Inside a New DTC Brand’s Ad Engine, Run in 6 Steps: Operators Build (E10)

We're gonna show you stuff today. So if you are listening to this, this is a better episode to watch. We're gonna walk you through the Winks creative supply chain. Um, we're gonna tell you how we're doing it as a new brand. We're gonna contrast that with how you would do this at some larger scale since Cody and I are used to operating at some larger scale. So I think we can give you guys decent perspective. And if you like what you're hearing and you want some of these resources, I would suggest that you go to 9operators.com. Sign up for the newsletter. It's free. The guy that runs this whole company, Aaron, I'm sure some of you know him, loves to give this away whenever we have something cool to show. So that's your way to get it. When you hear creative supply chain, what the hell does that mean in your head?

I love the term. I mean, the same way you can't sell product without the raw materials, the packaging, all of that stuff that has to happen that is essential to the supply side of your business. Um, creative supply chain is the essential ingredients of the demand side of your business. And so if you want to generate sales, you need a ton of creative, but it doesn't just happen overnight, right? It doesn't just show up as a live ad in your ad account. There is a pretty lengthy process that has to happen to have winning and successful ads.

For people who are watching, I have shared my screen now. I created a visual of our creative supply chain as a new brand. And I would say this, like, this is not to be prescriptive. This is not the only way to do things. This is not the best way to do things. I'm sure everybody has an opinion. on how they approach this. And we're not even talking about like measurement here. So, right. So like that part, I'm just not even on here, but I wanted to create a visual of like how I think we think about making more ads or in our, I guess like more messages. Right. So Cody, I think the main thing, like at least where I start, and this might just be copywriting background, I tend to not start from like a format or a concept, I usually start from a message, right? So like in Wink's case, there's like the melatonin hangover, there's the 3:00 AM wake up, there's the cortisol spike. Like there's all of these sort of triggers, these messages. And I would even call these like, these are the why nows, you know, like why am I gonna buy and why am I gonna buy right now? It's like, did you just experience a melatonin hangover? That's like a why now. It's a very like current pain. So first step in creative supply chain for me is message. Like what angle, who am I talking to? All of that stuff is like step one. And to contrast this, at Pela, we only had 4 that we spent time on. Like we really had 4 major sort of buckets like made in Canada, uh, microplastic-free, compostable. Uh, there was one other and I forget what it is, but every piece of creative basically slotted in under those. And we found lots of ways to vary it up. But the way I would describe creative supply chain is like, if you're looking at this visual, right, there's 6 steps and really the first 5 is how you get a lot of variety in your creative. Would you agree with that statement, Cody? Like, this is like combining these is how you get variety.

Yeah, it's, I mean, it's the most important. So like we could just bucket that as like research, right? But like, that's, I think the most important part that how I always vision is like you have a few different personas or avatars and under them you have a few different like angles or pillars. And you know, in the beginning you don't know what's gonna work. You hopefully stack them. But if you want to scale, and if you look at all the brands that have scaled really well, right? Like a Mars Men, like a Gruen's, like whatever they're doing this, you know, this like post-Andromeda persona strategy where they're able to talk to a very wide variety of people.

When you say stack, can you, what does that mean? Are you taking like 2 of the same messages and stacking it for a persona?

I guess it's like a persona stack because you have different levels and everyone calls these things very different things, but you have your persona or an avatar and that's your who. And you should have 2 to 4 personas. Again, depending on the size of your business, the larger you get, the more you need to scale and you have to reach out to more people, but you should have 2 to 4 personas. So those are the people, the avatars that you're going after, right? You should know about them, their day, their problems, their beliefs, their desires, what they've tried before, where they are in the funnel. That's where I've been spending a lot of time lately. Just Believe it or not, manual research, like obviously assisted with, with agents, but like actually reading Reddit for hours on end. And same thing with Tribe even. That's what I've realized. Like I need to go deeper into like the, our customers, our emotions, stuff like that. And then once I have something, then you can give it to an agent and be like, do more of that, make me more statics. But it's very hard to automate it. And I think that's a mistake we made in the beginning. Like we had a lot of ads. We kind of, we knew, we know how to build a supply chain and automate it. Too many ads. Yeah.

Yeah.

That's where most people should be spending 80% of their time. I, I firmly believe it.

[Sponsor Content] This episode is brought to you by our friends at Aftersell. Let me start by saying there is something that all of the best operators have figured out, and that is a new customer is expensive as hell to acquire. So every extra dollar you can get out of that new customer, well, that's just basically free money. Ridge uses Aftersell. Simple Modern uses Aftersell. Jones Road uses Aftersell. Peely uses it. Upsells, downsells in the cart checkout after purchase, plus ROCT. So the moments right after someone buys make you money instead of doing nothing. Same traffic, more revenue. Sounds kind of obvious. We're building Winks right now. That's why you're listening to this show. And Aftersell went into the stack on day one, not a bolt-on for later when we decided we wanted to finally get smarter. Look, Black Friday's coming. You know it, I know it. Your traffic's gonna go up, your ad costs go up even more. Every single visit is worth more than it was last month if you do it right. So don't just spend more to buy more customers, make every visitor worth more. That's why we partnered with Aftersell. The production part is so, so cheap and there's so much optionality now for the production part.

