Ecom Growth Insider · Podcast episode

Camilo Castañeda: Media Buying Is Thirty Percent, the Creative Does the Targeting

With Camilo Castañeda · Published 2025-05-09 · 43 minutes · Hosted by Andrej Tumachowitsch

Camilo Castañeda: Media Buying Is Thirty Percent, the Creative Does the Targeting

Camilo Castañeda runs Epic Ads Lab. He started as a video editor making high-production product commercials, watched that style stop converting when TikTok arrived, and rebuilt around research-led creative instead. He works with brands spending anywhere from five thousand to five hundred thousand dollars a month.

His central claim in this episode is that the creative does the targeting. He puts media buying at about thirty percent of the outcome: bad buying still hurts, but the right targeting inside the wrong creative cannot work. So the leverage sits in what the ad says and who it says it to.

This is for operators who are testing plenty of creative and still cannot tell why it is not landing. You get where he finds angles, how he splits budget between what works and what is new, the three metrics he would keep if he could only keep three, and an honest read on where AI creative actually helps.

What you will take away

• Most brands advertise what they think the USP is. Camilo's example is a jewellery brand whose ads say the product is waterproof. Only people who want waterproof jewellery watch that. Everyone shopping on design, on feeling good, on status, scrolls past. The fix is to advertise what the customer cares about at their stage of awareness rather than what the brand is proud of.
• One-star reviews are the best research you have. He builds angles from what humans actually say: comment sections, forums, reviews, and the one-star reviews on competitors in particular, because those tell you exactly why somebody chose a different brand. He puts research at roughly sixty to seventy percent of the strategy.
• The other thirty percent is swings. The rest is deliberately trying something new, usually borrowed from an adjacent niche. For a collagen brand he would look at what works in protein, in creatine, and then further out again, and remix it rather than copy a direct competitor.
• Branding is something you earn with direct response first. Camilo's rule is to prove market fit with direct response, get profitable, and only then spend on brand. Good branding lowers your cost of advertising over time, but it is expensive, and small brands routinely spend attention on logos and on-brand polish before they have proven anything converts.
• Three metrics, if you only get three. Cost per result, click-through rate, and hook rate. He would swap click-through rate for unique outbound click rate once a brand is scaling rather than testing. Early on the question is simply whether anybody is clicking at all.
• Ad fatigue is mostly a limiting belief. He has seen ads stay profitable at a frequency around 2.8. His view is that you will not saturate a whole market, that every market has sub-segments, and that being afraid of fatigue is really a creative supply problem. Make more angles rather than switching off something that is still working.
• Split the budget seventy thirty. Seventy percent of spend on what is proven, thirty percent on new tests. For a brand under ten thousand a month he suggests one hundred to one hundred and fifty dollars a day as a floor, spread across around twenty creative angles, while accepting that is a thin spread.
• AI creative wins have not scaled yet. He has had AI-generated creative produce results, but none that scaled past what a real person delivers, and the ones that worked were edited so the AI seams were cut around. His read is that AI belongs in the testing round to find the message cheaply, then you bring in real creators once you know what works.

Jump to a chapter

• 00:00 - From video editor to Epic Ads Lab
• 02:48 - High production or UGC, and why polish stopped converting
• 04:31 - The editing community, and where he hires from
• 07:48 - Where creative strategy goes wrong
• 09:58 - One-star reviews, and where the angles come from
• 11:33 - How he decides what to test next
• 12:39 - Branding is something you earn with direct response
• 16:24 - When nothing works: creative, offer, or site?
• 19:22 - Three metrics, if you only get three
• 21:30 - Which creative styles work at which stage
• 24:37 - Media buying is thirty percent
• 26:50 - Creative fatigue is mostly a limiting belief
• 29:34 - Seventy thirty: proven spend against new tests
• 32:58 - What AI creative can and cannot do yet
• 37:48 - Would you accept an AI ad from Nike?
• 40:27 - One piece of advice
• 41:37 - Does the founder have to be on camera?
• 42:27 - Where to find Camilo

Full transcript

Lightly edited from the episode captions for readability. The episode is the record - where they differ, the audio wins. Lightly edited from the recording captions, which carry timings but no speaker labels, so the transcript runs unattributed.

