Dylan:
I think they're both killer for either, to be honest, and they're really cost-effective. Promo is stock video footage. You can add overlay tag. It's pretty cheap too. I wish I used it more, but I just have a lifetime membership, so I'm stuck with it. That's a great option there. But then Billo, B-I-L-L-O, I think it's dot app, I would jump on this train. We're testing two of their creatives right now and it's working really well. The reason I say that is user-generated content, where it's at, it's different for YouTube probably-
Brett:
Sure.
Dylan:
But people love seeing a guy like you and I on an ad versus, shit, Tom Cruise. I mean, it's more relevant to them. User-generated content is where I think the platform's going. So you don't always need to have professional footage. I mean, on our podcast, we had Truff Hot Sauce. We had a 20K video versus a free video. The paid video, substantially when it comes to actual sales and whatnot. So you don't always need to have professional quality stuff, as long as it's relevant. You can just do iPhones-
Brett:
Billo is going to allow you get user-generated content?
Dylan:
Exactly. You send a product to somebody, they record it for 60 bucks, and boom, it's usable for ads.
Brett:
Nice. Yeah. We see, on the YouTube side, we talked about this one when you interviewed me, but you can still use user-generated content. You just can't use one 15-second UGC video by itself. You need to do a UGC mashup or a UGC mixed into a slightly higher production value video. So it does look different, but yeah, I love that resource. I'd never heard of it before. Yeah, we talk to clients all the time. You need good UGC.
Dylan:
It's nice. Yeah. Because, I mean, creatives, it's what moves the needle. I mean, a good way to put this, and I talked about this a little bit earlier on a separate podcast, but you don't always need new creative. It could be new angles, to where we have a sleep product. I think I may have talked about this on our podcast together, but what we're currently doing now is, hey, fall asleep faster. Boom, that angle's hidden. People love it. People have so much trouble falling asleep.
Dylan:
Now that's going to hit a cap. At 3K a day, we really can't scale past that. So what we did, exact same creative, but two new angles. Angle one is fall asleep and stay asleep and no more tossing and turning. That's pretty scalable too. Exact same audience, different subsection within the audience, essentially. The third angle is tired of getting up in the middle of night to go pee. That's a fricking huge one. So rather than us trying to get innovative and have different crazy creatives, the creative tests for that specific week, we're just testing different copy angles.
Dylan:
I'll be honest, when it comes to creative tests, I picture creative as more of the ad itself versus the static image or just video, to where it sometimes is the copy side of things. That's a great example of diversifying the different angles with the existing and creative we have that's already working pretty well.
Brett:
Yeah. And then, so first of all, love those headlines, those, really directed at a problem-solution type of structure. And so, that's just a proven winner as well. That's always worked in advertisement. Problem-solution has always worked. I love that, and this is something important to remember, is that sometimes you get these audiences that convert. But there are subsets of those audiences where one angle really works better than another. And if you're not testing and not properly testing, you don't know that.
Brett:
Sometimes you may have, oh, hey, the fall asleep faster like, "That's working. Let's just do that." Well, but if you hadn't tested the you're already tired of getting up in the middle of night to go the bathroom, then you wouldn't know that, hey, actually, there's a subset that that converts a lot better for. So yeah, really thoughtful.
Dylan:
Yeah. I mean, what's working best is broad targeting. I'm not going in for sleep interests or snoring interests. While those do work for quick wins, I don't think they're scalable and consistent, just depending on how Facebook looks at it or even we look at it. But I love having three or four active ads per audience, and I'll throw in different angles there because depending on who's clicking on the ad and converting, Facebook knows that.
Dylan:
I mean, somebody who clicks on the one from peeing in the middle of the night, to having tossing and turning, those are complete different people. So Facebook will find people similar to them based off those ads. So it's not only optimizing at the campaign ad set level. It's also the ad level, depending on who's clicking on it.
Brett:
Yep. That's awesome. Let's talk a little bit about campaign structure then. You alluded to a couple of things a minute ago, that I want to dive into. You talked about using 10% of your budget for testing and 60 to 70%, I think you said, to controls, and you've got remarketing in there. Break that down a little bit. I know every client's going to be a little bit different, and when you see results, you're going to push or shift or whatever. But what does that budget breakdown typically look like?
Dylan:
Yeah. Basically, if we're spending a thousand bucks a day, 30K a month, I would... Trying to think what those ratios are. It's a shitty one. We'll say a hundred K a month, basically, to make it easier. So if we're spending-
Brett:
Yeah. It's way easier to do percentages when you're based on a hundred of something. Yeah.
