Hosted By

Rabah Rahil
CMO at Triple Whale

Guests

Ash Melwani
Co-Founder & CMO of MyObvi
Cody Plofker
Director of E-Commerce Jones Road Beauty

How To Optimize Your Ads Post iOS 14

In this episode of adspend we go over how to navigate and optimize paid media and ads in a post iOS 14 landscape. Ā #Adspend

Notes & Links

šŸ“§Subscribe to Whale Mail for exclusive industry insights and in-depth marketing breakdowns: https://www.getrevue.co/profile/tripl...šŸ¦

Follow us on Twitter for Industry insights https://twitter.com/triplewhale

ā€Follow the people featured in this episode here:

- Rabah's Twitter: https://twitter.com/rabahrahil
- Cody's Twitter: https://twitter.com/codyplof
- Ash's Twitter: https://twitter.com/ashvinmelwani

Transcription

Ash Melwani (00:00):

I would rather like I would, I'd sell the business and I'd be dope. Like there's no way that I'm gonna do this every week. I need to give Facebook 40 creatives. Let it decide if it decides this is a winner. Great. Put it in scaling. Like you just have to make it easier on yourself. Like you can't be like so scientific and like, well, like I gotta have statistical significance on like this adset versus like in like, just make it simple.

Rabah Rahil (00:35):

All right, boys, we're back for number six. I'm a little sunburnt. I'm peeling, like crazy, almost as tan as Ash, but Hey, Hey, I'm keeping up. I'm still beating Cody. So here we are. We're back for another episode of ad spend. Cody is back. There might be some feistiness. He has overcame the Ronna, like a legend. He has a new house on the way. He also has another thing on the way a baby Cody.

Cody Plofker (00:58):

Yeah,

Rabah Rahil (00:58):

Right? Yeah. So how exciting is that

Cody Plofker (01:01):

A girl though? So not

Rabah Rahil (01:03):

Me. Yeah, but even even more exciting. So congrats there. Ashe is looking fresh in his, uh, fancy hip hoodie as always. Um, and so I think today we're gonna get into a little bit of how to optimize post iOS 14. Um, I'll drop some knowledge, Ko drops. The knowledge Ash will drops. The knowledge will kind of circle it in a, a VA and then give it to you guys in a beautiful podcast format. So without further ado, Cody take us away.

Cody Plofker (01:28):

Yeah. Dope. Uh, this was kind of my idea what I wanted to chat about. Cause it's definitely something I have been struggling with. I am no expert at it. I don't know if anybody is an expert at it, uh, to be totally honest. So I just would love to chat about it cause I'm sure there's a lot of people out there who are struggling with this who are frustrated. It's no surprise that Facebook attribution is absolutely terrible. I mean, I, I tweeted the other day, but like I was just looking at our, our, uh, our ex you know, our exclusions. And I was looking at our, our Clavio 180 day purchasers compared to pixel 180 day purchasers. And the pixel is like a third of the size. Yeah. Like it is crazy how bad attribution is. And, and in, in both ways, in terms of over reporting and under reporting. So I think you absolutely need first party data. You need to use Mer, but I think a lot of people say that and you know, there's only so much you can say on Twitter. So I, I figured let's just talk about it and Ash, and I will share what we're actually doing to optimize our ads. And then Rob can obviously share everything that he's seeing being inside of triple oil.

Rabah Rahil (02:34):

You're getting good at this Cody. That was pretty smooth. I just kind of tossed it over to you. Yeah, I was. So once you get a studio, it's over, it's over people. He's hot.

Cody Plofker (02:42):

He's hot. So Ash, would you talked, obviously you dropped some heat, you know, two pods ago about your creative testing, obviously with that much creative testing you're doing, you must be looking at, what's working a lot of optimization. What's your process for that? How much using Facebook, how much you GA how much using triple oil?

Ash Melwani (03:01):

Yeah. So I think one of the biggest things that I've learned to stop doing is looking at Facebook as like your, your north star. Right. Um, I've kind of done the test with like all the different attribution windows, and I've seen over reporting under reporting. Um, some situations where the reporting is kind of matched up to, you know, these other things like GA and triple, um, I've kind of had to really just pick one and stick with it and see if the overall kind of starts to follow. Right. So what I've been doing recently is literally only looking at the data from triple whale pixel, right. Um, so, you know, two pods ago, I mentioned, you know, the whole creative testing, you know, we launched everything on a Sunday, uh, at midnight and I let it run for seven days. Right. So basically no

Cody Plofker (03:54):

Matter just, just for people here and no matter what, even no matter what terrible, you're letting it run.

Ash Melwani (04:01):

Like, so actually it's crazy. Cuz last week I had a really bad test that just wasn't performing. So I did cut it like maybe two, three days earlier than I usually would have. Right. And it, it, that's just how it's like, you might have a really good test. You might have an okay test and you might just have a shit test and that's it. Right. Um, but the way that I'm deciding that is I'm looking strictly at triple oil, right. So for example, yesterday I even tweeted or no, sorry, Sunday, right? Uh, day one of testing, uh, for the week thousand dollars budget. Um, and Facebook said I had $183 CPA. Okay. Um, obviously, you know, you're not gonna get data for 72 hours, so you got, you gotta let it run this and that. But like, I don't want to rely on waiting 72 hours and I'm sure there's other advertisers who don't have that budget to let something run for 72 hours at like a thousand dollars a day.

Ash Melwani (04:55):

Right. So you need to know what direction these ads are going in as soon as possible, just so that, you know, and it depends on the appetizer and your budgets. Right. So looking at triple whale, at least for my testing, when I'm testing top of the funnel and RA you can kind of get into this later, but I'm really looking at first click attribution. Right? Yeah. I want to know which of my ads created that awareness and eventually led to a conversion. Right. I wanna know what brought somebody into my ecosystem and later on converted them. Right.

Cody Plofker (05:26):

That's cause you don't, you want them to kind of take a lot of touch points and buy later on. Right.

Ash Melwani (05:32):

<laugh> I

Rabah Rahil (05:32):

Prefer, I told you he was back folks. I tried to tell you he was back.

Ash Melwani (05:38):

Um, no, but I mean, I'm optimizing, so at least my tests are on a seven day click. Right. But I'm really, I really want somebody to click or I want, I wanna know which ad introduce them to the brand. Right. For sure. Whether they go down the funnel,

Cody Plofker (05:52):

Obviously I'm just joking. Yeah. No, but I'm, I'm surprised that you're using first click click. That's your main, main north car. Yeah.

Ash Melwani (05:58):

Yeah. And what's what I have been seeing is that over since January, until now are because of now consistently done this process from January until now the testing creatives every week, following triple well, using that first click attribution as the north, I've seen our NCPA drop. Like sounds good. If you go and pull up our, our monthly it's literally like this. So now I have reason to believe that is just the way that's, it's gonna work for us on an overall level. So I'm gonna continue to do that. Right. Um, so basically, you know, I let it run for a week. I'm looking at, you know, I'm monitoring first click, you know, which has the lowest N CPA of all the ads, which one's getting the most spend. Um, and then I'm really just taking those ads and moving it into a scaling into my scaling campaign. Right. If that ad is beating the lowest performing ad in my scaling campaign, I'll just replace it. Right. Kind of recess the learning. And then the next week we'll see how it performs with a whole, technically a new environment.

Cody Plofker (07:01):

Right. And so you're only doing that once a week. Like you're only moving something into scaling once a week at the very most

Ash Melwani (07:08):

Mm-hmm <affirmative> otherwise once a week or maybe every two weeks, depending on if, if the test is beating my control. Right. That, that, that would be the only reason to move something over. Yeah. Or if my control is starting to perform badly,

Cody Plofker (07:23):

What about, so what, what are you doing if you're halfway through the week and you're, your results are terrible. Are you doing anything to that scaling campaign? Are you touching it or you're just letting it ride out. You're just focusing on creative testing and you fix it next week. What, what are you do?

Ash Melwani (07:36):

So here's where I think we're in a good place because now I'm constantly pumping out creatives. I don't have to worry about the weekly performance because I will have new stuff to replace it. So my weeks are a lot less volatile cuz they're technically newer creatives that have been proven. Right. Whereas before, if I wasn't swapping out creatives, as often, this stuff would probably be running for like months on end. And if I have a shit week it's because it's probably hitting that fatigue level. Right. So

Cody Plofker (08:09):

Then, but you're being proactive because you're being, because you're being so proactive about it. You're not really fine in that like steep drop off.

Ash Melwani (08:17):

Exactly. The only time we ever actually see anything is God knows what reason Mondays are our worst days. Like you'll have a Sunday, that's just a monster day. And then the next day, as soon as 12:00 AM it it's like conversion rate drops, NCPA increases. And like Monday is just the worst day and Tuesday it gets back to normal and Wednesday gets better. Thursday gets better. Friday gets better. And like, it's the weirdest cycle that I've seen. And I have no idea, but like that, I just know that Mondays aren't the best day. And I have to just let things run for the rest of the week. And it's like, okay, well at least I know that I'm doing everything I can on the creative side that I'm always gonna swap out new creatives for new winners and this and that. So it can't be creative. That's the issue. It's just, I don't know. It could be Facebook thing. It's just, maybe it's our demographic that just doesn't buy abundant. I don't know. I just, I, I can't believe, or I'm not led to believe it's like a creative issue, you know what I mean?

Rabah Rahil (09:13):

Okay. Okay. I have a thesis there. If I can interject mm-hmm <affirmative> cause our, our Mondays are horrible too. I mean, we're obviously B2B SAS, so it's not necessarily alignment, B2C or DDC. Um, but I actually think, and just anecdotally, like my Mondays are hell, I actually think people work on Mondays and that's my thesis. Like nobody shops on Mondays. That's like the one Workday, even if you're not aware, like workhorse, you actually do work on Monday. That's that's my, my big brain thesis. I think

Cody Plofker (09:40):

So. I mean, that would make sense.

Ash Melwani (09:42):

Yeah. Yeah. No, I, I, it's just been very noticeable for the last two years and like, if we over at the office, we're like, oh, today's slow. Oh, it's Monday, it's a Monday. You know what I mean? It's just, it is what it is. You just can't do anything about it. You just gotta let it run, you know? Cause if you make decisions on based on one day and that too, being a day that, you know, historically doesn't do well, you're gonna screw your shit up. So like keep it going, you know, just have, have trust in yourself that you're doing what you can do to kind of keep things less bottle tile. Okay.

Cody Plofker (10:12):

Okay.

Rabah Rahil (10:13):

Can I ask you something or, or go ahead, Cody. No, no, no. You go, have you tried, um, using like, uh, cost caps or bid caps where you keep the budget, but then you just modulate those cost caps and bid caps where like on Mondays, for example, you would, um, put the cost caps or bid caps, um, basically cost controls, lower those. So they'll still play if you, if the market's good, but if not, they'll kind of save you some money. And then on Tuesday through Friday, what have you, you just skyrocket those cost caps to meet, you know, maybe three, four X or C target CPA. And then cuz cuz you won't mess with the algorithm in that much or in that way. You're really not challenging the algorithm as much as you are because those budgets are already kind of hard coded. You're just, um, telling the stock broker essentially don't be so drunk here is only what you can spend.

Ash Melwani (11:02):

I've I think Septe last September till around December, I was fully all in on class caps, right? Yeah. Um, after that in January, what I was like here is my process. Here's what I'm gonna do. And I'm not gonna deviate from that. Right. I have like simplified our entire media grind process. Like I don't wanna mess around with cost caps. I don't wanna mess around the bids. I don't wanna have to go into the ad account more than once a week. Right. Wow. All I wanna do is I just wanna see if my NCPA is lining up with our KPIs. Right. For example, last, uh, last Monday and the Monday before, right? Our NCPA was above average for a Monday, right? Like our NCPA goal is 55. Okay. Our NCPA last Monday was $70 the week before last Mon that previous Monday was $70. And then throughout the week it lowers.

Ash Melwani (11:53):

Right. But this was the first Monday. We were actually at 55, which means it wasn't the worst Monday. But because we're, we're still making all these tweets and we introduced TikTok, right. We're scaling TikTok and those CPAs are so much lower than Facebook right now. It's bringing our overall average down. So now it's like, okay, love it. I'm keeping my spend where Facebook is right now. And I'm going so hard on TikTok so that it's bringing everything down and now I just gotta like rinse repeat. So that's where I'm like, okay, well this Monday could be worse. Right? Like a week ago I was $15 higher. Now I'm like at my goal and now throughout the rest of the week, I'm actually under my goal. So that's kind of like, you just gotta keep working at it and like nitpicking at it, trying to keep bringing it down as much as you can. I mean, set yourself a goal and then just work towards that, you know, just get a little bit better every week.

Rabah Rahil (12:43):

I love that.

Cody Plofker (12:44):

So I just wanna get back on track a little bit with like the optimization and like how you're looking at things. So data delays, right? You're not even looking at Facebook triple well, you're able to get some data back sooner cuz I, it doesn't have to get delayed pushing back. You're looking at, um, there still are going to be delays though, because obviously on the first day it's not gonna cover all the conversions that happen. If a lot of them are happening outside of the first day. Are you looking at any, do you like to look at, uh, any leading indicators? Do you look like, look at any on platform AME? Do you like to look at any cost per add cart or you're just CPA,

Ash Melwani (13:18):

Just CPA, at least for Facebook mm-hmm <affirmative> with TikTok. I think there has been more of a correlation in terms with like your CCPC and CTR to CPA versus those metrics compared to a CPA based on Facebook. And that again that's

Cody Plofker (13:37):

And I, I think that's where we're different. There's no right or wrong, but like I'm heavily into those, those metrics,

Rabah Rahil (13:42):

Software metrics

Cody Plofker (13:43):

Doing a ton of ADA analysis, like thumb stop. Uh, we recently did a correlation analysis and found that like three second views divided by clicks. I think what is, what it was, was like the highest correlation to CPA for us. So like if people have watched for three seconds, what percentage of those people clicked? And that had a pretty high correlation, much higher than thumbs stop, much higher than click through rate. So that's kind of a big one that, that we've been looking at. Um, I don't know what call it like a effectiveness.