00:00 - From video editor to Epic Ads Lab

Today I'm here with Camilo. Camilo is an ad creative expert with hundreds of thousands followers across all of your social media channels supporting a bunch of e commerce brands and helps them scale using hyper targeted ad creatives and at the same time also has one of the largest, if not the largest video editing communities on this world. Today we'll basically be going deep into what actually works in 2025 when it comes to e commerce ad creatives, what mistakes brands are making, how to scale using at creatives, UGC and also where the industry in general is headed. Hi Camilo, and to kick things off, do you just want to give tell us a little bit about you and your background? Of course.

Thank you for that intro. I love that. Yeah. My name is Camilo. I run an agency called Epic Ads Lab.

I used to be a video editor whilst I was trying to become an engineer. So I became an engineer and I worked in industry consulting and then I, I was an editor making YouTube videos, showing what I was learning and that led into doing product commercials which led into working with brands, then which led into building an agency and now we're here just making videos every single day. Nice. That sounds fun. So was it like your, your plan to get into, into e commerce or did you just end up there?

I. It was a bit of a 5050 because I, I think I always wanted to be, I actually wanted to be a YouTuber. I wanted to be like a, like a YouTube creator that made cool videos and people go wow, that's a cool video. And then I realized my or the part that I wanted to get to was to make these really fancy product commercials, like really nice looking studio lights, camera angles, like I know everything about how to use a camera and a gimbal and whatnot. And then I was doing that for a bit and it bought up my following.

And then I was getting e commerce brands saying hey, can you do this commercial for my brand? So I was getting products shipped to my house and during COVID this happened a lot and then we kind of went into doing loads and loads of that until I think TikTok came out and then product high production videos just, just didn't Convert as well, because everyone was doing that UGC type organic. And so someone will see something that's really polished and then they'll scroll past. So I had to pivot and then that's kind of how we've been. But like, I think there's, it comes in waves because now you want to have some high production stuff, but we can get into that in a bit.

Yeah, yeah. And that was like one of my next questions, like, are you still doing some high production things or is it now like purely ugc low, low production?

02:48 - High production or UGC, and why polish stopped converting

It depends a lot on the brand. So if we see that a brand has ample UGC or ample organic type videos and maybe we, we think that we should try something high end because it promotes a certain desire, then we will try it. I think overall most brands just need a lot of diversity, a lot of different types. Because if you go on my feed on Instagram, there's a lot of highly produced videos, you know, filmmaking, high production commercials, blah, blah, blah, blah. That's what I like.

But if you go on to my girlfriends, hers is like, you know, girls on the day vlogging, like super low tier. So that's where you need to have both so that the high production one show up on mine and the low ones show up on hers. And so we, it depends a lot on the research we do for the client and whether we think it's going to be a good fit. Yeah, that makes sense. And which of those do you personally prefer, like on the one side in terms of like creating them, but then also in terms of like seeing them, watching them?

Oh, that's a good question. I think I like to make, I like both high production ones is a lot of work because like you have to get things, a lot of things right, lighting and as we scale it, we have to get teams to do them. UGC is a lot easier but is very technical in the research process. You have to be very technical with what words you use, what kind of person you find, et cetera. So there's good aspects of both.

We do, I'd say about 70% organic type content, 30% of the high produce stuff.

04:31 - The editing community, and where he hires from

Yeah. Okay. And is your goal or is your dream still to become a YouTuber or have you like shifted away from, from that? Well, weirdly, weirdly I became a YouTuber because I was on, on my main channel, I was teaching everything that I was learning and that's. And at one point I started doing like car videos, filming cars and then my channel exploded.

So I, I had, you know, at One point I was getting half a million views a month for like six months straight. And it was, it was crazy. But I kind of, I got the play button, as you can see, and I got a lot of fame then I was like, I don't really want to do car videos. Like I kind of like doing the E commerce stuff. I'm getting like jobs from the E commerce stuff.

And what I did switch it to was building my editing community, which has helped me a lot to find good talent because I think good editing talent is really hard. But yeah, basically became a YouTuber. Not like a Mr. Beast or like a super creator, but YouTube is a big part of what, what I do and it's taken a bit of a step back. But yeah, I think I, I did start a new channel with the ads for.

The Ads channel for, for getting ad clients. But yeah, I guess hopefully I have a second play button. Nice. That sounds cool. Yeah.