Dylan:
I know. I was like, "Why would I pick 30K?" But yeah. If we're spending 100K a month, I'll easily allocate 10K of that to testing, and those testing campaigns, those maybe a hundred bucks a day, and I can have multiples of those. So it's pretty snazzy there to where you have enough wiggle room to... Because if I'm spending a hundred bucks a day, that's 3K a month. I could have three or four hefty tests with that basically.
Dylan:
That's typically my sweet spot. Now, with the retargeting, I'll push 20% of my overall budget to that. We've got a hundred K. We took 10K for testing. Let's take another 20K out for retargeting. This depends on the conversion cycle, of course, and how much volume you are bringing for your colder audiences, because I do have one account where we only spend 10% for retargeting just because anything more will break it pretty much. So it's case by case. But I think a good rule of thumb is 20% of your overall spend. Then, of course, the other 70%, I would say all for cold audiences.
Brett:
Yeah. Very similar when we run YouTube as well. Then the re-marketing could vary if it's a long sales cycle versus couple of days sales cycle. Then those re-marketing percentages could shift a little bit. So yeah, that's really cool. We got the budget breakdown. Then what are you doing now in terms of campaigns structure? What would you add to that, now that we've got our budget parameters laid out?
Dylan:
Yeah. I typically have two or three control campaigns, and one will be... I'll put one in the ad set in it, to be honest. I'll test out one, which is wide open, maybe age breakdown. If it's no one buys under 25, I'll just make it 25, 65 plus. I give the algorithm full room to just do its thing. It's really just creative at that game. We just input every other week, depending on what hits for the creative tests. The control campaigns, I don't touch ever. The only thing I ever do is bump the budget up 20% or add in some fresh creative from our creative testing campaign that's proven itself.
Dylan:
With those control campaigns, they have really big audiences. As mentioned, we'll go broad with literally zero. So it'll be 70 million plus. We'll do 10% look-like audiences. Those are really, really fat. Those are like 20 or 30 million people. Then we'll use an interest-based campaign which could have one or maybe two or even three ad sets with interest stacked, based off what we're looking at. But those are all 30 million plus audience size as well.
Dylan:
So the bigger the audience size, the more consistent and scalable it is from what we've noticed. We don't even really touch those as mentioned, unless we're doing 20% budget increments daily, which I love to do daily. Anything more, it usually breaks it to an extent for a couple of minutes.
Brett:
So maybe scaling up budget as much as 20% a day. If you go faster than that, you're probably going to break something.
Dylan:
That's per campaign. If I have three top of funnel control campaigns at 500 bucks a day each, I could easily add on an extra 300 bucks that day. If you're doing that every single day, that you can scale pretty quick.
Brett:
Yeah, yeah, yeah. For sure. That compounds pretty fast.
Dylan:
Exactly. I mean, a good example. I had one account I launched last July. Month one, they spent 5K. Month two, they spent 10K. Then 15K, then 18, 24, 42K. We got the ad account shut down, only spent 40K the next month. But then this past February, spent 70K and then we're planning to spend 140K this month. That's how we're scaling it up. I can make those bigger increments hinges now that we're spending 70K plus a month. But it's a little bit more risky with the smaller spends because Facebook fluctuates so much.
Dylan:
That worked really well for us. So if you are looking to find a scalable model, and this is a subscription, by the way, so it's not like a three X or anything. It's like a 1.2X, just because -
Brett:
That's all you need when you've got long lifetime value and good recurring monthly revenue.
Dylan:
Exactly. That's how we're scaling up that account. The biggest way we're doing it is I'm straight up duplicating every campaign and just adding in new creative, basically. I'll just have creative tests week three one, and then that will just run for the week. Sometimes it works, sometimes it doesn't. If it's working, I'll scale that baby up. But I started out at a hundred bucks a day, and still go to 120, then 140, then 160, and then all the way to 200. Then I go to 240, 280, and vice versa there.
Dylan:
And so, you can't scale it up somewhat fast. But if I just get a little thought that this could be a really good one, I will start testing. But it's at 500 bucks a day, just to see what happens so it gets that data faster. Then I'll shut it off in two days if it doesn't work. So it depends on how much wiggle room I really have.
Brett:
Sure. Totally makes sense. When you encounter a prospect or a client that's plateaued or stalled a little bit, what are some of the common mistakes, some of the common reasons that you find that, hey, these are the reasons why this account is not scaling like it should be?
Dylan:
I would say retargeting and exclusions. The retargeting should always do way better than your top of funnel. So, I mean, whenever I see a top-of-funnel campaign getting a four or five X, I'm like, "They're not excluding." I can almost guarantee that.