Rabah Rahil (14:12):

So it's how is that different than a click rate though? Like it can't be right? Yeah. Function the same, same bit different. Yeah. It's a function of click rate. It's so interesting. I'm not saying you're wrong. I'm just kind, yeah.

Cody Plofker (14:23):

I don't know. I don't know if I totally understand it. Yeah.

Rabah Rahil (14:26):

When I think of thumb stop ratio, I think of a pattern interrupt where it's like, you're literally interrupting a pattern like the scroll. Well

Ash Melwani (14:32):

What's your formula for a thumb stop ratio.

Rabah Rahil (14:34):

It's sorry. Yeah. We should always be clear about

Cody Plofker (14:37):

That three second views divided by

Rabah Rahil (14:38):

Impressions, right? Yeah, exactly. I impressed. So that way I know I serve this person to add, but then they stop for at least three seconds. So that, so the difference makes sense to me.

Cody Plofker (14:47):

Clickthrough rates is from impressions. It's, you know, it's clicks, divided by impressions. This is right. Would be, I forget if it's clicks first or impressions first, it's always mess it up. But this one would be clicks divided by three second views. So it's kind of like once you got somebody, how likely are they to click? The other one is just how likely are they to click if you saw the ad? Yeah. So because essentially click through rate and I don't, I don't know if I totally understand this completely, you know, to answer your question, but click the rate. You're also actually really measuring your thumb stop cuz it's it's your, your measure, your number you're dividing by is impressions. So it's not just desire. You know, a lot of people use click through rate as like the, the D and ADA, but it's got just that cuz an element of it is also the stop and this is just how effective is your ad once you get them

Rabah Rahil (15:32):

To be fair. Let me just push back a little bit. Yeah. I can click without interacting with the ad. Whereas a thumb stop ratio is literally I have to consume for three seconds before it would trigger that thumb stop, but I take what you're saying. I'm just kind of being pedantic here where you can get shown to ad and I go, boom. And I slap it. The other thing is, uh, click rate is applicable across all creative types. Whereas thumb stop ratio has to be exactly

Cody Plofker (15:58):

It's

Rabah Rahil (15:58):

That's yeah. That's

Cody Plofker (15:59):

And that's probably

Rabah Rahil (16:00):

Two videos.

Cody Plofker (16:00):

We do a ton of video. We do almost no static and ASTA is a ton of static. So like it's a lot harder. Um, just to, uh, throw a disclaimer out there. Like those are, I guess leading indicator is two CPA CPA. Yep. I don't optimize off of those. If our CPA is bad, I don't want a good click the rate with a bad CPA. I think I just have to throw it out there. Cause if you talk about click through rate on Twitter, if you talk about thumb stop or anything else, people are like, don't optimize off those. Like those are garbage metrics like that. Just click bait. And it's like, it's, it's a lot more nuanced than that. You know what I mean? Mm-hmm <affirmative> like, it's really hard to have a really good CPA if you're getting an $8 click, you know what I mean hundred percent. So like it does matter to a certain degree and when you're getting that data a so not reliable, but also so delayed from Facebook's end. It is kind of nice to know if you're on the right track. So I will definitely kill something early. And I, if those AME are crap, like I just know if we have a 15% thumb stop or a 0.5% click through rate. Like no matter what the CPA's not gonna be good, but you

Ash Melwani (17:03):

A little more CPA first.

Cody Plofker (17:06):

Oh totally. Definitely. Yeah. Yeah. So it's like a order

Rabah Rahil (17:08):

Operations.

Cody Plofker (17:09):

Exactly. Totally. And then those are, to me, those are numbers that I might look at to try to help me improve CPA. I'm not gonna look at those in absence of CPA. I'm not gonna look at it in spite of it, but I'll look at those numbers to try to help me improve CPA. And usually as long as you're doing it, not click baby, getting those numbers better improves your cost per click and then it, it can kind of make sure you're having qualified traffic again with the caveat of it's not clip Baie or anything like that.

Ash Melwani (17:37):

Got it. So let me just clarify. So you did a correlation test based on, I guess a ratio. What are you pointing? This ratio? What do you wanna call

Rabah Rahil (17:46):

It? Yeah, this, the code metric

Cody Plofker (17:48):

Effectiveness is what is what I put, I don't know. Add effectiveness. That's

Ash Melwani (17:52):

Let's call let's

Cody Plofker (17:53):

Call.

Ash Melwani (17:54):

No, it's gotta be better. A hundred percent. It

Cody Plofker (17:55):

Has to be better. Um, you guys name it? Not me. The, the plot, the Plager nah, no, no. You don't wanna name it. Come on. Nah, no, one's gonna be able to say it right. That's

Ash Melwani (18:06):

True. Okay. So let's let's just, what is it? It's um, okay, so clicks divided by three second views. Okay. Did you run a correlation test? So you did a correlation test between that metric and CPA versus did you do the same correlation test with clicks divided by impressions

Cody Plofker (18:23):

Or sorry.

Ash Melwani (18:23):

So we what's that thumb stop ratio next

Cody Plofker (18:24):

Three seconds divided

Ash Melwani (18:25):

By impressions.

Cody Plofker (18:26):

Yep. Yeah. So we did it essentially. We did it on all those is we compared all of them to CPA. Okay. And found their

Ash Melwani (18:33):

And what did those look like compared to

Cody Plofker (18:36):

There were what you were saying? So nothing was statistically significant, right? Nothing is, you know, above, uh, 95%. Right? Sure. Statistical significance. However, there were some, there were some decent correlations, but the strongest one that we found was maybe 75% correlated. I don't remember exactly what was, was that one.

Ash Melwani (18:57):

Okay. Interesting. I, I fully believe it, man. I mean like, that's cool if you, if, if you have somebody that is kind of clicking, cause they've been, you know, they've related to that hook, right. You're pulling in for a certain reason. Right. And then they're probably more likely to buy X, whatever it is. Right. I think the same thing has happened for us. I know we're doing static and you're doing video, but like our static, our creatives are really nailing that singular angle. And, and we're just trying to iterate those angles because that is probably the reason people are coming in. Right. Where like, you know, I mentioned this two pods ago, it's like before I used to be, this is great for this, this and this. And then when we kind of just went, this is perfect for this, we got a little bit more qualified audience and we kind of tailored everything to that and conversion increased and our CPAN dropped a little bit. So I totally agree that like that hope that three seconds is like super crucial because it could just make or break and it could attract a different audience too. Mm-hmm <affirmative> so it's like super crucial to definitely measure that. If, if CPAs or like you're saying the delayed data is, is not coming in, that I think is, is definitely a good leading

Cody Plofker (20:06):

Computer. It'll just gimme, it'll give a little bit a it's that's on the creative analysis side. So then you look at all right. You know, Hey, this ad did. Okay. But let me, you know, click through rate was good. Average watch time was pretty good. So it was, it was engaging, right? Like my messaging is decent, but you know, conversion, rate's not there. A lot of people would look at that and be like, ah, it's a landing page issue. I don't necessarily think that's always true. I think a lot of the, the conversion is more of a persuasion thing. And it's what you're saying or not saying to sell them in the, the video as well. But again, you just look at those metrics and be like, all right, now, if we get, you know, we keep everything the same, but we get a better thumb stop.

Cody Plofker (20:40):

How does everything else improve? And theoretically it should. So a, that the iterative process is, is one reason I look at those metrics, but, but the other reason is again, if we have some correlations, they're very soft correlations, not hard, but it'll give me a little bit more confidence to let those ads continue to run with the bad CPA. Because not that I want ads to run with bad CPA, but because I know that that's very delayed data and there's a chance our CPA gets a lot better in the platform because I know that these are good ads from an engagement perspective.

Rabah Rahil (21:12):

Yeah. That's

Ash Melwani (21:13):

Fascinating. Now,

Rabah Rahil (21:15):

Do you, um, do you ever run, uh, and I'm gonna get roasted on Twitter, but uh, a guy named Jordan is actually, he's a killer and he's saying he's finding some success, some success with this. Um, are you running any OG video view campaigns? Like people have watched over X percentage of these retargeting?

Cody Plofker (21:34):

No. So you're saying like top of funnel, video views, or are you saying like no

Rabah Rahil (21:37):

Video? Yeah. So people that have watched 50% more of the videos, 75% more video or a hundred percent of the video, are you retargeting them at all?

Cody Plofker (21:45):

Yeah. Yeah. I usually like to look, you know, consolidate it with some other stuff that's working, so

Rabah Rahil (21:50):

Okay. But

Cody Plofker (21:50):

Not exclusive IG engagers. Yeah. Yeah. Yeah. But, but we, I mean, we definitely have had that split up separately. Yeah. Have you seens one? Yeah, definitely. I mean again, like, okay, cool. Our, our style is very top of funnel. Like if you look at our ad library right now, we've got some new ones that are very educational. Uh, if you look at these thumb stop is really good at them. Average watch time is about double what our normal ads are. Exactly. Click three are, are better just by being educational. I, you know, comments and shares are way higher. Right. So I think that there's some value, but that's not something I can necessarily look at and be like, all right. That, that correlates to, you know, more cheaper acquisition costs or whatever, but theoretically it, it again, should build that middle funnel a lot better. And then also build more affinity with those people. Yeah.

Rabah Rahil (22:35):

You're gonna, and by, oh, I was just gonna say, and by definition, the ad's gonna win more auction.

Cody Plofker (22:40):

Yeah.

Rabah Rahil (22:40):

Because it's just not just the bid. You're gonna have the estimated action rate and then you're also gonna have a higher ranking, um, because it's gonna be a more engaged with ads. So you're, you're kind of winning on all fronts there. Do you look at any custom metrics like, uh, revenue per click or AOV or

Cody Plofker (22:55):

Anything like that? Yeah. I look at, I mean, I guess click to purchase is custom metric, right? Yeah, yeah. That's yeah. Yeah. So

Rabah Rahil (23:01):

Conversion

Cody Plofker (23:01):

Rate, click, click to purchase is obviously one of the main ones click, click to add to cart. Um, I don't really look at, uh, like add to cart, could purchase a ton. I have it in there. It's probably not a main one. Um, and, and RPS definitely.

Rabah Rahil (23:14):

Yeah.

Cody Plofker (23:15):

RPS AOV obviously.

Rabah Rahil (23:17):

Yeah. I love it.

Ash Melwani (23:19):

So a question if you're doing, so you're saying you're doing like a lot of educational content, right? Yeah. And a little bit more. And I guess the goal is to kind of build that middle, right. You want, you want a more educated consumer making a purchase with, you know, your brand. How are like, what is your like process of at least seeing if that works right? Like how are you measuring that? Like, I feel like before it's a lot easier, like prior to like all the data loss, but like now how are you figuring out if the money you're spending on that top of the funnel for education is actually playing out to your manage later down the line.

Cody Plofker (23:57):

I guess this is my, this is my whole like brand first performance marketing thesis. Like it's still conversion optimized campaigns. I'm still optimizing off of a CPA. But if you give me two ads with a, a ROAS of 1.8, you know, and one of them is just a static image ad to a product. And another one is like an educational video with a, a much better average watch time. I'll, I'll take that one. You know? So if, if, and that's kind of a gut feeling, I don't, I don't that's subjective. I don't even know how I would go about finding that data to be able to see if there's any correlation between like those ad running those ads and like Google brand search later on. Right. Or, or anything like that or overall Mer, you know, cause again, the theory would be no, that's fair.

Cody Plofker (24:38):

They're they're, they're seeing you, they're remembering you. They have a good feeling about you. The other thing I'll look at like a lot of these educational ones go to an educational landing page, which then go to a quiz. And so in triple we can look at cost per email sign up. So for example, one of these is going to an advertorial five reasons why that goes to a quiz. We're getting like $7 emails on purchase, optimized campaigns. So even though our, our return on ad spend is not, and we only launch these on, I launched these on Sunday. So it's early, you know, but even if our ROS is not quite the same, like, or if we are the same, that's a ton of incremental asset building that we're doing, whatever you want to call, call it that I think would, would get a better LTV to C over the long term. You know what I mean?

Rabah Rahil (25:18):

Yeah,

Cody Plofker (25:19):

Absolutely. Like our email list has, is huge. Now. Like I looked last month and hadn't looked in a while and it's like doubled in size just because of how much traffic we're spending to, to quiz that's worth something.

Rabah Rahil (25:32):

And you obviously have flows on the back end, et cetera, et cetera, where these touchpoint are gonna be way cheaper than, uh, proper paid media.

Cody Plofker (25:40):

Yeah, exactly. Hopefully we're trying to, you know, use paid media to build our brand instead of just using it for, for that purchase. But again, that being said, well, this is not brand, this is brand first performance. So we are still optimizing for purchases, we're bootstraps. So we can't just throw money out there and hope it comes back. Like we're still optimizing to a CPA and you know, like, yeah, we're still optimizing to a CPA.

Rabah Rahil (26:03):

However, so you optimized to a CPA as well. Yeah. Cause that's what Ash does. So because you have like an well, you're kind of a young business though. Right? So how do you, how do you nail down that CPA with, uh, LTV?

Cody Plofker (26:16):

Uh, I mean, we're also, I boots scrap to know ashes as well. So like we're not really going off of an LTV. We're we're also just preferences and kind of growth trajectory and organic traffic. We're we're not aggressive, like anybody that would come in here and would be like, you need to spend way more. We just choose not to, for, for several reasons. One is inventory. One is profitability, you know, for sure we're not trying to grow at all costs. So we're, we're really, I guess, optimizing to an Mer more than anything. Yep. That try in there that that's kind of the, the main thing. Obviously we know our contribution margin. We know we wanna break even, or be profitable on first purchase. So we're, we're kind of not going on a, a LTV payback period. We're kind of going off a contribution margin.