And do you still focus on also like growing the, the video editing community? Yeah, yeah. It is a, it is a big aspect of what we do. So I run it with a guy called Jack. He also has another YouTube channel.

We just started running ads to a paid product with an upsell to the community just to kind of hit that cold traffic because we kind of stole that our own personal brands. But so far we have about 700 members. We have two different tiers and we want to grow this to like a ousand member thing plus. And this community, I assume it helps you a lot with like creating the, the ads for the e commerce brands. Finding like good, good editors.

Yeah, for sure. It really keeps me ahead of what I have to stay ahead of what's working on social because we're showing editors, hey, this is what's working. This is what you need to learn. So if I'm very behind, then I'm telling editor stuff that's dated. And when I need to find talent is super easy.

I just go like most of the people that I've hired in the agency have been from VP plus and it's, it's mainly because these people are, you know, they, they're kind of thankful from what we teach them. Very hard working, which is very hard sometimes to find someone on upwork or fiverr who's just like dedicated to be a good editor and really hone their craft. Yeah, yeah. And I totally get like, like a lot of brands are really struggling with like finding high quality editors, creators, all of those things. And you have a huge advantage there just on the one hand because you have huge following and also have like the.

The authority but also a lot of people like following you who are interested in that. But then also you have the community where you can just basically pick the best editors and then ask them whether they want to work with you together.

07:48 - Where creative strategy goes wrong

For sure. Yeah. Absolutely. Awesome. But now let's get more into the E commerce side of things.

I mean you worked with a lot of huge E commerce brands and helped them with their creatives nowadays. What would you say? What do you see in terms of the mistakes that a lot of brands are making? Like why are there. Yeah.

What is wrong with their creative strategy? I think it. I guess it depends how you define the strategy because there's like there's the front end and there's the back end. They both do work together. So if you have a.

You have a very good ad but a terrible landing page doesn't matter. It still shows up as no conversions on the ad account. But let's. If you focus mostly on the attention acquisition part of the the of the creative strategy it would be. A lot of brands think that people.

They assume people care about them in the beginning so they would. They'll. They'll do things like generic messaging or care about their brand logo and image not realizing that. 0001 of the population is seeing their stuff. They think that oh they're going to spend this much like 5k on ad spend.

They think that it's going to be a lot of people to see it. But that's not really the case. Maybe the impressions kind of give them that. That idea. So I think the biggest issue is that they don't focus on the customer.

They don't focus on who they're serving because everybody's self interested. Everyone cares about what was in it. For me you know if I see a jewelry brand and the content they're making is it's just talking about how it's waterproof. It's like that's only. Only people that care about waterproof jewelry are going to watch that.

But if someone just cares about designs or to feel good or for status seeking scroll past and so they'll try that message because that's what they think is the USP of their business. People don't really care. And so what I think most brands need to do is really care about what that customers and potential customers care about at different stages of the marketing sophistication level. Yeah.

09:58 - One-star reviews, and where the angles come from

And how do do you or how do the brands find that out? A lot of research. Our biggest lever that we pull is how we do the research. So how we find angles, ideas, things people say, we find what humans say. So people in the comment section, on forums, reviews, one star reviews tell you everything that you need to know about a product, especially in, in the competitor brands.

Because some people might choose to go to a different brand because they left a one star review saying oh, I didn't like the fact that this product did this, blah blah, blah. So going from that kind of information is good. And then I would say that's about 60, 70% of a strategy. The other 30 is just taking swings, just like trying something new, just trying something you saw from a different industry, a different niche. We, we like to find things that work in a specific niche and then try to find things that work in adjacent niches.

So say you're, you're a collagen supplement brand. We will find things that work for other collagen products and competitors. But we'll also see what happens, what works in the protein shake supplements, in the creatine supplements. And then we also look at maybe like green tea or some other stuff, see what works for them and then see how we can implement and remix it. So I would say those are the two things that most brands don't do.

They just put a bunch of stuff they think is going to work or they just copy somebody else and then they're like why is this not working?

11:33 - How he decides what to test next

Yeah. And do you have like a specific framework that you follow when it comes to like testing those, those different angles or pain points? Does it really like depend on, on each individual case? So we do have some winning angles and creatives that, that we will, let's call it, repurpose with the information that we get from the research. So let's say, let's say we have an ad for a, so we have, we have a luxury calculator.