Brett:
Yeah. The funnel campaign is actually there's a section of it that's remarketing.
Dylan:
I see that a lot.
Brett:
I do too on the YouTube side. Yeah.
Dylan:
That's a big one. I think even ad formats. The amount of accounts I audit to where they're only using carousels, only using videos or only static images, I'm like, "You're throwing all your eggs in one basket, praying for the best," to where I could have the same copy, which I love to do, and I'll have that same copy with a video static and a carousel ad. But these different ad formats have different placements in the ad auction. So sometimes adding in a different creative format could lower your CPA like 10%. It's fricking bananas.
Brett:
Really? Do you have any rules of thumb for, "Okay. We're running this percentage of carousel ads, or this percentage of static ads, this percentage of video," or it's just all dependent on what's working?
Dylan:
It's depending on what's working, I would say 5% carousel ads though, 30% video, and the remainder being static ads, because I'm supposed to see-
Brett:
Yeah. I was going to see, and I'm not on Facebook, a ton. Not always a fun place to hang out these days. Great place for advertisers though. Static... I mean, I'm sorry, carousel ads, I don't see them that much anymore, like re-marketing mainly. But not beyond that necessarily. Okay. Super interesting.
Dylan:
Good little case study for that is I literally audited somebody last week and all they're running are carousel ads for their food. I just straight up said, "Make a slideshow using the same images," and that lowered their CPA 15%. And it took two days. So that was a quick win. Didn't change anything. Just took those images, turn into slideshow, and boom. Now it's converting like crazy.
Brett:
Nice. So slideshow. That's the type of thing that doesn't work on YouTube at all. But so, when you do that, are you using something like Animoto or another tool like that, just create a slideshow or...
Dylan:
I don't do it very often because I don't like the Facebook creative tools. But yeah, you would just select the images and you can make a slideshow with it, within Facebook natively. So they didn't even have an editor or anything, and it's like, "Hey, just take those five images in the carousel and put them in here." It did it for them basically.
Brett:
Nice, nice. Very interesting.
Dylan:
It's cool. But ad formats, exclusions, and I wouldn't even say too many campaigns.
Brett:
Okay.
Dylan:
Facebook pushes 50 conversions a week. I think that's blasphemy. I don't believe it personally. But I do think you can spread yourself too thin to where those cosmetic-
Brett:
Facebook, yeah. Can you explain that really quickly? They push 50 conversions a week. What do you mean by that?
Dylan:
50 sales. If you have a campaign, if want it to get-
Brett:
At the campaign level, so a campaign needs to be doing 50 sales a week?
Dylan:
Exactly, when using campaign budget optimization. That's when you set your budget at the campaign level. But if you're setting it at the ad set level, the ads, it has to get 50 conversions a week. A conversion can be anything in Facebook size, depending on what you're optimizing for. But of course purchases where it's at.
Brett:
Interrupt and talk about that for a minute. So 50 conversions, are you saying you actually think it could be less than 50 and does well or ... Okay. Got it. Yeah.
Dylan:
Yeah. We have like $20,000 products and we only get like three or four of those a month.
Brett:
Yeah.
Dylan:
I mean, the learning phase is a myth, in my opinion. But with those 50 conversions, where I was going with that, is if you have 20 campaigns and they're all getting five conversions, it's going to take forever to get out of that learning phase and longer to optimize, which isn't that big of a deal. But if you ended up having three campaigns instead of 50, where they had 50 conversions in each one, you're going to get a way better bang for your buck. Facebook will optimize faster.
Dylan:
It's basically, if I had some campaigns that's 10 bucks a day, or I could have two campaigns at 50 bucks a day. The 50 bucks a day ones are going to do way better any day of the week, just because they're getting data faster. You can optimize faster, make decisions faster, versus the others.
Brett:
Yeah. I agree. Then sometimes having fewer campaigns and you're really pushing and scaling, it's just easier to manage and easier to pull levers on. I think there's a lot of benefits that go into that. That's awesome. All right. Let's pivot a little bit. Let's talk about everybody's favorite topic, or depending on when you're listening to this, it maybe the worst topic that you could ever have imagined, or it could be like, "Oh, this is no big deal." I don't know. Let's talk about iOS 14. What are you telling clients now related to iOS 14? How are you preparing? Let's walk through that a little bit.
Dylan:
I changed up all my pricing because of that, basically. What we're seeing on Facebook, I'm only seeing this in one account right now, but I mean, we're getting a hundred sales a day, and we used to show 85 to 90 sales attributed to Facebook, but now we're only showing 45 to 50, if 50. So it's dropped about 40, 50% in tracking for that specific client when everything is business as usual. So I don't really have an explanation for that except this, but there's no way to gauge if that's it or not. So it makes it tricky.