Rabah Rahil (26:57):

Yeah. That makes sense. Does that makes sense?

Cody Plofker (26:58):

Yeah. And you can, and again, you can grow a lot faster optimizing off of a 90 day LTV or sure. A 365 day LTV. It's just not what we're choosing to do at the moment.

Rabah Rahil (27:07):

Yeah. Year LTVs. I do not recommend unless you have like a two to three year old business because, um, there's just a lot of fuckery that can get involved with, uh, extrapolations. Especially when you get out to a year, you also run into a lot of macro headwinds that you probably won't be able to predict, but yeah, that's super smart. Yeah, totally.

Cody Plofker (27:25):

I love it. So, alright. So here's my, what I realized this week and tell me if you guys kind of agree with this, if this makes sense or not. So I noticed, and I tweeted the other day that like the better an ad does when you launch it. The worse actually thes looks in Facebook for a while. And my theory is because a better ad Facebook's kind of put more spend behind it, right? Yes. And like at the levels, you know, Ash, Ashley and I are spending, you know, 20 plus K a day. Like when you put something in evergreen it's and it's Facebook likes it, it's gonna get spend, it's gonna rack up that spend really, really, really quickly. But the purchases are not gonna rack up that quick. They're gonna be more and more delayed. So it's like, it's like, you have this Seesaw and it's like in platform.

Cody Plofker (28:02):

Yeah, yeah. In platform. So you have the Seesaw where you're like waiting it down. It's like, you're trying to balance things in the old days as spend went up, you would get purchases. So you could tell pretty quickly now it's like your spends here. Your RO is like, not, it's like not coming in because it's so delayed. So it's like the, and so the other, the that's one and you really have to be patient on, is the stats look terrible? Just like, like, like I'm telling you like God awful. And, and I turned off awful lot of stuff in the last month and few weeks that I, I shouldn't have, and then obviously hit it with the delayed attribution. But the other thing, because now the conversions are reporting on time of correct me if I'm wrong, it's it's time of purchase and not time of impression. Correct?

Rabah Rahil (28:43):

Correct. So it's, uh, the way it's bifurcated is click an event before it used to be click. And so if you say day zero, yeah. You clicked on an ad and then bought on day three mm-hmm <affirmative> day zero would get that credit. Now, if you click on an ad on day zero, but then buy on day three, you're gonna get day three is gonna get attributed. And so in triple, well you have a click in event date, so yeah. You're absolutely right. So,

Cody Plofker (29:06):

So you can, those, those purchases, right? Especially for us, the majority of our purchases don't happen on the first day. It used to be that if somebody purchased it would go, Facebook would go back and then attribute those purchases to those first three days of the ads. Now it doesn't. So overall with that kind of new way of delayed attribution, the overall ROS is going to get better as, as you let things run. Correct. However, right now, if I went back in, even though our overall CPA is decent, if I went back in our first three days would still look awful.

Rabah Rahil (29:35):

Yeah.

Cody Plofker (29:36):

Yeah. So I, to me that that's fascinating. It's just something to keep in mind. It's like things are just going to look awful and the more you're spending on something, the worse they're gonna look, I think

Rabah Rahil (29:45):

Exactly

Ash Melwani (29:47):

This is, can I say something that might, I'm not, I don't wanna to run you the wrong way. This is where I think our first part, the second part we were talking about optimizing based on that one day clip, right? For you now you want that journey is gonna be a lot harder for you to understand what is being attributed to your spend, versus if I'm going for that one day, click and conversion, I don't have that. And I can make decisions a little bit more quicker than you can. And, and I'm saying either one is right, right. You're going for that brand performance marketing first, I'm going for like that direct response, I'm just acquiring to acquire. And like, those are the issues that you're gonna have. Right. And that's, that's all I wanna say about that, but I mean, I totally understand the whole, the premise of like, okay, well I wanna educate, I want people to go down this funnel, but that's kind of why I asked you the question earlier. It's like, how are you measuring that? Are you like, you kind of have to let things run and then look it back at it on like a seven day like window and be like, okay, well, is this good? Great. Let's keep going. But then like, you can't make decisions day to day, which is like, it's tough. Yeah. It's tough if you can't do that for

Cody Plofker (30:54):

Sure. But you're not, but you're also not making decisions day to day either. Like you said, you don't touch it for a week. Oh,

Ash Melwani (31:00):

Whoa. A hundred percent, but I'm getting, I'm gonna get data right away. So that I know for a fact something is working on day one. You don't necessarily know that.

Cody Plofker (31:13):

No, I don't. I don't. I don't. Yeah.

Ash Melwani (31:15):

And again, I'm saying I'm, I'm more so saying that I'm loo I'm not touching it for seven days because then after seven days worth of spend, I know this is a clear winner, but on day one, I know if something is heading in that trajectory, which is the only

Cody Plofker (31:28):

Yeah. And, and I don't. Yeah, you're right. I don't, so I think two things that are important, a you have to have like strong hands to do this. Like you have to like, get computer, have patience, being willing to put 10 K into an ad. And it looks like you're at like a 0.1, it looks like you've made a thousand back and like, you just have to be okay with that. But I also think that's why you also have to be testing in a, in a sandbox campaign testing in an environment where you're controlling the spend and you're, you're willing to essentially risk or lose that amount of spend and only graduating things when you have the confidence. Like, I think I would have a lot less confidence land launching a brand new thing with that budget. You know, I, I don't think I have the confidence to do that.

Rabah Rahil (32:07):

Yeah, absolutely. The other thing too, Cody, you can look at is when you do in triple well, you can actually look at quick date. So, um, to as Ashley's point when you guys are spending at the scale you're spending at quick, date's a much better way to look at, um, the attribution than event date. Sure. Because you're the event, date or event based attribution when you're spending at the scale you are and launching these new creatives, the, the credit could be going to the wrong people. If that makes sense. Like the click base actually is gonna be a better clearer read, um, when you're spending at that scale because, uh, you're, you're gonna pervert the data. I'm looking at event date, because again, day zero is actually what caused the purchase. But day three is what's going to get the yeah. Um, the credit

Cody Plofker (32:50):

I gotta, I gotta do that more. So cuz like, so cuz we're running a lot of, you know, lead gen again, still purchase optimized, but building our email list and we a, we have a ton of product launches, right? Like we launch multiple times a month just at, at this stage in our, you know, our journey, but also we've had a ton of inventory issues. So a ton of back in stocks,

Rabah Rahil (33:08):

Big deal, man.

Cody Plofker (33:09):

So like I know just from, you know, customer feedback on ads, CX social, that a lot of people will comment on ads and be like, this looks great, but I'm waiting until blank is back in stock. So I bet that there's like a big discrepancy between click and event date for that reason. Yeah. And it'll be really interesting to go back to click date instead of looking at event date. Cause like for example, we had a launch today or we had a back in stock today, like today's through the roof. Yeah. And I know some of these customers, we acquired them as at least acquired their attention and acquired their email four days ago or whatever,

Rabah Rahil (33:42):

A hundred percent. Uh, that's kind of, if we can kind of segue a little bit into some attribution models cause Ash and I had a, a pretty interesting discussion on that. So there's this ultimately a few in triple well there's uh, first click last click fractional linear. Um, and then triple well, uh, plus views. And so ultimately the way I think about them is I'm gonna use that first click attribution model, um, when I wanna scale and see what my best prospecting ads are. So what, what are kind of to Ashe's point when somebody has a first click, what caused that purchase on the first click? Last click is gonna be the inverse where I'm gonna scale up my retargeting campaigns based off last click and then triple whale plus attribution is, or plus view through is gonna be kind of your best overall ads, but cover your ears.

Rabah Rahil (34:27):

Cody ear Musk. This has view through conversions. And so you want to take that with a grain of salt and then fractional is just gonna show you fractional linear, excuse me, you're just gonna spread that across the channels and you're gonna see what channels in your ecosystem are actually, um, driving the conversions. And so I think that's a really interesting mental model in terms of how you're gonna scale your campaigns and look at how to leverage those attribution models. Cause I get that question a lot. I don't think, um, until Ash and I started jamming about it, that we had a actual, a concrete answer on that.

Cody Plofker (34:56):

So what are the, what are the pros and cons like, should there be one attribution model for your business or should you use different ones at, at, at different times and what are, what are some of the pros and cons of these different attribution models or limitations?

Rabah Rahil (35:10):

Yeah, great question. So I guess I should probably ratchet back up for people that might not, uh, be hip to it. So first click means it's gonna attribute all the, um, actions to the first click. The first touch point of that journey. Last click is gonna attribute the, uh, all the success to the last touch point of that journey. Um, fractional is gonna, uh, spread that out across any channels that were involved, um, equally. So this is there's different fractional models. We use fractional linear. So it's basically gonna there's four touch points. All four of those touch points are gonna get 25% of the conversion. Um, and then triple whale, uh, plus view through is gonna be last platform touch. So it'll exclude, uh, organic and email. Um, and then it'll also bring in view throughs as well. When you have triple attribution plus view throughs. Um,

Cody Plofker (35:55):

I love those views.

Rabah Rahil (35:57):

Yeah. Hey, Hey, they're real. They don't get me wrong. They're real. But it's just, um, you wanna scale again in terms of the attribution model that will help you understand what is closer to the truth in terms of the participation of that ad. Um, and so kind of to your question, Cody, I think that the problem with first click is that, um, it might not close. And so if you have a longer again, uh, sales cycle, if you look at first click for retargeting, probably it's not gonna be the path because this wasn't the person that got the person close. This was the person that got 'em into the funnel. And so your educational videos might be fantastic and prospecting. And so that's why I like to use first click or prospecting, whereas retargeting, you wanna see last click is gonna show you who's the closer mm-hmm <affirmative>, who's the person that is closing that sale. And so that's more in my opinion, applicable to that retargeting. So you're maybe statics or whatever, but you wanna see that retargeting. And then again, the triple whale plus view through is gonna show you the kind of best overall ads that are working there. And then the fractional is gonna show you what channels are actually causing, um, purchases. So that, or that's kind of just how I think of it.

Cody Plofker (37:05):

So say you're like eight sleep, right? Like let's say you're selling mattresses with obviously a long sales cycle, you know, you would be using for your prospecting first click. Like that's kind of what you'd be relying on because correct. You know, they're probably not buying from that ad directly, but you want to know what's feeding them.

Rabah Rahil (37:22):

Yep. So when I'm, uh, looking at the success of my, when I'm looking to scale my prospecting ads, I'm gonna use that first click attribution. And then conversely, when I'm looking to scale my retargeting ads, I'll switch that attribution model to last click. Cause I wanna see what ads are actually closing people.

Cody Plofker (37:39):

What if you were, you know, I don't know, but what if you're a low AOV product, like a $30 product, you know? Yep. Pretty low sub sub $50 AOV. What, what would you recommend

Rabah Rahil (37:48):

Mostly first click? Cause you're probably not gonna have the opportunity to do retargeting because you're gonna have economics that just don't warrant that. So I wanna see what can cause a purchase or at least get somebody into my funnel and then I can do the code thing and look at my email signups hit 'em on email, but,

Cody Plofker (38:04):

But you wouldn't do last click for that, for that?

Rabah Rahil (38:07):

No. I mean,

Cody Plofker (38:08):

Because it's a shorter funnel or like that

Rabah Rahil (38:10):

Would exactly, I mean you could kind of have six on one or the other. It's more so the layers of where they ads are. So I wouldn't be running retargeting in a low AOV, um, type of business. And so then that makes kind of last click, a moot point. But into your point, it's, it'll be a little bit of six and one half dozen, the other with first click last click, but it's more so where the ads are in the funnel is when I choose to use the attribution model. And honestly, in that sense, I might just be looking at triple wa plus view throughs. Cause I'm just trying to figure out what the hell's driving some semblance of performance because mm-hmm <affirmative>, you don't have that sophisticated of a marketing ecosystem. You're basically an AOB kind of break even at that point. Like, so whatever's getting me on basically ultimately the cheapest first touch points, get me in.

Ash Melwani (38:56):

Well, to that point, what if like, I would imagine first click, I think might kind of match up, right. And that's,

Rabah Rahil (39:02):

It's gonna be very close. Yeah. It'd be very close to your first click and your last click because there's not a lot of, uh, differentiation in the journey and there's not gonna be multiple touch points usually.

Cody Plofker (39:13):

Yeah, definitely. What about tell, when is fractional important? Like when does fractional help?

Rabah Rahil (39:20):

For me, fractional is useful to understand, um, channel kind of contribution. So not necessarily like

Cody Plofker (39:26):

Between paid channels

Rabah Rahil (39:27):

Between paid channels. Exactly. So you're gonna see like when you're you use that fractional attribution, you're gonna see as Facebook driving it is Google driving it and so on and so forth. And so you're gonna see that RO as switch. Um, and you gotta remember fraction will have kind of more than the whole because it'll give credit across all those channels. And so you're, if you have 10 purchases, you could have 20, um, purchases spread out across all these different channels. And so that's why it's not necessarily a great quantifiable attribution model, but it's very good at qualitative where it's like qualitatively this channel's driving and this channel's driving and this one isn't. And so that's why use when I would apply fractional where it's more so to understand at the channel level contribution more so than the campaign adder ATS it.

Cody Plofker (40:11):

So in general, are there any data delays on triple well, like obviously Facebook is delayed up to 72 hours. It's like time of conversion.

Rabah Rahil (40:19):

Yeah, pretty much. Yeah. So that's, what's nice about it.

Cody Plofker (40:22):

And what is like, so on like triple attribution plus view is like, what is the attribution window? Is it, is it seven days? Is it unlimited?

Rabah Rahil (40:30):

It's gonna be seven day view one day click. Okay. Or, I mean seven day click one day view. Yeah, exactly. Like the Facebook does open up, uh, that 20 day, 8, 28 day view or 28 day click one day view window. And you'll be able to see kind of, uh, longer look backs, but ultimately it's gonna be based on the, uh, the account and everybody's account is essentially set to seven day click one day view.