Okay. And we have winning concepts for other tech products. We will find them, we will obviously remix it and use the research for this mainly, mainly based on the concept idea. So whether it's going to be, I don't know, a street interview or it's a pain agitate pain or it's a desire driven type creative. And then what goes in it, what people say, what words are used, what copy comes from the research.

Yeah, that makes sense.

12:39 - Branding is something you earn with direct response

And you already briefly touched on the branded branding ads aspect versus what I would think, what I would say is more like the direct response approach where you would call out the pain, call out something specifically. What is your take on that? Are there cases where brands should do more Branding or branded ads and direct response or is it all direct response? No, there is a case for sure. I think brands that need to focus on branding, logos and all that stuff come when you hit a certain amount of spend and a certain size because branding, ultimately good branding really lowers your cost of advertising because somebody will come back because they like the brand or somebody will hear about the brand or did you hear about like, like Udi for example?

That's a very good brand. Like you know it, you, you've heard it, you know, it's something that people have. You will see that same product in different brands but you call, call it an Oodie. The same thing as like here we have like a type of vacuum, it's called a Henry. So like, like we, that's super strong branding.

Like even something that looks remotely like it, you'd be like, that's a Henry. So I, I would say only focus on branding if it's, if you've hit and you've proven a market fit with direct response, you can be profitable and then you can spend money on branding and you can spend money on creative that is more brand orientated, that is maybe more for maybe you don't get the highest cost per result, the highest roas, but it's there. So people can kind of like McDonald's put these signs everywhere. It's just to remind you that they exist, that they're, they are. So what would the big, the bigger brands do?

Like it works for them. You know, you get reminded that oh yeah, I could eat McDonald's, I'm going to love it. Yeah, totally agree with that. I mean we also work with brands from all sizes pretty much. And I notice back in the days I used to think that direct response is the only way to go and that's what every brand should do.

And I was always thinking that if Nike, if Apple, if they would do more direct response, they would have better results. But now I slowly start to realize that at a certain point, at a certain stage it just makes sense to go with a, with a branded approach. Yeah, but I think most brands overestimate it and do it too early, too quickly. Agree. I notice that especially with like the, the smallest brands or the smallest clients that we, that we work with that they usually put way more focus and attention on that all the creatives look on brand and look nice and have the logo.

When in their case they shouldn't care about that at all. Like for them they should focus everything on getting conversions, getting sales, proving what works and just testing out what works. But after you've reached a certain point, then it also makes sense too. Yeah, I think it comes down to can you afford branding because it's expensive, because it's reminding people of, of your specific business and how they should act towards it. And then I think it also comes down to how unique is your product, where it has to have a good brand because, you know, everyone knows about dropshipping.

Everybody can find the same person, the same manufacturer. If you manage to take a product like say this calculator and then you install chat GPT API into it and you're the only one that's done that, you know, being branded as one of the first or the second ones to do that is, is the most important thing that you can do.

16:24 - When nothing works: creative, offer, or site?

Maybe an early, early stage, but I think it's really rare. It only happens to those brands that can rely on its uniqueness. Yeah, I agree. And now let's say you, you work with a brand and you test a bunch of creatives, do a lot of like, different variations, angles, scripts, whatever, and nothing works. Like, nothing, nothing gets results.

How do you know whether it's like an issue with the creatives or whether it's the product, the website, the offer? Yeah, great question. So we will look at different metrics. If we've tried all the creatives that we thought were going to work and it didn't, we would look at conversion rates from ad to site and kind of see if that's the issue, how many add to cards but no purchases. Maybe get some Shopify stats as well and then we'll see.

Maybe, maybe the ad isn't aligning with the website. Maybe there's something on the website and the offer doesn't look good enough. Maybe it's AOV problem, maybe the AOV is too low or maybe it's too high for what it is. And then we kind of compare that with what we see from other, other brands do. On the creative side, we will look at.

Okay, is it that these ads just the hook rates suck? Is it that people aren't watching it? Is it that it's not getting any clicks? There's a lot of metrics you look at. Typically, I believe that there's no wasted ad spend because you still get data.