Dylan:
But what we're telling all of our clients is, hey, we're going to be looking at Google Analytics a lot more now, to look at seven-day windows and day-to-day spend in revenue versus-
Brett:
Are you guys using UTMs to track, or you're just not going to-
Dylan:
Exactly.
Brett:
Okay. Yeah.
Dylan:
We went way harder with UTMs where we implemented it for everybody, basically, just because Facebook's not showing the true monetary value of most things. So rather than even doing reporting in Facebook, we do it all through Google Analytics. The other part of this is we're probably going to see less results than we typically do because Facebook looks at 28-day click one day view versus seven-day click one day view.
Dylan:
Now, that's not much. It could result in a few extra sales. A lot of people don't convert too quick. To combat this, shouts out to Nick Shackelford, this is two for the day for you.
Brett:
Yeah. That guy's awesome, man. He and I have spoken at the same event, some events. Nick Shackelford, real deal. I believe, I believe a good soccer player too. So anyway, for what that's worth.
Dylan:
He's a goalie, I think, because I'm a goalie, and I think-
Brett:
Oh, is that right? Okay. Okay. Oh yeah, yeah. But you're a hockey goalie. I'm getting totally sidetracked here. My daughter's playing lacrosse. None of these sports I've ever played, but lacrosse, soccer, hockey. I was a basketball guy. I always had that. But fascinated by these other sports that I have less experience with.
Dylan:
They're fun. But yeah. Nick Shackelford, I'm meeting him in a couple of weeks coming to Austin for Geek Out, shouts out, Geek Out. But yeah, he created a 28-day click multiplier. Essentially, I would go in through all of our ad accounts, look at six to eight months of data, gauge what's our revenue on a one-day click window, seven-day click, 28-day click? And that's for cold and warm audiences. So you're basically looking at past data to gauge, if I had a 2X on a seven-day click window, what would I be on a 28-day click?
Dylan:
You would have to multiply this by 1.23, for example. So we have all those figures of what it could be. But what I've actually found is that the last year has just been a fricking mess, so the COVID, the elections, the freeze that hit Texas. I noticed sales were down 40 to 60% in Texas, and some of the weeks.
Brett:
Yeah, yeah, yeah, yeah. No away around that.
Dylan:
So there are a lot of other variables involved to make it trickier to gauge if that's really the case there. But the 28-day click multiplier, that's one part of it. But I think the biggest thing is we're looking at more blended metrics, ecosystem metrics, with UTM codes, to gauge, hey, Snapchat's doing this, Facebook's doing this. There's probably going to be some overlap here, but not too much.
Dylan:
So we just changed our whole reporting model to stick with percent of ad spend versus return on ad spend and CPA because we just don't have that info like we used to. I'll probably go to it in the future once all the kinks get worked out. But we haven't seen much change on our end. I usually always look at stuff on a one-day click and seven-day click window. I mean, for those accounts that have that, nothing's changed pretty much.
Dylan:
I think the hardest part is for how I'm relaying it to clients, is having to have them get their domain verified, set up specific pixel events and stuff to where it's a pain in the ass. So it needs to be done in order for the Facebook pixel to attract this stuff. But Facebook didn't make it easy. I think the hardest part of all of this is the grunt work I have to deliver to my clients to be like, "Hey, you just put this [inaudible 00:35:50] tag code in your website somewhere. I don't know where, but it's supposed to go there."
Dylan:
I think the implementation side is probably the hardest part, but we were just proactive. We had a lot of resources involved and had calls with all of our clients to give them expectations, to play it safe to say we're proactive. But at the end of the day, the big stuff hasn't even been released yet. I mean, I think it's just the beginning.
Brett:
It totally is. And then, we're doing a similar thing. We're sending out news information from Google to our clients and helping them get ready. We're not seeing any major changes yet, but I really like a few things you talked about there, one, blended metrics. We're going to have to get more creative about how we track and attribute and how we look at overall sales lift because, yeah, there's going to be some cases where Facebook maybe isn't given all the credit that it deserves, or YouTube is not given all the credit it deserves, but we need to be able to get some blended metrics to get a better idea.
Brett:
I love that 28-day click multiplier. We do something similar, but reverse where we're looking at like, "Hey, based on a one-day CPA, here's what we expect that CPA to be in seven days or in 30 days." So it's similar thing like, "Hey, if the CPA today is a hundred, well, that's cool because we're expecting it to be 50 by the time we get to 14, 15 days," based on what the historical data and stuff.