Cody Plofker (40:54):

Got it, got it. Okay. So, so then if you see any delayed, you know, attribution in there, it's because of people purchasing within that window, but outside of like that day. Yeah. It's not as of like a delay.

Rabah Rahil (41:05):

Exactly. And so you will seek again, like longer look back. So if you look at kind of the, the LTBs, you'll be able to say thirty, sixty, a hundred eighty, but again, it's gonna be more so to your point, like we're catching this and then attributing it to it, but it's not gonna be, um, within the attribution window, which is really what you care

Cody Plofker (41:21):

About. Yeah, yeah. For sure. Yeah. Cause, cause again, that's why I wasn't sure. I didn't think there were any delays, but like our triple ROS does continue to get better as, as you know, we launch stuff and it goes on and obviously it's because our it's

Rabah Rahil (41:32):

Just cuz you're so good.

Cody Plofker (41:33):

No, no, it's the opposite. I can't get, I'm not like Ash, I can't get them to buy on the first purchase. So we have to wait a little bit.

Rabah Rahil (41:39):

<laugh> he won't give this up. It's like a dog with a bone. I love this guy. Absolutely.

Cody Plofker (41:44):

I'm saying I'm not good at, I'm saying I'm not as good as Ash.

Rabah Rahil (41:48):

No, I, I love it, but sorry, if we got kind of nerdy there folks, but uh, I think it's really important to understand kind of the, the attribution models of when and when not to use them and um, ping us offline or whatever, if you do have more questions, but are we got nine minutes? Cody has a hard cutoff. He has to go fix, uh, his driveway pavement. They were off four feet. And so he's gonna go out there and fix it up. Uh, what else do we guys wanna get into?

Ash Melwani (42:13):

Um, I think what else was on that list? Oh,

Rabah Rahil (42:19):

Um, but you guys wanna do some, some trivia. I have some, some data. Yeah.

Cody Plofker (42:23):

I got one question for Ash, but then the, and then yeah, so Ash, you're saying, I was just thinking about this week, cuz you're telling me, you know, you don't touch anything in, in your main scaling campaign. Right. And you're testing 40 creatives a week. How are you getting enough budget to test those 40 creatives? Like it seems like either you're spending a ton on creative testing or you're really spreading out that budget to, to cover 40.

Ash Melwani (42:48):

So here's the exact budget layout and campaign structure if this helps anyone. Um, but basically I have two ad accounts. Each ad account is for a separate product. Um, one's like our weight loss product and one, one reading product in each account. I have a scaling campaign around 75 to 10,000 a day. Um, then within that is a testing campaign at a thousand, right? So it's about 10% of the scaling. So I mean it's really 10%. It's only 10% and it typically does drive decent conversions on most weeks that there's weeks where it's just really shit and I'll bring up our average, you know, acquisition cost. But I mean, as we keep making iterations, it tends to start, it'll tend to get better. Right. So as we get closer to, okay, well this angle is working and, and it it's really broken down across like four 40 creatives are broken down across four different angles per product.

Ash Melwani (43:57):

So it's really not, it's like five it's five ads per angle per product. Right. So the thing is out of all those out of like all let's say between the two testing campaigns, there's only really gonna be two or three that get all the spend. Right. Yeah. And I know people are like, well, how are you? How do you know the other creators are not getting spent this and that? Like, I don't care. Like if Facebook says, this is what needs to get the spend and this is what's getting engagement and it's converting, like that's the winner. Like I don't have time to be like, okay, well let's set up an ad set. Let's put this here and this here and this here. And like give every single creative enough budget. Like, no,

Cody Plofker (44:33):

You should do 40 assets. You should do 40 assets. One <laugh>

Ash Melwani (44:37):

I will, I would rather like I would, I'd sell the business and not be dope. Like there's no way that I'm gonna do this every week. I need to give Facebook 40 creatives. Let it decide if it decides this is a winner, great. Put in scaling. Like you just have to make it easier on yourself. Like you can't be like so scientific and like, well, like I gotta have statistical significance on like this adset versus like enough, like just make it simple.

Cody Plofker (45:02):

And so some of those creatives are like not getting any spend really.

Ash Melwani (45:06):

Yeah. And, and the reason it's not is because Facebook doesn't think it's engaging. Yeah. You know?

Rabah Rahil (45:11):

Yep. And, and to be fair too. So I used to, when I ran my agency, I used to do a lot of dynamic creative testing. Um, and that was a big issue because, uh, Facebook was basically a first mover. So the first, uh, in, in dynamic creative. And so like the first thing that hit dynamic creative, which is eat up all the spend. And so what I would do is pull out kind of the winners and then let those other people fight it out. But what Ash is doing is actually way better because you, you, that is optimized exactly right. Where, whatever Facebook likes it's gonna get and it's gonna spread out that, that optimization pretty, pretty effectively. Whereas in dynamic creative, I had multiple winners a lot of times in that. And so I had to pull them out and then scale that and then run that dynamic creative test again. But Ash is essentially circumventing that by giving basically everybody a free chance and it's not gonna be as much of a winner take all issue. Um, as when you do do dynamic creative testing, it's a big, big issue where basically like one thing wins and then you'll ne nothing ever sees any sort of spend at all.

Ash Melwani (46:12):

Yeah. And to be fair there, there's going to be weeks where we don't have time for, you know, the designers to pop out creative. I'll probably go back and find the, the creatives that didn't get any spend and maybe put them into another environment where it's with the other creatives that didn't get spent and then see what happens there right at the end of the day, if something performs and it's, you know, creating a better CPA than something that's in the scaling campaign. Great. That's all I care about. Move on to the next. But like I said, you know, now we have a backlog of creatives that may, may make it to the next round of testing, you know, but that's just how we're doing it. And it's, it's helping us find more stability. And um, I mean our, our CPA's coming down, so

Rabah Rahil (46:54):

That's huge. That's huge. Okay. Let's get into a couple trivia questions and then we'll sign off cuz it's Cody, we're pushing up against it. Okay. Uh, we tracked about what we 578 million in ad spend. Um, what do you think the top spending channel was and how much Cody? You can go first and then Ash

Cody Plofker (47:14):

Facebook.

Rabah Rahil (47:14):

Okay. How much? Five. So 5 70, 78 is the whole, yeah,

Cody Plofker (47:19):

I'm saying 60% of that. I, I'm not gonna do public math, but I'm gonna say 60% of

Rabah Rahil (47:23):

That. I'll do it quickly for you. Whoops.

Ash Melwani (47:28):

That 300 million something

Rabah Rahil (47:30):

Around that. Look at man. That's not bad. Three forty seven, three forty seven four. We'll call it three, three forty eight. Okay. Ash, where are you at big

Cody Plofker (47:37):

Spin channel?

Ash Melwani (47:39):

Cody said 60%. I, I think it's closer to like 75 80.

Rabah Rahil (47:43):

Okay. Let's go. I'll give you 80. Oops.

Cody Plofker (47:46):

So that's yeah, you're probably, you're probably more right.

Rabah Rahil (47:49):

You can't change it. I'm thinking by

Cody Plofker (47:52):

Mine prior on

Rabah Rahil (47:53):

4 62. Okay. Four. Okay. Well, if we're doing prices right rules, you would bust Ash, but you are definitely way closer. Uh, Facebook was 450.3 million, um, in quarter one. Geez. And again, this is gonna be a quarter over quarter comparison, but it was down 26%, but there's some seasonality in there where, um, you know, Q1 is always gonna be, um, less than Q4, but okay. What's the second, second channel.

Cody Plofker (48:23):

Ooh,

Rabah Rahil (48:23):

Google or TikTok? Google.

Cody Plofker (48:25):

I'll give you Google. Okay,

Rabah Rahil (48:26):

Cool. Where you at? How much? Yeah, so 4 27. What is

Cody Plofker (48:31):

It?

Rabah Rahil (48:31):

Yeah. Or four 50, excuse me. Four 50 is gone. So you got,

Ash Melwani (48:36):

You have a hundred million left

Rabah Rahil (48:38):

Hundred 28.

Cody Plofker (48:40):

So I'm saying if it's 1 28, I'm saying like 17 of it went to talk. The rest went to go Google.

Rabah Rahil (48:49):

Okay. So is that 111? Okay. Where are you at Ash?

Ash Melwani (48:54):

Yeah, I think that split makes sense, like hundred to Google. The rest of the TikTok

Rabah Rahil (48:58):

Guys. You guys are a little better on this one. So I'll

Cody Plofker (49:01):

Go. Yeah, I'll go. One 11 Ash. We'll go a hundred.

Rabah Rahil (49:03):

Okay. One. Oh, you're one 11. Boom. You guys are tied now. So I guess the TikTok, well, I guess you can do the math. So the TikTok won't even be the ad breaker, but yeah. So buck a buck 17, basically 116 point 78 million. Um, and again, this is across about 2,700 merchants in, um, 60 different countries. So, um, and that was up 31% quarter over quarter, which is very interesting to me. Um, and then TikTok, even though it is growing like, uh, a weed it's 110% growth quarter over quarter, um, basically came in at 10.9 million. So it's still just an absolute drop in the bucket, which is was

Cody Plofker (49:38):

Crazy. That'd be interesting to see though, like, you know, as you track that, how that's gonna go court.

Rabah Rahil (49:42):

Well, for me, it'll be interesting when we have Q2 over Q1, because Q4 just has a lot of seasonality perversions where it's gonna be one of the most expensive, like it was the most expensive CPMs are way down compared to Q4. So it's obviously gonna be the most expensive and usually the highest scale cuz that's when people spend the most. But I thought I'd give you guys a little, little taste of some fun data here. Um, Cody, we have one minute sign off, tell the people how to sign. I

Cody Plofker (50:05):

Don't have like a hard stop. So I got a few more minutes.

Rabah Rahil (50:07):

Okay. Well I got a piece. I got a hard stop. Well, you got me on the Ker kind of hydration and now I, you know, it's just that you're just look

Cody Plofker (50:18):

How good your skin looks. Look how hydrated

Rabah Rahil (50:20):

You look. I'm. So I'm burnt though, that that Marine layer got me. It was freezing in San Diego, but I didn't realize you can get my hand, got sunburn. Ah, hang was, I was a wreck. I was a wreck. Um, I'm gonna try to get these guys out to Miami geek out. Let's go pees. Um, Cody tell where the people were to follow you. Uh, tell 'em about your newsletter. What else you got going? Your masterclass? Let 'em all know.

Cody Plofker (50:40):

Hit me up on Twitter. Uh, Cody plot on Twitter, Lincoln bio. You can find my newsletter, uh, actually just about to release like a, a personal website. So that'll have a link. Oh cool. In there as well. But uh, yeah, that, and then if you wanna wanna kind of, you know, borrow some of my time on, uh, on mentor pass, you know, you check it out, schedule a half hour hour call with me just to get any, any questions answered. But if not, just hit me up on Twitter.

Rabah Rahil (51:04):

Fantastic. He's a wonderful feed and a great newsletter as well. For sure. Ash, let the people know,

Ash Melwani (51:10):

Uh, follow me on Twitter, Ashton. Melani I'm gunning for 10 K followers hopefully soon. <laugh>

Rabah Rahil (51:16):

Jeez, you savages. That's incredible. You guys are massive. Let's go. Let's get him over the 10 K folks. Come on, come on. We'll give you, we'll give you some swag or something like that. We'll hook you up. We'll get you out there. <laugh> my hobby swags. Awesome. Yeah. You guys launched some really cool stuff.

Ash Melwani (51:30):

Yeah. Some new stuff coming out, but yeah, let's see. Uh, let's see how it goes.

Rabah Rahil (51:35):

Cool. What, no product plugs or nothing. You guys just launched the cool collagen stuff, right? The, the new flavors or anything.

Ash Melwani (51:41):

Oh, we just launched new. Oh, we're launching new gummies this week. Uh, Friday. We're doing super fruit gummies. Uh, so get your fruits and vegetables in and then ashwaganda gummies. So keep going.

Rabah Rahil (51:53):

Oh, love. It'll be good.

Ash Melwani (51:56):

Very good. Cody. How'd you like, uh, the collagen. You

Cody Plofker (51:58):

Picked it up, dude. I loved it. I, I loved it. Mix it, mix it in with some almond milk. Yeah, I got the, I think it was like the cookie one. I don't know if it was that

Ash Melwani (52:05):

It

Rabah Rahil (52:05):

Wasn't

Cody Plofker (52:05):

Donut one. It was enden one. Yeah. Yeah. It was. I, I really liked, it tastes great. Like good looks dessert mix. Yeah. Yeah.

Rabah Rahil (52:13):

And you don't even have to order it on the site. Right. You guys are in what GNCs or vitamin shops where

Ash Melwani (52:17):

You at shop vitamin shop nationwide 600 stores nationwide. Um, just launch two weeks ago. Lets go. Um, so if you're buy one, go check it out. Send me a picture between me a picture. Um, but yeah, just yeah, let's check it out.

Rabah Rahil (52:33):

<laugh> beautiful. All right, folks. Uh, if you do wanna get more involved with triple oil, we are tri triple oil.com. We have a wonderful newsletter that goes out, uh, whale mail Tuesday, Thursday, and then we're on the bird app at triple oil boys. Thank you for another session. Thank you for all the time in the fantastic questions. Cody, happy to have you back. Congrats again on the bun of the oven. Well done, sir. Thank you. Um, and Ash love the hoodie we got, I, I need some swag help here. Uh, we, we are dying for some swag and you, you are the, the man. So I'm gonna pick your brain somewhere on that. Uh, folks, subscribe, share, do whatever you need to do. Spread the gospel farm wide. Um, and that's all we got for it. Another one in the books. Thanks guys.

Speaker 4 (53:12):

Please

Ash Melwani (53:13):

Take care.