You still get data whether it's good or whether it's bad. So we just take that and we kind of model it and we look at what went wrong, what we try next, and then we'll iterate and then if again it's ran, it doesn't work then there's some other issue going on that we have to look at. Kind of like I said, you have the front end and the back end so they have to work together. Yeah. And are there also cases where you support the brands with the back end or you tell them the offer doesn't work, the page doesn't work?

Yeah. I mean I'm not a CRO expert and typically I would recommend other CRO people for it just because I know they'll be the best at it. But I can make recommendations as to like look your AOV needs to be higher or you only have what you only sell in one product and you can maybe add a bundle. Is there something that we can do? Is it.

Can you put the two best selling products together and make a bundle that typically huge swings. Or maybe we just say your prices is too low. Like there's like five of the brands, they are way more expensive. With the right ads they come into your site. Like we can probably make this work a lot better.

So we do provide some recommendations. Yeah. Nice. Yeah that makes sense. And let's say you could only look at three metrics when you're analyzing creatives.

19:22 - Three metrics, if you only get three

Like which ones would you look at? Oosh. Three metrics probably cost per result I guess at what stage of, of the of the business is it at? Because I was gonna think definitely a click through rate but it depends what stage the click through rate. So say it's just like early testing.

I'll just look at click through rate all because I will tell you is anybody even clicking on, on the, on the video opening the comments like there's anything with the ad or the interacting. If it's later like a highest scaling stage I'll look at you know, unique outbound click through rate instead. But so it'll probably be cost per result click through rate and then hook rate would be the three I think are the most. You had to only choose three. But what about you like cause you, you run a lot of ad accounts.

What do you look at? Great question. Just pulling the UNO reverse card also depends on the brand and the stage. I would say either CPA or return on ad spend depending on the brand. Usually the CPA or the cost per new customer, cost per acquisition, then the link click through rate and the third one if it's a video I think also the hook score or the hook rate.

Yeah. If not potentially like the CPM or the cost per click. Yeah but it's tough, it's tough to only look at three metrics because they all tell a story, you know, that saw a specific story that it tells.

21:30 - Which creative styles work at which stage

Yeah, yeah, it does. And when it comes to the advertisement, like the what kind of styles of creatives do you recommend or do you implement for brands? Because we have like ugc, we have statics, we have founders ads, expert ads, influencer ads. Like, what do you see work best and has like the, the highest ROI on average? I think it depends a lot on the stage of the brand.

Let's say you're a 0 to 10k a month brand. You wanna, the most important thing is you want to see what works in terms of the messaging. What's the message that say. You're saying a supplement brand, your supplement brand for women to reduce bloating. Is it that I hate how bloating looks?

I have bloating feels. Is it that I want to go to a toilet more often? I don't know like which is the messaging first? I think that's the first, the first step. And you can do that with, you know, a handful of UGC concepts, make it really easy.

I would say, you know, 10k to 50k a month. Plus if you're a brand that has a strong founder, you definitely need some founder ads. I see that as a way to build a cult following. Build. There's a lot of people that resonate these days with a lot of authenticity, especially in ads.

And they're looking for those ads that are, oh, here's the, here's the founder. He's getting on camera like he's talking about this. I want to follow that. I want to, I resonate with this especially in this day and age, which is a lot more entrepreneurial. So I'll say I would definitely put founder ads in, in that stage.

Statics are always really good for the bottom of the funnel. Although we have had one ad account scale ridiculously with one static ad, it does depend on the product. I would say for ugc, I think UGC is too generic. I think there's a, there's the type of UGC you use because you can have like a video sales letter type ugc. We take somebody from generic like unaware stage down to like super aware stage.

So I think you'll have to categorize, categorize it based on, on the awareness stage and then the kind of concepts that you do. I think three interviews work really well. They're very good type of creative. We have had really good success with a few of them. I think, I mean AI voiceovers with the right visuals can cook really well.

I'M a big fan of those, especially once we know the messaging. And then I would say higher produced type content works well for educational products that need, they need education. Like if you're, if you're like a tech product, like you need something that looks really pristine, clean, but also explains the techie stuff for those people that are techie. So I, I really, that's always, that's what we say. Hyper targeted ad creatives, because they have to be targeted.

The content does the targeting.

24:37 - Media buying is thirty percent

The, the media buying is like, I'd say like 30. Because if you have bad media buying, it still makes, it makes a huge impact. But you only get so much if you have terrible creative because if you, if you put the wrong target in, it's not gonna. The wrong creative, the wrong targeting in the creative is just not going to work. Yeah.