Dylan:
The other cool part with this is I have access to every channels marketing efforts now. So I get all the reports from the Google Teams. If I start noticing some discrepancies, I don't like throwing people under the bus. But when someone's not transparent and unethical about it, I have no problem throwing them under the bus there to where... I've had it happen twice where they were totally looking at the wrong KPIs, and turns out, they burned over 70K in cash and thought they were profitable. They weren't.
Dylan:
I was like, "Dude, they have been giving you the complete wrong figures. Let me do this for you. I'll take it over." It got me some clients. It's helped out in a way there, but I like seeing other people's marketing efforts. I mean-
Brett:
Yep. Yep.
Dylan:
It's interesting.
Brett:
Sure. Yeah. The blended metrics and digging in and yessing all the channels tied together, that's super important. I think it's actually going to reinforce one of the first points you made, which is the ad is everything. As privacy issues become tighter and there's less transparency or less ability to track, the ads are just going to become more and more important, like what you say, how you say it, how you deliver that ad is going to be more and more important. So get it right now. That's for sure.
Dylan:
Oh yeah. Something I spoke about in another podcast is I didn't know this would affect our ads, but what words you're using. If you're using, "Hey, buy this today," you could be like, "Hey, grab goodies today," or something. So specific-
Brett:
"Buy this today," sounds salesy and pushy, but "Get your goodies today," sounds fun.
Dylan:
Exactly.
Brett:
Like playing around with that. Yeah
Dylan:
But yeah, it's an auction. If a million ads with "Buy this today" versus "Get your goodies today," you could convert better at "Get your goodies today". What I've noticed is whenever we're using different lingo, our CPMs are about 10 to 15% cheaper, which on a massive scale, that's fricking huge.
Brett:
That totally impacts the rest of the math down the line. So if you're worried about RoAS and CPA, which you should, impacting your CPM can definitely impact all the rest of it because that's the leading metric there, leading versus lagging. Awesome. Dylan, this has been fantastic. We are officially running up against time here, so want to wrap up a little bit.
Brett:
But two things. One, plug the podcast. Why do you enjoy doing the podcast? If you have any recent takeaways from the podcast, and we've already shared some, so no worries if nothing else. But talk about the podcast a little bit. And then, I want you to tell people about your company and how they can learn more about you there.
Dylan:
Yeah. For the podcast, Rich Ad Poor Ad, check it out. I think you just go to Google and type in Rich Ad Poor Ad. It's all under FunnelDashes network, essentially. We have a lot of great guests who spend 50K a day, a dollar a day, organic affiliate marketers. We have brand owners talk about their brand directly, agency owners, media buyers. We have literally a bit of everything on there and it's not from beginners, Billie, Jean, Neil Patel, for, again, Brett Curry.
Dylan:
We have some legends on our podcast. I mean, there are a ton of actionable things, because we talk about what's working, what's something that crashed and burned, and some financial tip. So it's 20 or 30 minutes of good, valuable golden nuggets to be dropped. As mentioned, I do it just to network. We're probably going to monetize it. I didn't think I would ever say that, but I'm doing it strictly meet people like Brett and just level up my personal game as well.
Brett:
Sweet. Love it. Love it. So yeah, check it out, Rich Ad Poor Ad. Then your company, Koality Media, how can people learn more about your company? Any resources? Anything you want to plug there?
Dylan:
I would say find me on Twitter at marketerdylan or Facebook, you're going to search Dylan Carpenter. The website is called Koality Media, K-O-A-L-I-T-Y, media.co. I had an intern a free three years ago. I don't use it, but there may be some goodies on there.
Brett:
Hey, that's often the sign of the like, hey, you're just a practitioner. You're just doing it, man. You're running the ads. You don't care about your own website. You're out there making it happen.
Dylan:
Exactly. Yeah, if you have questions, I'm an open book. Ask me anything, but not too crazy, hopefully. But yeah. Find me on Twitter or Facebook and I love having conversations.
Brett:
Sweet. All right, man. Dylan Carpenter, ladies and gentlemen, rocking it. Thanks for taking the time. This has been a lot of fun. Really excited to see where everything goes with iOS 14 and all these changes. But you dropped some amazing golden nuggets that are going to help people no matter what happens here in the future. So really appreciate it, man.
Dylan:
Nice. Appreciate it, man. I loved it. Thanks for having me on.
Brett:
Absolutely. As always, we appreciate you tuning in. Hey, we'd love your feedback. Let us know what you'd like to hear more of, what you'd like to hear less of. If you have not done so, and you feel so inclined, we'd love that review on iTunes. With that, until next time, thank you for listening.
.png)
.png)