ā€

Podcast

How To Optimize Your Ads Post iOS 14

March 18, 2024

53:29

Hosted By

Rabah Rahil
CMO at Triple Whale

Guests

Ash Melwani
Co-Founder & CMO of MyObvi
Cody Plofker
Director of E-Commerce Jones Road Beauty

Episode Description

In this episode of adspend we go over how to navigate and optimize paid media and ads in a post iOS 14 landscape. Ā #Adspend

Notes & Links

šŸ“§Subscribe to Whale Mail for exclusive industry insights and in-depth marketing breakdowns: https://www.getrevue.co/profile/tripl...šŸ¦

Follow us on Twitter for Industry insights https://twitter.com/triplewhale

ā€Follow the people featured in this episode here:

- Rabah's Twitter: https://twitter.com/rabahrahil
- Cody's Twitter: https://twitter.com/codyplof
- Ash's Twitter: https://twitter.com/ashvinmelwani

Transcription

Ash Melwani (00:00):

I would rather like I would, I'd sell the business and I'd be dope. Like there's no way that I'm gonna do this every week. I need to give Facebook 40 creatives. Let it decide if it decides this is a winner. Great. Put it in scaling. Like you just have to make it easier on yourself. Like you can't be like so scientific and like, well, like I gotta have statistical significance on like this adset versus like in like, just make it simple.

Rabah Rahil (00:35):

All right, boys, we're back for number six. I'm a little sunburnt. I'm peeling, like crazy, almost as tan as Ash, but Hey, Hey, I'm keeping up. I'm still beating Cody. So here we are. We're back for another episode of ad spend. Cody is back. There might be some feistiness. He has overcame the Ronna, like a legend. He has a new house on the way. He also has another thing on the way a baby Cody.

Cody Plofker (00:58):

Yeah,

Rabah Rahil (00:58):

Right? Yeah. So how exciting is that

Cody Plofker (01:01):

A girl though? So not

Rabah Rahil (01:03):

Me. Yeah, but even even more exciting. So congrats there. Ashe is looking fresh in his, uh, fancy hip hoodie as always. Um, and so I think today we're gonna get into a little bit of how to optimize post iOS 14. Um, I'll drop some knowledge, Ko drops. The knowledge Ash will drops. The knowledge will kind of circle it in a, a VA and then give it to you guys in a beautiful podcast format. So without further ado, Cody take us away.

Cody Plofker (01:28):

Yeah. Dope. Uh, this was kind of my idea what I wanted to chat about. Cause it's definitely something I have been struggling with. I am no expert at it. I don't know if anybody is an expert at it, uh, to be totally honest. So I just would love to chat about it cause I'm sure there's a lot of people out there who are struggling with this who are frustrated. It's no surprise that Facebook attribution is absolutely terrible. I mean, I, I tweeted the other day, but like I was just looking at our, our, uh, our ex you know, our exclusions. And I was looking at our, our Clavio 180 day purchasers compared to pixel 180 day purchasers. And the pixel is like a third of the size. Yeah. Like it is crazy how bad attribution is. And, and in, in both ways, in terms of over reporting and under reporting. So I think you absolutely need first party data. You need to use Mer, but I think a lot of people say that and you know, there's only so much you can say on Twitter. So I, I figured let's just talk about it and Ash, and I will share what we're actually doing to optimize our ads. And then Rob can obviously share everything that he's seeing being inside of triple oil.

Rabah Rahil (02:34):

You're getting good at this Cody. That was pretty smooth. I just kind of tossed it over to you. Yeah, I was. So once you get a studio, it's over, it's over people. He's hot.

Cody Plofker (02:42):

He's hot. So Ash, would you talked, obviously you dropped some heat, you know, two pods ago about your creative testing, obviously with that much creative testing you're doing, you must be looking at, what's working a lot of optimization. What's your process for that? How much using Facebook, how much you GA how much using triple oil?

Ash Melwani (03:01):

Yeah. So I think one of the biggest things that I've learned to stop doing is looking at Facebook as like your, your north star. Right. Um, I've kind of done the test with like all the different attribution windows, and I've seen over reporting under reporting. Um, some situations where the reporting is kind of matched up to, you know, these other things like GA and triple, um, I've kind of had to really just pick one and stick with it and see if the overall kind of starts to follow. Right. So what I've been doing recently is literally only looking at the data from triple whale pixel, right. Um, so, you know, two pods ago, I mentioned, you know, the whole creative testing, you know, we launched everything on a Sunday, uh, at midnight and I let it run for seven days. Right. So basically no

Cody Plofker (03:54):

Matter just, just for people here and no matter what, even no matter what terrible, you're letting it run.

Ash Melwani (04:01):

Like, so actually it's crazy. Cuz last week I had a really bad test that just wasn't performing. So I did cut it like maybe two, three days earlier than I usually would have. Right. And it, it, that's just how it's like, you might have a really good test. You might have an okay test and you might just have a shit test and that's it. Right. Um, but the way that I'm deciding that is I'm looking strictly at triple oil, right. So for example, yesterday I even tweeted or no, sorry, Sunday, right? Uh, day one of testing, uh, for the week thousand dollars budget. Um, and Facebook said I had $183 CPA. Okay. Um, obviously, you know, you're not gonna get data for 72 hours, so you got, you gotta let it run this and that. But like, I don't want to rely on waiting 72 hours and I'm sure there's other advertisers who don't have that budget to let something run for 72 hours at like a thousand dollars a day.

Ash Melwani (04:55):

Right. So you need to know what direction these ads are going in as soon as possible, just so that, you know, and it depends on the appetizer and your budgets. Right. So looking at triple whale, at least for my testing, when I'm testing top of the funnel and RA you can kind of get into this later, but I'm really looking at first click attribution. Right? Yeah. I want to know which of my ads created that awareness and eventually led to a conversion. Right. I wanna know what brought somebody into my ecosystem and later on converted them. Right.

Cody Plofker (05:26):

That's cause you don't, you want them to kind of take a lot of touch points and buy later on. Right.

Ash Melwani (05:32):

<laugh> I

Rabah Rahil (05:32):

Prefer, I told you he was back folks. I tried to tell you he was back.

Ash Melwani (05:38):

Um, no, but I mean, I'm optimizing, so at least my tests are on a seven day click. Right. But I'm really, I really want somebody to click or I want, I wanna know which ad introduce them to the brand. Right. For sure. Whether they go down the funnel,

Cody Plofker (05:52):

Obviously I'm just joking. Yeah. No, but I'm, I'm surprised that you're using first click click. That's your main, main north car. Yeah.

Ash Melwani (05:58):

Yeah. And what's what I have been seeing is that over since January, until now are because of now consistently done this process from January until now the testing creatives every week, following triple well, using that first click attribution as the north, I've seen our NCPA drop. Like sounds good. If you go and pull up our, our monthly it's literally like this. So now I have reason to believe that is just the way that's, it's gonna work for us on an overall level. So I'm gonna continue to do that. Right. Um, so basically, you know, I let it run for a week. I'm looking at, you know, I'm monitoring first click, you know, which has the lowest N CPA of all the ads, which one's getting the most spend. Um, and then I'm really just taking those ads and moving it into a scaling into my scaling campaign. Right. If that ad is beating the lowest performing ad in my scaling campaign, I'll just replace it. Right. Kind of recess the learning. And then the next week we'll see how it performs with a whole, technically a new environment.

Cody Plofker (07:01):

Right. And so you're only doing that once a week. Like you're only moving something into scaling once a week at the very most

Ash Melwani (07:08):

Mm-hmm <affirmative> otherwise once a week or maybe every two weeks, depending on if, if the test is beating my control. Right. That, that, that would be the only reason to move something over. Yeah. Or if my control is starting to perform badly,

Cody Plofker (07:23):

What about, so what, what are you doing if you're halfway through the week and you're, your results are terrible. Are you doing anything to that scaling campaign? Are you touching it or you're just letting it ride out. You're just focusing on creative testing and you fix it next week. What, what are you do?

Ash Melwani (07:36):

So here's where I think we're in a good place because now I'm constantly pumping out creatives. I don't have to worry about the weekly performance because I will have new stuff to replace it. So my weeks are a lot less volatile cuz they're technically newer creatives that have been proven. Right. Whereas before, if I wasn't swapping out creatives, as often, this stuff would probably be running for like months on end. And if I have a shit week it's because it's probably hitting that fatigue level. Right. So

Cody Plofker (08:09):

Then, but you're being proactive because you're being, because you're being so proactive about it. You're not really fine in that like steep drop off.

Ash Melwani (08:17):

Exactly. The only time we ever actually see anything is God knows what reason Mondays are our worst days. Like you'll have a Sunday, that's just a monster day. And then the next day, as soon as 12:00 AM it it's like conversion rate drops, NCPA increases. And like Monday is just the worst day and Tuesday it gets back to normal and Wednesday gets better. Thursday gets better. Friday gets better. And like, it's the weirdest cycle that I've seen. And I have no idea, but like that, I just know that Mondays aren't the best day. And I have to just let things run for the rest of the week. And it's like, okay, well at least I know that I'm doing everything I can on the creative side that I'm always gonna swap out new creatives for new winners and this and that. So it can't be creative. That's the issue. It's just, I don't know. It could be Facebook thing. It's just, maybe it's our demographic that just doesn't buy abundant. I don't know. I just, I, I can't believe, or I'm not led to believe it's like a creative issue, you know what I mean?

Rabah Rahil (09:13):

Okay. Okay. I have a thesis there. If I can interject mm-hmm <affirmative> cause our, our Mondays are horrible too. I mean, we're obviously B2B SAS, so it's not necessarily alignment, B2C or DDC. Um, but I actually think, and just anecdotally, like my Mondays are hell, I actually think people work on Mondays and that's my thesis. Like nobody shops on Mondays. That's like the one Workday, even if you're not aware, like workhorse, you actually do work on Monday. That's that's my, my big brain thesis. I think

Cody Plofker (09:40):

So. I mean, that would make sense.

Ash Melwani (09:42):

Yeah. Yeah. No, I, I, it's just been very noticeable for the last two years and like, if we over at the office, we're like, oh, today's slow. Oh, it's Monday, it's a Monday. You know what I mean? It's just, it is what it is. You just can't do anything about it. You just gotta let it run, you know? Cause if you make decisions on based on one day and that too, being a day that, you know, historically doesn't do well, you're gonna screw your shit up. So like keep it going, you know, just have, have trust in yourself that you're doing what you can do to kind of keep things less bottle tile. Okay.

Cody Plofker (10:12):

Okay.

Rabah Rahil (10:13):

Can I ask you something or, or go ahead, Cody. No, no, no. You go, have you tried, um, using like, uh, cost caps or bid caps where you keep the budget, but then you just modulate those cost caps and bid caps where like on Mondays, for example, you would, um, put the cost caps or bid caps, um, basically cost controls, lower those. So they'll still play if you, if the market's good, but if not, they'll kind of save you some money. And then on Tuesday through Friday, what have you, you just skyrocket those cost caps to meet, you know, maybe three, four X or C target CPA. And then cuz cuz you won't mess with the algorithm in that much or in that way. You're really not challenging the algorithm as much as you are because those budgets are already kind of hard coded. You're just, um, telling the stock broker essentially don't be so drunk here is only what you can spend.

Ash Melwani (11:02):

I've I think Septe last September till around December, I was fully all in on class caps, right? Yeah. Um, after that in January, what I was like here is my process. Here's what I'm gonna do. And I'm not gonna deviate from that. Right. I have like simplified our entire media grind process. Like I don't wanna mess around with cost caps. I don't wanna mess around the bids. I don't wanna have to go into the ad account more than once a week. Right. Wow. All I wanna do is I just wanna see if my NCPA is lining up with our KPIs. Right. For example, last, uh, last Monday and the Monday before, right? Our NCPA was above average for a Monday, right? Like our NCPA goal is 55. Okay. Our NCPA last Monday was $70 the week before last Mon that previous Monday was $70. And then throughout the week it lowers.

Ash Melwani (11:53):

Right. But this was the first Monday. We were actually at 55, which means it wasn't the worst Monday. But because we're, we're still making all these tweets and we introduced TikTok, right. We're scaling TikTok and those CPAs are so much lower than Facebook right now. It's bringing our overall average down. So now it's like, okay, love it. I'm keeping my spend where Facebook is right now. And I'm going so hard on TikTok so that it's bringing everything down and now I just gotta like rinse repeat. So that's where I'm like, okay, well this Monday could be worse. Right? Like a week ago I was $15 higher. Now I'm like at my goal and now throughout the rest of the week, I'm actually under my goal. So that's kind of like, you just gotta keep working at it and like nitpicking at it, trying to keep bringing it down as much as you can. I mean, set yourself a goal and then just work towards that, you know, just get a little bit better every week.

Rabah Rahil (12:43):

I love that.

Cody Plofker (12:44):

So I just wanna get back on track a little bit with like the optimization and like how you're looking at things. So data delays, right? You're not even looking at Facebook triple well, you're able to get some data back sooner cuz I, it doesn't have to get delayed pushing back. You're looking at, um, there still are going to be delays though, because obviously on the first day it's not gonna cover all the conversions that happen. If a lot of them are happening outside of the first day. Are you looking at any, do you like to look at, uh, any leading indicators? Do you look like, look at any on platform AME? Do you like to look at any cost per add cart or you're just CPA,

Ash Melwani (13:18):

Just CPA, at least for Facebook mm-hmm <affirmative> with TikTok. I think there has been more of a correlation in terms with like your CCPC and CTR to CPA versus those metrics compared to a CPA based on Facebook. And that again that's

Cody Plofker (13:37):

And I, I think that's where we're different. There's no right or wrong, but like I'm heavily into those, those metrics,

Rabah Rahil (13:42):

Software metrics

Cody Plofker (13:43):

Doing a ton of ADA analysis, like thumb stop. Uh, we recently did a correlation analysis and found that like three second views divided by clicks. I think what is, what it was, was like the highest correlation to CPA for us. So like if people have watched for three seconds, what percentage of those people clicked? And that had a pretty high correlation, much higher than thumbs stop, much higher than click through rate. So that's kind of a big one that, that we've been looking at. Um, I don't know what call it like a effectiveness.