And how important do you think is the communication between the media buyers and the creative team? Or should it be the same person or the same team working on both of those or. Yeah. How to ensure the best results? I think a media buyer who understands creative is going to be an absolute killer.

So a media buyer who understands like this video would do well because I saw it work in another account or whatever. That would be the ultimate Gigachad level of marketer. I do think a happy medium is you have the creative team, you have the media buying team. They meet once a week, they tell us what's working, they tell us what they think, we tell you what we think and we come up with a plan. I think that's the best happy medium.

But we have worked with some brands who the media buyers were doing stuff that only worked in 2018 and it makes our creative look bad. So we started making a creative testing playbook. So it's just everything that we've learned from our experience and other media buyers, even stuff that you have taught me, I've written that in the playbook so just so we can cover all the bases. Nice. Yeah, that makes sense.

And I mean, I assume it's very frustrating if you have like very, very good creatives or you, you. Yeah, you create a bunch of creatives and then you see that the, the media buyers are just messing everything up and they're using like outdated strategies or not. You're not using the creatives the right way. Now the worst is when they shut down the creative because the one that got the most spend didn't get any results and the other ones didn't get any spend yet. I'm just like, why do we do all of this.

26:50 - Creative fatigue is mostly a limiting belief

Yeah, I know that feeling. But yeah, no, that's great. And I'm curious to hear your thoughts on creative fatigue because I know that some of our clients, they're very, very afraid of creative fatigue and they in my opinion, want us to turn off ads too quickly because the performance of the ads is still. The performance is great, but they think that everyone has already seen the ad in the target audience and they are really, really afraid of annoying someone and they don't want the same person to see an ad multiple times and therefore want to turn it off. And then we have clients on the other extreme that just don't provide enough content, not focused enough on creatives and having enough assets where we just have to use the same creatives for multiple months, sometimes even a year of using the same videos and just like repurposing as much as we can.

Yeah, yeah. So what are your thoughts on, on that? I think, I think the first one I think is a limiting belief. I don't think you're going to saturate a whole market. I think every market has sub segments and I don't like.

I think it's fine if you have a frequency of above 2. I think it's fine because we have seen some ads are still profitable with frequency of like 2. 8 or something ridiculous. I think it's comes down to it's a creative problem. Like if you're afraid of ad fatigue, just make more ads so that you can and make more angles and change it up so that if I don't think anybody gets annoyed when they see a fresh take of something, I don't think so.

And I think, you know, if, if ad fatigue was, was as scary as they make it seem, then every single brand would, would be having those catastrophic problems. But in reality is the brands are continuously iterating on creative and just, you know, just making sure that the metrics stay good and don't have that limiting belief. Do the best. I think for people that don't make new creative like they're just miss. They're leaving a lot of money on the table because you know, if it's just.

I think that's worse because you're not allowing other people who consume content in a different way to see that stuff. I see content as more like, like a slot machine token and but the token works on different slot machines and say you have a static. That's like one token. You put that on one slot machine and you can win 0 to 10k in dollars and then you have the other one.

29:34 - Seventy thirty: proven spend against new tests

And you can have a different amount of earnings. It just depends on the machine you use from the tokens. Yeah, that makes sense. And how do you kind of find the balance between reiterating what already works or angles, ads, scripts, whatever, that have already proven themselves to work versus testing completely new things? Yeah, that's.

That's a very good question. That's something that we still work a lot on. I would say at the moment it's about 30% of stuff is fresh, new swings that we, the creative team, just like have that, oh, what if we try this? But the majority of stuff is don't reinvent the. Don't reinvent the wheel.

Try something that's worked. We just know the research is going to do a lot of the heavy lifting. Because what we really want is somebody to look at an ad and have that feeling that Facebook is spying on them. They're like, how do they know I have that problem? Like, how is it?

That's what we want. And when we take new swings, we have to make sure we're using the context and the research really well. It's just how we demonstrate it. So for some brands, a brand new swing might be taste testing different toothpastes or rather than just doing what everybody else is doing. And it's just someone talking about why it's so good and why I replaced my old one.