Rabah Rahil (14:12):

So it's how is that different than a click rate though? Like it can't be right? Yeah. Function the same, same bit different. Yeah. It's a function of click rate. It's so interesting. I'm not saying you're wrong. I'm just kind, yeah.

Cody Plofker (14:23):

I don't know. I don't know if I totally understand it. Yeah.

Rabah Rahil (14:26):

When I think of thumb stop ratio, I think of a pattern interrupt where it's like, you're literally interrupting a pattern like the scroll. Well

Ash Melwani (14:32):

What's your formula for a thumb stop ratio.

Rabah Rahil (14:34):

It's sorry. Yeah. We should always be clear about

Cody Plofker (14:37):

That three second views divided by

Rabah Rahil (14:38):

Impressions, right? Yeah, exactly. I impressed. So that way I know I serve this person to add, but then they stop for at least three seconds. So that, so the difference makes sense to me.

Cody Plofker (14:47):

Clickthrough rates is from impressions. It's, you know, it's clicks, divided by impressions. This is right. Would be, I forget if it's clicks first or impressions first, it's always mess it up. But this one would be clicks divided by three second views. So it's kind of like once you got somebody, how likely are they to click? The other one is just how likely are they to click if you saw the ad? Yeah. So because essentially click through rate and I don't, I don't know if I totally understand this completely, you know, to answer your question, but click the rate. You're also actually really measuring your thumb stop cuz it's it's your, your measure, your number you're dividing by is impressions. So it's not just desire. You know, a lot of people use click through rate as like the, the D and ADA, but it's got just that cuz an element of it is also the stop and this is just how effective is your ad once you get them

Rabah Rahil (15:32):

To be fair. Let me just push back a little bit. Yeah. I can click without interacting with the ad. Whereas a thumb stop ratio is literally I have to consume for three seconds before it would trigger that thumb stop, but I take what you're saying. I'm just kind of being pedantic here where you can get shown to ad and I go, boom. And I slap it. The other thing is, uh, click rate is applicable across all creative types. Whereas thumb stop ratio has to be exactly

Cody Plofker (15:58):

It's

Rabah Rahil (15:58):

That's yeah. That's

Cody Plofker (15:59):

And that's probably

Rabah Rahil (16:00):

Two videos.

Cody Plofker (16:00):

We do a ton of video. We do almost no static and ASTA is a ton of static. So like it's a lot harder. Um, just to, uh, throw a disclaimer out there. Like those are, I guess leading indicator is two CPA CPA. Yep. I don't optimize off of those. If our CPA is bad, I don't want a good click the rate with a bad CPA. I think I just have to throw it out there. Cause if you talk about click through rate on Twitter, if you talk about thumb stop or anything else, people are like, don't optimize off those. Like those are garbage metrics like that. Just click bait. And it's like, it's, it's a lot more nuanced than that. You know what I mean? Mm-hmm <affirmative> like, it's really hard to have a really good CPA if you're getting an $8 click, you know what I mean hundred percent. So like it does matter to a certain degree and when you're getting that data a so not reliable, but also so delayed from Facebook's end. It is kind of nice to know if you're on the right track. So I will definitely kill something early. And I, if those AME are crap, like I just know if we have a 15% thumb stop or a 0.5% click through rate. Like no matter what the CPA's not gonna be good, but you

Ash Melwani (17:03):

A little more CPA first.

Cody Plofker (17:06):

Oh totally. Definitely. Yeah. Yeah. So it's like a order

Rabah Rahil (17:08):

Operations.

Cody Plofker (17:09):

Exactly. Totally. And then those are, to me, those are numbers that I might look at to try to help me improve CPA. I'm not gonna look at those in absence of CPA. I'm not gonna look at it in spite of it, but I'll look at those numbers to try to help me improve CPA. And usually as long as you're doing it, not click baby, getting those numbers better improves your cost per click and then it, it can kind of make sure you're having qualified traffic again with the caveat of it's not clip Baie or anything like that.

Ash Melwani (17:37):

Got it. So let me just clarify. So you did a correlation test based on, I guess a ratio. What are you pointing? This ratio? What do you wanna call

Rabah Rahil (17:46):

It? Yeah, this, the code metric

Cody Plofker (17:48):

Effectiveness is what is what I put, I don't know. Add effectiveness. That's

Ash Melwani (17:52):

Let's call let's

Cody Plofker (17:53):

Call.

Ash Melwani (17:54):

No, it's gotta be better. A hundred percent. It

Cody Plofker (17:55):

Has to be better. Um, you guys name it? Not me. The, the plot, the Plager nah, no, no. You don't wanna name it. Come on. Nah, no, one's gonna be able to say it right. That's

Ash Melwani (18:06):

True. Okay. So let's let's just, what is it? It's um, okay, so clicks divided by three second views. Okay. Did you run a correlation test? So you did a correlation test between that metric and CPA versus did you do the same correlation test with clicks divided by impressions

Cody Plofker (18:23):

Or sorry.

Ash Melwani (18:23):

So we what's that thumb stop ratio next

Cody Plofker (18:24):

Three seconds divided

Ash Melwani (18:25):

By impressions.

Cody Plofker (18:26):

Yep. Yeah. So we did it essentially. We did it on all those is we compared all of them to CPA. Okay. And found their

Ash Melwani (18:33):

And what did those look like compared to

Cody Plofker (18:36):

There were what you were saying? So nothing was statistically significant, right? Nothing is, you know, above, uh, 95%. Right? Sure. Statistical significance. However, there were some, there were some decent correlations, but the strongest one that we found was maybe 75% correlated. I don't remember exactly what was, was that one.

Ash Melwani (18:57):

Okay. Interesting. I, I fully believe it, man. I mean like, that's cool if you, if, if you have somebody that is kind of clicking, cause they've been, you know, they've related to that hook, right. You're pulling in for a certain reason. Right. And then they're probably more likely to buy X, whatever it is. Right. I think the same thing has happened for us. I know we're doing static and you're doing video, but like our static, our creatives are really nailing that singular angle. And, and we're just trying to iterate those angles because that is probably the reason people are coming in. Right. Where like, you know, I mentioned this two pods ago, it's like before I used to be, this is great for this, this and this. And then when we kind of just went, this is perfect for this, we got a little bit more qualified audience and we kind of tailored everything to that and conversion increased and our CPAN dropped a little bit. So I totally agree that like that hope that three seconds is like super crucial because it could just make or break and it could attract a different audience too. Mm-hmm <affirmative> so it's like super crucial to definitely measure that. If, if CPAs or like you're saying the delayed data is, is not coming in, that I think is, is definitely a good leading

Cody Plofker (20:06):

Computer. It'll just gimme, it'll give a little bit a it's that's on the creative analysis side. So then you look at all right. You know, Hey, this ad did. Okay. But let me, you know, click through rate was good. Average watch time was pretty good. So it was, it was engaging, right? Like my messaging is decent, but you know, conversion, rate's not there. A lot of people would look at that and be like, ah, it's a landing page issue. I don't necessarily think that's always true. I think a lot of the, the conversion is more of a persuasion thing. And it's what you're saying or not saying to sell them in the, the video as well. But again, you just look at those metrics and be like, all right, now, if we get, you know, we keep everything the same, but we get a better thumb stop.

Cody Plofker (20:40):

How does everything else improve? And theoretically it should. So a, that the iterative process is, is one reason I look at those metrics, but, but the other reason is again, if we have some correlations, they're very soft correlations, not hard, but it'll give me a little bit more confidence to let those ads continue to run with the bad CPA. Because not that I want ads to run with bad CPA, but because I know that that's very delayed data and there's a chance our CPA gets a lot better in the platform because I know that these are good ads from an engagement perspective.

Rabah Rahil (21:12):

Yeah. That's

Ash Melwani (21:13):

Fascinating. Now,

Rabah Rahil (21:15):

Do you, um, do you ever run, uh, and I'm gonna get roasted on Twitter, but uh, a guy named Jordan is actually, he's a killer and he's saying he's finding some success, some success with this. Um, are you running any OG video view campaigns? Like people have watched over X percentage of these retargeting?

Cody Plofker (21:34):

No. So you're saying like top of funnel, video views, or are you saying like no

Rabah Rahil (21:37):

Video? Yeah. So people that have watched 50% more of the videos, 75% more video or a hundred percent of the video, are you retargeting them at all?

Cody Plofker (21:45):

Yeah. Yeah. I usually like to look, you know, consolidate it with some other stuff that's working, so

Rabah Rahil (21:50):

Okay. But

Cody Plofker (21:50):

Not exclusive IG engagers. Yeah. Yeah. Yeah. But, but we, I mean, we definitely have had that split up separately. Yeah. Have you seens one? Yeah, definitely. I mean again, like, okay, cool. Our, our style is very top of funnel. Like if you look at our ad library right now, we've got some new ones that are very educational. Uh, if you look at these thumb stop is really good at them. Average watch time is about double what our normal ads are. Exactly. Click three are, are better just by being educational. I, you know, comments and shares are way higher. Right. So I think that there's some value, but that's not something I can necessarily look at and be like, all right. That, that correlates to, you know, more cheaper acquisition costs or whatever, but theoretically it, it again, should build that middle funnel a lot better. And then also build more affinity with those people. Yeah.

Rabah Rahil (22:35):

You're gonna, and by, oh, I was just gonna say, and by definition, the ad's gonna win more auction.

Cody Plofker (22:40):

Yeah.

Rabah Rahil (22:40):

Because it's just not just the bid. You're gonna have the estimated action rate and then you're also gonna have a higher ranking, um, because it's gonna be a more engaged with ads. So you're, you're kind of winning on all fronts there. Do you look at any custom metrics like, uh, revenue per click or AOV or

Cody Plofker (22:55):

Anything like that? Yeah. I look at, I mean, I guess click to purchase is custom metric, right? Yeah, yeah. That's yeah. Yeah. So

Rabah Rahil (23:01):

Conversion

Cody Plofker (23:01):

Rate, click, click to purchase is obviously one of the main ones click, click to add to cart. Um, I don't really look at, uh, like add to cart, could purchase a ton. I have it in there. It's probably not a main one. Um, and, and RPS definitely.

Rabah Rahil (23:14):

Yeah.

Cody Plofker (23:15):

RPS AOV obviously.

Rabah Rahil (23:17):

Yeah. I love it.

Ash Melwani (23:19):

So a question if you're doing, so you're saying you're doing like a lot of educational content, right? Yeah. And a little bit more. And I guess the goal is to kind of build that middle, right. You want, you want a more educated consumer making a purchase with, you know, your brand. How are like, what is your like process of at least seeing if that works right? Like how are you measuring that? Like, I feel like before it's a lot easier, like prior to like all the data loss, but like now how are you figuring out if the money you're spending on that top of the funnel for education is actually playing out to your manage later down the line.

Cody Plofker (23:57):

I guess this is my, this is my whole like brand first performance marketing thesis. Like it's still conversion optimized campaigns. I'm still optimizing off of a CPA. But if you give me two ads with a, a ROAS of 1.8, you know, and one of them is just a static image ad to a product. And another one is like an educational video with a, a much better average watch time. I'll, I'll take that one. You know? So if, if, and that's kind of a gut feeling, I don't, I don't that's subjective. I don't even know how I would go about finding that data to be able to see if there's any correlation between like those ad running those ads and like Google brand search later on. Right. Or, or anything like that or overall Mer, you know, cause again, the theory would be no, that's fair.

Cody Plofker (24:38):

They're they're, they're seeing you, they're remembering you. They have a good feeling about you. The other thing I'll look at like a lot of these educational ones go to an educational landing page, which then go to a quiz. And so in triple we can look at cost per email sign up. So for example, one of these is going to an advertorial five reasons why that goes to a quiz. We're getting like $7 emails on purchase, optimized campaigns. So even though our, our return on ad spend is not, and we only launch these on, I launched these on Sunday. So it's early, you know, but even if our ROS is not quite the same, like, or if we are the same, that's a ton of incremental asset building that we're doing, whatever you want to call, call it that I think would, would get a better LTV to C over the long term. You know what I mean?

Rabah Rahil (25:18):

Yeah,

Cody Plofker (25:19):

Absolutely. Like our email list has, is huge. Now. Like I looked last month and hadn't looked in a while and it's like doubled in size just because of how much traffic we're spending to, to quiz that's worth something.

Rabah Rahil (25:32):

And you obviously have flows on the back end, et cetera, et cetera, where these touchpoint are gonna be way cheaper than, uh, proper paid media.

Cody Plofker (25:40):

Yeah, exactly. Hopefully we're trying to, you know, use paid media to build our brand instead of just using it for, for that purchase. But again, that being said, well, this is not brand, this is brand first performance. So we are still optimizing for purchases, we're bootstraps. So we can't just throw money out there and hope it comes back. Like we're still optimizing to a CPA and you know, like, yeah, we're still optimizing to a CPA.

Rabah Rahil (26:03):

However, so you optimized to a CPA as well. Yeah. Cause that's what Ash does. So because you have like an well, you're kind of a young business though. Right? So how do you, how do you nail down that CPA with, uh, LTV?

Cody Plofker (26:16):

Uh, I mean, we're also, I boots scrap to know ashes as well. So like we're not really going off of an LTV. We're we're also just preferences and kind of growth trajectory and organic traffic. We're we're not aggressive, like anybody that would come in here and would be like, you need to spend way more. We just choose not to, for, for several reasons. One is inventory. One is profitability, you know, for sure we're not trying to grow at all costs. So we're, we're really, I guess, optimizing to an Mer more than anything. Yep. That try in there that that's kind of the, the main thing. Obviously we know our contribution margin. We know we wanna break even, or be profitable on first purchase. So we're, we're kind of not going on a, a LTV payback period. We're kind of going off a contribution margin.

Rabah Rahil (26:57):

Yeah. That makes sense. Does that makes sense?

Cody Plofker (26:58):

Yeah. And you can, and again, you can grow a lot faster optimizing off of a 90 day LTV or sure. A 365 day LTV. It's just not what we're choosing to do at the moment.

Rabah Rahil (27:07):

Yeah. Year LTVs. I do not recommend unless you have like a two to three year old business because, um, there's just a lot of fuckery that can get involved with, uh, extrapolations. Especially when you get out to a year, you also run into a lot of macro headwinds that you probably won't be able to predict, but yeah, that's super smart. Yeah, totally.

Cody Plofker (27:25):

I love it. So, alright. So here's my, what I realized this week and tell me if you guys kind of agree with this, if this makes sense or not. So I noticed, and I tweeted the other day that like the better an ad does when you launch it. The worse actually thes looks in Facebook for a while. And my theory is because a better ad Facebook's kind of put more spend behind it, right? Yes. And like at the levels, you know, Ash, Ashley and I are spending, you know, 20 plus K a day. Like when you put something in evergreen it's and it's Facebook likes it, it's gonna get spend, it's gonna rack up that spend really, really, really quickly. But the purchases are not gonna rack up that quick. They're gonna be more and more delayed. So it's like, it's like, you have this Seesaw and it's like in platform.

Cody Plofker (28:02):

Yeah, yeah. In platform. So you have the Seesaw where you're like waiting it down. It's like, you're trying to balance things in the old days as spend went up, you would get purchases. So you could tell pretty quickly now it's like your spends here. Your RO is like, not, it's like not coming in because it's so delayed. So it's like the, and so the other, the that's one and you really have to be patient on, is the stats look terrible? Just like, like, like I'm telling you like God awful. And, and I turned off awful lot of stuff in the last month and few weeks that I, I shouldn't have, and then obviously hit it with the delayed attribution. But the other thing, because now the conversions are reporting on time of correct me if I'm wrong, it's it's time of purchase and not time of impression. Correct?

Rabah Rahil (28:43):

Correct. So it's, uh, the way it's bifurcated is click an event before it used to be click. And so if you say day zero, yeah. You clicked on an ad and then bought on day three mm-hmm <affirmative> day zero would get that credit. Now, if you click on an ad on day zero, but then buy on day three, you're gonna get day three is gonna get attributed. And so in triple, well you have a click in event date, so yeah. You're absolutely right. So,

Cody Plofker (29:06):

So you can, those, those purchases, right? Especially for us, the majority of our purchases don't happen on the first day. It used to be that if somebody purchased it would go, Facebook would go back and then attribute those purchases to those first three days of the ads. Now it doesn't. So overall with that kind of new way of delayed attribution, the overall ROS is going to get better as, as you let things run. Correct. However, right now, if I went back in, even though our overall CPA is decent, if I went back in our first three days would still look awful.

Rabah Rahil (29:35):

Yeah.

Cody Plofker (29:36):

Yeah. So I, to me that that's fascinating. It's just something to keep in mind. It's like things are just going to look awful and the more you're spending on something, the worse they're gonna look, I think

Rabah Rahil (29:45):

Exactly

Ash Melwani (29:47):

This is, can I say something that might, I'm not, I don't wanna to run you the wrong way. This is where I think our first part, the second part we were talking about optimizing based on that one day clip, right? For you now you want that journey is gonna be a lot harder for you to understand what is being attributed to your spend, versus if I'm going for that one day, click and conversion, I don't have that. And I can make decisions a little bit more quicker than you can. And, and I'm saying either one is right, right. You're going for that brand performance marketing first, I'm going for like that direct response, I'm just acquiring to acquire. And like, those are the issues that you're gonna have. Right. And that's, that's all I wanna say about that, but I mean, I totally understand the whole, the premise of like, okay, well I wanna educate, I want people to go down this funnel, but that's kind of why I asked you the question earlier. It's like, how are you measuring that? Are you like, you kind of have to let things run and then look it back at it on like a seven day like window and be like, okay, well, is this good? Great. Let's keep going. But then like, you can't make decisions day to day, which is like, it's tough. Yeah. It's tough if you can't do that for

Cody Plofker (30:54):

Sure. But you're not, but you're also not making decisions day to day either. Like you said, you don't touch it for a week. Oh,

Ash Melwani (31:00):

Whoa. A hundred percent, but I'm getting, I'm gonna get data right away. So that I know for a fact something is working on day one. You don't necessarily know that.

Cody Plofker (31:13):

No, I don't. I don't. I don't. Yeah.

Ash Melwani (31:15):

And again, I'm saying I'm, I'm more so saying that I'm loo I'm not touching it for seven days because then after seven days worth of spend, I know this is a clear winner, but on day one, I know if something is heading in that trajectory, which is the only

Cody Plofker (31:28):

Yeah. And, and I don't. Yeah, you're right. I don't, so I think two things that are important, a you have to have like strong hands to do this. Like you have to like, get computer, have patience, being willing to put 10 K into an ad. And it looks like you're at like a 0.1, it looks like you've made a thousand back and like, you just have to be okay with that. But I also think that's why you also have to be testing in a, in a sandbox campaign testing in an environment where you're controlling the spend and you're, you're willing to essentially risk or lose that amount of spend and only graduating things when you have the confidence. Like, I think I would have a lot less confidence land launching a brand new thing with that budget. You know, I, I don't think I have the confidence to do that.

Rabah Rahil (32:07):

Yeah, absolutely. The other thing too, Cody, you can look at is when you do in triple well, you can actually look at quick date. So, um, to as Ashley's point when you guys are spending at the scale you're spending at quick, date's a much better way to look at, um, the attribution than event date. Sure. Because you're the event, date or event based attribution when you're spending at the scale you are and launching these new creatives, the, the credit could be going to the wrong people. If that makes sense. Like the click base actually is gonna be a better clearer read, um, when you're spending at that scale because, uh, you're, you're gonna pervert the data. I'm looking at event date, because again, day zero is actually what caused the purchase. But day three is what's going to get the yeah. Um, the credit

Cody Plofker (32:50):

I gotta, I gotta do that more. So cuz like, so cuz we're running a lot of, you know, lead gen again, still purchase optimized, but building our email list and we a, we have a ton of product launches, right? Like we launch multiple times a month just at, at this stage in our, you know, our journey, but also we've had a ton of inventory issues. So a ton of back in stocks,

Rabah Rahil (33:08):

Big deal, man.

Cody Plofker (33:09):

So like I know just from, you know, customer feedback on ads, CX social, that a lot of people will comment on ads and be like, this looks great, but I'm waiting until blank is back in stock. So I bet that there's like a big discrepancy between click and event date for that reason. Yeah. And it'll be really interesting to go back to click date instead of looking at event date. Cause like for example, we had a launch today or we had a back in stock today, like today's through the roof. Yeah. And I know some of these customers, we acquired them as at least acquired their attention and acquired their email four days ago or whatever,

Rabah Rahil (33:42):

A hundred percent. Uh, that's kind of, if we can kind of segue a little bit into some attribution models cause Ash and I had a, a pretty interesting discussion on that. So there's this ultimately a few in triple well there's uh, first click last click fractional linear. Um, and then triple well, uh, plus views. And so ultimately the way I think about them is I'm gonna use that first click attribution model, um, when I wanna scale and see what my best prospecting ads are. So what, what are kind of to Ashe's point when somebody has a first click, what caused that purchase on the first click? Last click is gonna be the inverse where I'm gonna scale up my retargeting campaigns based off last click and then triple whale plus attribution is, or plus view through is gonna be kind of your best overall ads, but cover your ears.

Rabah Rahil (34:27):

Cody ear Musk. This has view through conversions. And so you want to take that with a grain of salt and then fractional is just gonna show you fractional linear, excuse me, you're just gonna spread that across the channels and you're gonna see what channels in your ecosystem are actually, um, driving the conversions. And so I think that's a really interesting mental model in terms of how you're gonna scale your campaigns and look at how to leverage those attribution models. Cause I get that question a lot. I don't think, um, until Ash and I started jamming about it, that we had a actual, a concrete answer on that.

Cody Plofker (34:56):

So what are the, what are the pros and cons like, should there be one attribution model for your business or should you use different ones at, at, at different times and what are, what are some of the pros and cons of these different attribution models or limitations?

Rabah Rahil (35:10):

Yeah, great question. So I guess I should probably ratchet back up for people that might not, uh, be hip to it. So first click means it's gonna attribute all the, um, actions to the first click. The first touch point of that journey. Last click is gonna attribute the, uh, all the success to the last touch point of that journey. Um, fractional is gonna, uh, spread that out across any channels that were involved, um, equally. So this is there's different fractional models. We use fractional linear. So it's basically gonna there's four touch points. All four of those touch points are gonna get 25% of the conversion. Um, and then triple whale, uh, plus view through is gonna be last platform touch. So it'll exclude, uh, organic and email. Um, and then it'll also bring in view throughs as well. When you have triple attribution plus view throughs. Um,

Cody Plofker (35:55):

I love those views.

Rabah Rahil (35:57):

Yeah. Hey, Hey, they're real. They don't get me wrong. They're real. But it's just, um, you wanna scale again in terms of the attribution model that will help you understand what is closer to the truth in terms of the participation of that ad. Um, and so kind of to your question, Cody, I think that the problem with first click is that, um, it might not close. And so if you have a longer again, uh, sales cycle, if you look at first click for retargeting, probably it's not gonna be the path because this wasn't the person that got the person close. This was the person that got 'em into the funnel. And so your educational videos might be fantastic and prospecting. And so that's why I like to use first click or prospecting, whereas retargeting, you wanna see last click is gonna show you who's the closer mm-hmm <affirmative>, who's the person that is closing that sale. And so that's more in my opinion, applicable to that retargeting. So you're maybe statics or whatever, but you wanna see that retargeting. And then again, the triple whale plus view through is gonna show you the kind of best overall ads that are working there. And then the fractional is gonna show you what channels are actually causing, um, purchases. So that, or that's kind of just how I think of it.

Cody Plofker (37:05):

So say you're like eight sleep, right? Like let's say you're selling mattresses with obviously a long sales cycle, you know, you would be using for your prospecting first click. Like that's kind of what you'd be relying on because correct. You know, they're probably not buying from that ad directly, but you want to know what's feeding them.

Rabah Rahil (37:22):

Yep. So when I'm, uh, looking at the success of my, when I'm looking to scale my prospecting ads, I'm gonna use that first click attribution. And then conversely, when I'm looking to scale my retargeting ads, I'll switch that attribution model to last click. Cause I wanna see what ads are actually closing people.

Cody Plofker (37:39):

What if you were, you know, I don't know, but what if you're a low AOV product, like a $30 product, you know? Yep. Pretty low sub sub $50 AOV. What, what would you recommend

Rabah Rahil (37:48):

Mostly first click? Cause you're probably not gonna have the opportunity to do retargeting because you're gonna have economics that just don't warrant that. So I wanna see what can cause a purchase or at least get somebody into my funnel and then I can do the code thing and look at my email signups hit 'em on email, but,

Cody Plofker (38:04):

But you wouldn't do last click for that, for that?

Rabah Rahil (38:07):

No. I mean,

Cody Plofker (38:08):

Because it's a shorter funnel or like that

Rabah Rahil (38:10):

Would exactly, I mean you could kind of have six on one or the other. It's more so the layers of where they ads are. So I wouldn't be running retargeting in a low AOV, um, type of business. And so then that makes kind of last click, a moot point. But into your point, it's, it'll be a little bit of six and one half dozen, the other with first click last click, but it's more so where the ads are in the funnel is when I choose to use the attribution model. And honestly, in that sense, I might just be looking at triple wa plus view throughs. Cause I'm just trying to figure out what the hell's driving some semblance of performance because mm-hmm <affirmative>, you don't have that sophisticated of a marketing ecosystem. You're basically an AOB kind of break even at that point. Like, so whatever's getting me on basically ultimately the cheapest first touch points, get me in.

Ash Melwani (38:56):

Well, to that point, what if like, I would imagine first click, I think might kind of match up, right. And that's,

Rabah Rahil (39:02):

It's gonna be very close. Yeah. It'd be very close to your first click and your last click because there's not a lot of, uh, differentiation in the journey and there's not gonna be multiple touch points usually.

Cody Plofker (39:13):

Yeah, definitely. What about tell, when is fractional important? Like when does fractional help?

Rabah Rahil (39:20):

For me, fractional is useful to understand, um, channel kind of contribution. So not necessarily like

Cody Plofker (39:26):

Between paid channels

Rabah Rahil (39:27):

Between paid channels. Exactly. So you're gonna see like when you're you use that fractional attribution, you're gonna see as Facebook driving it is Google driving it and so on and so forth. And so you're gonna see that RO as switch. Um, and you gotta remember fraction will have kind of more than the whole because it'll give credit across all those channels. And so you're, if you have 10 purchases, you could have 20, um, purchases spread out across all these different channels. And so that's why it's not necessarily a great quantifiable attribution model, but it's very good at qualitative where it's like qualitatively this channel's driving and this channel's driving and this one isn't. And so that's why use when I would apply fractional where it's more so to understand at the channel level contribution more so than the campaign adder ATS it.

Cody Plofker (40:11):

So in general, are there any data delays on triple well, like obviously Facebook is delayed up to 72 hours. It's like time of conversion.

Rabah Rahil (40:19):

Yeah, pretty much. Yeah. So that's, what's nice about it.

Cody Plofker (40:22):

And what is like, so on like triple attribution plus view is like, what is the attribution window? Is it, is it seven days? Is it unlimited?

Rabah Rahil (40:30):

It's gonna be seven day view one day click. Okay. Or, I mean seven day click one day view. Yeah, exactly. Like the Facebook does open up, uh, that 20 day, 8, 28 day view or 28 day click one day view window. And you'll be able to see kind of, uh, longer look backs, but ultimately it's gonna be based on the, uh, the account and everybody's account is essentially set to seven day click one day view.