And when it comes to the budget, do you have any ballpark or recommendation? How much percentage you would like for a brand to spend on stuff that they already know that works, and how much should they spend on testing completely new things? That's a good question, actually. So I would say if you're less than 10k a month, hundred to $150 a day minimum, you can test that between like 20 creatives, 20 different creative angles. It's not the best because you're, you are diluting.

But I know that some brands are starting very small. So I would say you start there and I would say maybe, probably keep it. 70, 30, 70% of your spend goes to things you know it's gonna work and it's proven. And then the other 30% will go into testing new things. And if obviously something new that we tested works really well, you can graduate that to a scaling campaign.

And you know, that's, that's good. It just depends on. That's like more of like what you would do. That's your speciality. Yeah.

I mean, it, it also depends on how much you already have. That works. Like, obviously if you're just starting out, you don't have anything that works. And you obviously have to spend the majority on, like, testing new things. But once you're at a point where you have a bunch of ads or angles that you just know that they work, then it definitely makes sense, as you mentioned, to have like 70, 80% of the budget towards what works, but then still make sure that you spend at least like 20 to 30% on testing new things.

Yeah. Because otherwise, at some point, the things that you that you know, that work, they will saturate or they will fatigue. And if you don't test new things, you'll never know when you're able to find something better and something new.

32:58 - What AI creative can and cannot do yet

I agree. Yeah. Yeah, 100%. And I mean, nowadays, especially in the last couple of months or one, two years, a lot of things are changing with AI and with, like, fake UGC creators, AI generated creators, AI generated ads. How are you seeing that trend moving forward and how are you kind of like, adapting to that?

This is a great question. Very. There's a lot of buzz about this right now. I'll just tell you my entire experience, just flat out. So we have tried a lot of AI U GC, a lot of that ChatGPT image to Sora Video person holding the product.

It is a very much hit or miss. Like, we have had some ads that are just doing really well, and we're like, this is great. But also we edited them in a way where any. At any point it looks fake. There's a cut or there's some B roll or there's something.

Because the majority of people, what they do is they just. They go to, let's say, arcads, they get the AI ugc, put a caption, post it, and they're like, why is it not working? It's like, well, because it looks too much like AI right now, so. Or they try the AI video stuff and they're like, why isn't it working? It's like, well, because it still looks too much like AI.

So you still have to take that footage and you still have to edit it in a way, have good pacing, make it look as raw, organic, or make it look as polished as you need to. So you have to be very careful with that because obviously some people, not just boomers, will see an AI video and go, oh, this is real, or, oh, this is fake. It can happen on both sides. But I would say we've had some winners from it, but they haven't scaled. Like, they've got some cool results, you know, purchases or whatever, but none of them have really scaled past what a real human does.

In my opinion, the future is not an AI. UGC versus Real. UGC or AI versus Real, I think is. It's. It's an and not an or.

In my opinion, I think nothing will be a human being who can do a challenge video or go and speak to other human beings. And I think most humans seek that human interaction and seeing another human on the other side, even if the AI gets really good, I think people will create that. Also, I think there'll be regulations. I think, you know, something bad will happen where there's the need to have regulations. And I think the majority of the time AI UGC will be used for testing.

So you have like, hey, let's do a round of AI tests, find the messaging. Great. We've saved ourselves on costs for creators. Now we know this is working, let's use the creators. Now we know the creators are working, let's do some whitelisting.

Now we know whitelisting is working, let's do brand. That's probably the process that will work. Yeah. So you think creators will still have a job? I think so.

I think it'll be competitive. I think what it will do is the same thing will happen to them. What's happening to video editors right now? Like three years ago, you can be an average video editor doing Alex Ramosi captions. Now you can't.

There's an AI that does it. Sub magic insert AI software that does it. You have to be a story, a visual storyteller. You have to get good. You have to increase your skill as an editor.

You have to know when to cut, what footage to use. It doesn't. It can't look like a bot. Edited this. And I think the same thing with ugc.

If a UGC creator looks too much like a bot and it's just saying things that sound too much like ads, why would I use you? I should use somebody who looks more organic or is a good storyteller, has a good presence. I think creators will have jobs. It'd just be more competitive and harder for you to stand out and harder for you to just be average. Way harder.

Yeah, yeah, I agree with that. And I also think that AI will obviously become better and better over the next couple of months and years. And at some point it will be very, very difficult to distinguish a real human being versus an AI avatar. But I think in the near future, it will be primarily useful for testing a lot of things. Just because even if you don't have a huge budget if you don't have a lot of creators, you can just spin up some ads and test like 20, 30, 50 different hooks, different pain points, different messaging, and just see what at least resonates a little bit better, a little bit more, and then use that as a guideline to create real ads or to hire creators.

37:48 - Would you accept an AI ad from Nike?

Yeah, but also think about this, right? How would you feel if Nike did an AI UGC ad? Depends how it looks like. I mean, on the one hand it would be weird because they can afford it. I think Coca Cola did this one really bad one where they got completely roasted and ripped apart.

Exactly. Yeah. Yeah. I think if you get to that we talked about earlier, that's that branding stage. Like, why are you doing like.

I think AI is. It's more of a means to solve the problem, which is advertising is expensive. Let's do a cheap way to test it. And I think it will be associated with maybe brands that are like early stages, if they get caught or if it's. It was too obvious.

But I think, yeah, like how Coca Cola got roasted completely, you know, you'd almost be like, why? What are you doing? Just pay somebody. Yeah, yeah. I think moving forward, it will be more interesting if brands use AI, but then they also, like, show that it's AI generated.

Or maybe they will have to do it and there will be like, restrictions where they have to say that it's AI generated. But I think you could create like really cool ads by mixing AI stuff and real stuff and just using the different facets, but not just completely using AI to try to recreate something that a creator could do. Yeah, exactly. Like something cool. Put a shoe in a cool environment.

Or if you're selling, I don't know, like a baseball bat, make the fire go into a ball. And that's AI And I think that's fine because it's fun and it looks cool. It's going to allow you to be more creative with the content. As you know, CGI exists and we see ads with CGI and we don't go a cgi, that sucks. The CGI is good.

We go, oh, that's a really cool ad. I love how they did that. So I think it would just make the barrier to entry for a lot of brands a lot easier. I think we are both way too young for that. But do you think when movies and the industry started using CGI there were like, people against that or afraid that it will make things not real?

Yeah, yeah. You got to think about. I mean, I don't even know Because I mean, you've probably seen a movie where the CGI was a bit weird and you're like, this looks fake. Yeah. And it kind of tarnishes the, the image.

But the story is good, you know, it's fine, can't complain.

40:27 - One piece of advice

Yeah, agree. Now the last question that I have for you, if you just could have one, give one piece of advice to an E commerce brand founder looking to scale this year using ads, using creatives, what would it be? Don't look at yourself as a brand trying to make content or make ads. Look at yourself as a media production company that monetizes with a product. I would say that would be a belief shift and a perspective shift.

Like think of yourself as how Mr. Beast sells Feastables. Right. Rather than it's feastables. Like if you, if you look at it that way, then media, which is the game currency, is the new.

Sorry, attention is the new currency. You have to start with attention. So you have to think about the media, you know, media production house. What you sell though is your product or service. And do you think E commerce brands nowadays need to.

Or the founders also need to build a personal brand?

41:37 - Does the founder have to be on camera?

Not even they need to, but I think it helps immensely, especially for that kind of branding aspect. I think you can get really far with just founder led content. We've seen it a lot and that's why it's getting popular. Because you'd be like, oh, that's the guy from, he was the founder of this thing. But I think it, it that only works if the product is unique enough.

I, I don't think it's gonna, you know, if there's two candle candle businesses and one as a founder that's like I made a candle because I like can, like is that like you don't care? But if it's something super unique like a, a drink that has certain chemicals that make you sharp and whatever that hasn't been done before, then you go, okay, cool. His story was he couldn't focus and whatever and blah blah, blah.

42:27 - Where to find Camilo

So then you get sold into that more. Yeah, that makes sense. Awesome. Yeah, this was great. Really, really valuable.

Thank you. Where can people follow you and get in touch with you if they want to learn more about you or about Epic AdSlab? I think Twitter and YouTube are the two best platforms. Twitter is Cammy Sees and YouTube is Cammy's Ad lab. And yeah, I post a lot of videos, make a lot of tweets.

Thank you, Andre. Yeah, I'll definitely follow you. I do follow you, but I would recommend everyone to follow you. Thanks, man. Appreciate you.

Awesome. Thank you. Sweet.