Cody Plofker (40:54):

Got it, got it. Okay. So, so then if you see any delayed, you know, attribution in there, it's because of people purchasing within that window, but outside of like that day. Yeah. It's not as of like a delay.

Rabah Rahil (41:05):

Exactly. And so you will seek again, like longer look back. So if you look at kind of the, the LTBs, you'll be able to say thirty, sixty, a hundred eighty, but again, it's gonna be more so to your point, like we're catching this and then attributing it to it, but it's not gonna be, um, within the attribution window, which is really what you care

Cody Plofker (41:21):

About. Yeah, yeah. For sure. Yeah. Cause, cause again, that's why I wasn't sure. I didn't think there were any delays, but like our triple ROS does continue to get better as, as you know, we launch stuff and it goes on and obviously it's because our it's

Rabah Rahil (41:32):

Just cuz you're so good.

Cody Plofker (41:33):

No, no, it's the opposite. I can't get, I'm not like Ash, I can't get them to buy on the first purchase. So we have to wait a little bit.

Rabah Rahil (41:39):

<laugh> he won't give this up. It's like a dog with a bone. I love this guy. Absolutely.

Cody Plofker (41:44):

I'm saying I'm not good at, I'm saying I'm not as good as Ash.

Rabah Rahil (41:48):

No, I, I love it, but sorry, if we got kind of nerdy there folks, but uh, I think it's really important to understand kind of the, the attribution models of when and when not to use them and um, ping us offline or whatever, if you do have more questions, but are we got nine minutes? Cody has a hard cutoff. He has to go fix, uh, his driveway pavement. They were off four feet. And so he's gonna go out there and fix it up. Uh, what else do we guys wanna get into?

Ash Melwani (42:13):

Um, I think what else was on that list? Oh,

Rabah Rahil (42:19):

Um, but you guys wanna do some, some trivia. I have some, some data. Yeah.

Cody Plofker (42:23):

I got one question for Ash, but then the, and then yeah, so Ash, you're saying, I was just thinking about this week, cuz you're telling me, you know, you don't touch anything in, in your main scaling campaign. Right. And you're testing 40 creatives a week. How are you getting enough budget to test those 40 creatives? Like it seems like either you're spending a ton on creative testing or you're really spreading out that budget to, to cover 40.

Ash Melwani (42:48):

So here's the exact budget layout and campaign structure if this helps anyone. Um, but basically I have two ad accounts. Each ad account is for a separate product. Um, one's like our weight loss product and one, one reading product in each account. I have a scaling campaign around 75 to 10,000 a day. Um, then within that is a testing campaign at a thousand, right? So it's about 10% of the scaling. So I mean it's really 10%. It's only 10% and it typically does drive decent conversions on most weeks that there's weeks where it's just really shit and I'll bring up our average, you know, acquisition cost. But I mean, as we keep making iterations, it tends to start, it'll tend to get better. Right. So as we get closer to, okay, well this angle is working and, and it it's really broken down across like four 40 creatives are broken down across four different angles per product.

Ash Melwani (43:57):

So it's really not, it's like five it's five ads per angle per product. Right. So the thing is out of all those out of like all let's say between the two testing campaigns, there's only really gonna be two or three that get all the spend. Right. Yeah. And I know people are like, well, how are you? How do you know the other creators are not getting spent this and that? Like, I don't care. Like if Facebook says, this is what needs to get the spend and this is what's getting engagement and it's converting, like that's the winner. Like I don't have time to be like, okay, well let's set up an ad set. Let's put this here and this here and this here. And like give every single creative enough budget. Like, no,

Cody Plofker (44:33):

You should do 40 assets. You should do 40 assets. One <laugh>

Ash Melwani (44:37):

I will, I would rather like I would, I'd sell the business and not be dope. Like there's no way that I'm gonna do this every week. I need to give Facebook 40 creatives. Let it decide if it decides this is a winner, great. Put in scaling. Like you just have to make it easier on yourself. Like you can't be like so scientific and like, well, like I gotta have statistical significance on like this adset versus like enough, like just make it simple.

Cody Plofker (45:02):

And so some of those creatives are like not getting any spend really.

Ash Melwani (45:06):

Yeah. And, and the reason it's not is because Facebook doesn't think it's engaging. Yeah. You know?

Rabah Rahil (45:11):

Yep. And, and to be fair too. So I used to, when I ran my agency, I used to do a lot of dynamic creative testing. Um, and that was a big issue because, uh, Facebook was basically a first mover. So the first, uh, in, in dynamic creative. And so like the first thing that hit dynamic creative, which is eat up all the spend. And so what I would do is pull out kind of the winners and then let those other people fight it out. But what Ash is doing is actually way better because you, you, that is optimized exactly right. Where, whatever Facebook likes it's gonna get and it's gonna spread out that, that optimization pretty, pretty effectively. Whereas in dynamic creative, I had multiple winners a lot of times in that. And so I had to pull them out and then scale that and then run that dynamic creative test again. But Ash is essentially circumventing that by giving basically everybody a free chance and it's not gonna be as much of a winner take all issue. Um, as when you do do dynamic creative testing, it's a big, big issue where basically like one thing wins and then you'll ne nothing ever sees any sort of spend at all.

Ash Melwani (46:12):

Yeah. And to be fair there, there's going to be weeks where we don't have time for, you know, the designers to pop out creative. I'll probably go back and find the, the creatives that didn't get any spend and maybe put them into another environment where it's with the other creatives that didn't get spent and then see what happens there right at the end of the day, if something performs and it's, you know, creating a better CPA than something that's in the scaling campaign. Great. That's all I care about. Move on to the next. But like I said, you know, now we have a backlog of creatives that may, may make it to the next round of testing, you know, but that's just how we're doing it. And it's, it's helping us find more stability. And um, I mean our, our CPA's coming down, so

Rabah Rahil (46:54):

That's huge. That's huge. Okay. Let's get into a couple trivia questions and then we'll sign off cuz it's Cody, we're pushing up against it. Okay. Uh, we tracked about what we 578 million in ad spend. Um, what do you think the top spending channel was and how much Cody? You can go first and then Ash

Cody Plofker (47:14):

Facebook.

Rabah Rahil (47:14):

Okay. How much? Five. So 5 70, 78 is the whole, yeah,

Cody Plofker (47:19):

I'm saying 60% of that. I, I'm not gonna do public math, but I'm gonna say 60% of

Rabah Rahil (47:23):

That. I'll do it quickly for you. Whoops.

Ash Melwani (47:28):

That 300 million something

Rabah Rahil (47:30):

Around that. Look at man. That's not bad. Three forty seven, three forty seven four. We'll call it three, three forty eight. Okay. Ash, where are you at big

Cody Plofker (47:37):

Spin channel?

Ash Melwani (47:39):

Cody said 60%. I, I think it's closer to like 75 80.

Rabah Rahil (47:43):

Okay. Let's go. I'll give you 80. Oops.

Cody Plofker (47:46):

So that's yeah, you're probably, you're probably more right.

Rabah Rahil (47:49):

You can't change it. I'm thinking by

Cody Plofker (47:52):

Mine prior on

Rabah Rahil (47:53):

4 62. Okay. Four. Okay. Well, if we're doing prices right rules, you would bust Ash, but you are definitely way closer. Uh, Facebook was 450.3 million, um, in quarter one. Geez. And again, this is gonna be a quarter over quarter comparison, but it was down 26%, but there's some seasonality in there where, um, you know, Q1 is always gonna be, um, less than Q4, but okay. What's the second, second channel.

Cody Plofker (48:23):

Ooh,

Rabah Rahil (48:23):

Google or TikTok? Google.

Cody Plofker (48:25):

I'll give you Google. Okay,

Rabah Rahil (48:26):

Cool. Where you at? How much? Yeah, so 4 27. What is

Cody Plofker (48:31):

It?

Rabah Rahil (48:31):

Yeah. Or four 50, excuse me. Four 50 is gone. So you got,

Ash Melwani (48:36):

You have a hundred million left

Rabah Rahil (48:38):

Hundred 28.

Cody Plofker (48:40):

So I'm saying if it's 1 28, I'm saying like 17 of it went to talk. The rest went to go Google.

Rabah Rahil (48:49):

Okay. So is that 111? Okay. Where are you at Ash?

Ash Melwani (48:54):

Yeah, I think that split makes sense, like hundred to Google. The rest of the TikTok

Rabah Rahil (48:58):

Guys. You guys are a little better on this one. So I'll

Cody Plofker (49:01):

Go. Yeah, I'll go. One 11 Ash. We'll go a hundred.

Rabah Rahil (49:03):

Okay. One. Oh, you're one 11. Boom. You guys are tied now. So I guess the TikTok, well, I guess you can do the math. So the TikTok won't even be the ad breaker, but yeah. So buck a buck 17, basically 116 point 78 million. Um, and again, this is across about 2,700 merchants in, um, 60 different countries. So, um, and that was up 31% quarter over quarter, which is very interesting to me. Um, and then TikTok, even though it is growing like, uh, a weed it's 110% growth quarter over quarter, um, basically came in at 10.9 million. So it's still just an absolute drop in the bucket, which is was

Cody Plofker (49:38):

Crazy. That'd be interesting to see though, like, you know, as you track that, how that's gonna go court.

Rabah Rahil (49:42):

Well, for me, it'll be interesting when we have Q2 over Q1, because Q4 just has a lot of seasonality perversions where it's gonna be one of the most expensive, like it was the most expensive CPMs are way down compared to Q4. So it's obviously gonna be the most expensive and usually the highest scale cuz that's when people spend the most. But I thought I'd give you guys a little, little taste of some fun data here. Um, Cody, we have one minute sign off, tell the people how to sign. I

Cody Plofker (50:05):

Don't have like a hard stop. So I got a few more minutes.

Rabah Rahil (50:07):

Okay. Well I got a piece. I got a hard stop. Well, you got me on the Ker kind of hydration and now I, you know, it's just that you're just look

Cody Plofker (50:18):

How good your skin looks. Look how hydrated

Rabah Rahil (50:20):

You look. I'm. So I'm burnt though, that that Marine layer got me. It was freezing in San Diego, but I didn't realize you can get my hand, got sunburn. Ah, hang was, I was a wreck. I was a wreck. Um, I'm gonna try to get these guys out to Miami geek out. Let's go pees. Um, Cody tell where the people were to follow you. Uh, tell 'em about your newsletter. What else you got going? Your masterclass? Let 'em all know.

Cody Plofker (50:40):

Hit me up on Twitter. Uh, Cody plot on Twitter, Lincoln bio. You can find my newsletter, uh, actually just about to release like a, a personal website. So that'll have a link. Oh cool. In there as well. But uh, yeah, that, and then if you wanna wanna kind of, you know, borrow some of my time on, uh, on mentor pass, you know, you check it out, schedule a half hour hour call with me just to get any, any questions answered. But if not, just hit me up on Twitter.

Rabah Rahil (51:04):

Fantastic. He's a wonderful feed and a great newsletter as well. For sure. Ash, let the people know,

Ash Melwani (51:10):

Uh, follow me on Twitter, Ashton. Melani I'm gunning for 10 K followers hopefully soon. <laugh>

Rabah Rahil (51:16):

Jeez, you savages. That's incredible. You guys are massive. Let's go. Let's get him over the 10 K folks. Come on, come on. We'll give you, we'll give you some swag or something like that. We'll hook you up. We'll get you out there. <laugh> my hobby swags. Awesome. Yeah. You guys launched some really cool stuff.

Ash Melwani (51:30):

Yeah. Some new stuff coming out, but yeah, let's see. Uh, let's see how it goes.

Rabah Rahil (51:35):

Cool. What, no product plugs or nothing. You guys just launched the cool collagen stuff, right? The, the new flavors or anything.

Ash Melwani (51:41):

Oh, we just launched new. Oh, we're launching new gummies this week. Uh, Friday. We're doing super fruit gummies. Uh, so get your fruits and vegetables in and then ashwaganda gummies. So keep going.

Rabah Rahil (51:53):

Oh, love. It'll be good.

Ash Melwani (51:56):

Very good. Cody. How'd you like, uh, the collagen. You

Cody Plofker (51:58):

Picked it up, dude. I loved it. I, I loved it. Mix it, mix it in with some almond milk. Yeah, I got the, I think it was like the cookie one. I don't know if it was that

Ash Melwani (52:05):

It

Rabah Rahil (52:05):

Wasn't

Cody Plofker (52:05):

Donut one. It was enden one. Yeah. Yeah. It was. I, I really liked, it tastes great. Like good looks dessert mix. Yeah. Yeah.

Rabah Rahil (52:13):

And you don't even have to order it on the site. Right. You guys are in what GNCs or vitamin shops where

Ash Melwani (52:17):

You at shop vitamin shop nationwide 600 stores nationwide. Um, just launch two weeks ago. Lets go. Um, so if you're buy one, go check it out. Send me a picture between me a picture. Um, but yeah, just yeah, let's check it out.

Rabah Rahil (52:33):

<laugh> beautiful. All right, folks. Uh, if you do wanna get more involved with triple oil, we are tri triple oil.com. We have a wonderful newsletter that goes out, uh, whale mail Tuesday, Thursday, and then we're on the bird app at triple oil boys. Thank you for another session. Thank you for all the time in the fantastic questions. Cody, happy to have you back. Congrats again on the bun of the oven. Well done, sir. Thank you. Um, and Ash love the hoodie we got, I, I need some swag help here. Uh, we, we are dying for some swag and you, you are the, the man. So I'm gonna pick your brain somewhere on that. Uh, folks, subscribe, share, do whatever you need to do. Spread the gospel farm wide. Um, and that's all we got for it. Another one in the books. Thanks guys.

Speaker 4 (53:12):

Please

Ash Melwani (53:13):

Take care.

ā€

One Tab
To Replace Them All

Supercharge your growth with a purpose-built ecomOSĀ for brands and agencies.

Before ...
Shopify
Ads Manager
Klaviyo
TikTok
Pinterest
Bliss
Triple Whale