Your ad can only be as good as your insight and your, you know, and your understanding of your customer. And sometimes I'll like be reading these threads and I'm like, no wonder like that ad didn't perform because it's like you're just saying, hey, do you want better sleep? But it's like, you know about these people's like problems, like their identity, their, you know what I mean? Like when you get down to it. So highly recommend people spend a lot of time there. And then right from there, you have your messages. That's the most important thing. And like Meta will talk about this when you're chasing creative diversity. It's not like, yes, visual differences or format differences matter, but they matter way less than the messaging. Right. So you, that's why it's, you know, a pillar, right? That's why you have that as the pillar. Then you have the product.

That's why I put it as the first one. It's like you kind of have to start here, right? You like, you need to learn the language of your customer. Um, or maybe not even the language, but like you did, I would agree. I think you deeply need to understand the customer. And I actually like, I, my, my take on this, Cody, is like, this is very hard to do in a post-Andromeda world, because everybody is so different. Like we were talking about like a lot of message variety, right? Yeah. And that's how you actually come at scale. Let's say you're doing manual research right now. How are you then incorporating or using AI to help you with that? Because I think this is the thing that people are like, are trying to speedrun is customer research and messaging. And AI can be super useful here, but I think like you, you're probably right. You probably need a handful of personas and messages that you're like, I know these are right. How do I go deeper on them?

So I got Claude open again, just, you know, between the day going different stuff. I have a super long thread and I'm just like doing research, right? I found a thread, this guy Carl on X. I thought, I think it's just, I love how he thinks about it and has had some really good articles. So I'm like, all right,

cool.

I'm going to, I'm going to follow this. I really like how he thinks about it. Right. We're going to test funnels. I want to test 4 funnels and your persona, your angle, your ad to your lander, to the offer. It's like all of that together. And the reason I want to do this is because I feel like we went, we kind of skipped a few steps and went volume. And I feel like we've tested a lot of ads and I'm like, what have I really learned from that? Like I actually don't have some of the learnings that I want. And so I'm like, let me make some more concentrated strategic tests and then we can do it. And so, you know, for where it'll likely go from the research is like, we'll have a cortisol funnel, we'll have a melatonin funnel, right? We'll have a 3:00 AM wake-up funnel and maybe one other one.

And are those funnels persona-specific or are they more message-specific? Is that kind of how you're thinking of things now?

I would say I'm probably going more message, but I would say persona, like for example, the 3:00 AM funnel will probably be heavily like a, right? Like a perimenopause, probably like that is who has that problem. Yeah. But I don't know if it'll specifically call out like, all right, do you have perimenopause? It might just, it might call out the problem, but that'll be part of the ads. We'll test, hey, what resonates? Is it, you know, yeah, what resonates in the ad? So that's how we're thinking about it. Like these 4 different lanes, like I know CTC used to call them pipes, You can call them funnels. Like everyone has different definitions, but like, those are the 4 things. So I'm like, all right, cool. We just need to go deep on research.

Would you agree that to get to depth, you first need to start out by like trying a bunch of things? Because I think for people, like people listening or watching, we have tried a lot of creative and a lot of messages in a bunch of different places. And what I'm guessing, Cody, what you're doing now is you're taking A lot of that like scattershot looking at it and saying like, okay, we hit in some of these places. Now I want to go really deep on these messages.

You don't want to start from the beginning. So you want to go with what data and insights you have. And, you know, it's one thing at Jones Road, you have, you know, hundreds of thousands of reviews and orders. Like you have a lot of data to go off of, right? We, you know, we're probably 6 weeks into Winks. We didn't have that much in the beginning. So we took our best guess. But now we have some insights and we know, hey, this melatonin funnel spent, maybe it didn't hit our CPA we needed, but like that is an insight there or cortisol. So we're like, all right, cool. There's some directional insights here or data. Let's now go deeper. So 100%. I had Runneth go through all of our reviews. I had Runneth go through our ad account to be able to find what are the potential top angles for us to look at. And then I also had Runneth go through of our competitor reviews. So yeah.

Oh, that's super smart. So I think that's a great call out for people, uh, that are watching this, like you're a new brand or you're a younger brand, you don't have near the volume of data that somebody larger than you does, but you can still benefit from that data because like a lot of reviews get published, right? Just look at Amazon. Like there's, you know, we did an operators episode, uh, recently with, and Mike was kind of schooling us on Amazon. He's like, it's like the public ledger. Right? Like there's just so much data on Amazon and then a place like Reddit. I, what's your take on Reddit? I have, I have had mixed experiences using Reddit for customer research because in my experience, there's like a certain kind of person that uses Reddit.

Yeah, that's fair.

[Sponsor Content] And it's a very narrow worldview. Um, that doesn't mean it's not valuable, but I've seen people use Reddit as like This is my source of truth for how the world thinks and works. And I'm like, it's not. Reddit is basically the internet's pillow. People go there to scream into it. Convince me why I should use Reddit. Let's start there. Most AI tools right now are all promise and no delivery. You know exactly what I am talking about. Super fancy launch videos. This is why I'm loving what Rich Panel is doing. They have AI and support smashed together, made it practical, made it useful. Made it valuable to brands today, not on some future promise. And instead of asking you to spend weeks writing prompts and uploading help docs and babysitting, they flipped the whole script. Their AI builds your support team for you and everything it needs. I've watched it. It works. It's freaking amazing. Richpanel even guarantees 50% of your support volume will be automated by AI within 30 days or your money back. I call that a no-brainer offer. If you are interested, go to richpanel.com/demo, not for some generic demo, but to actually book a call and watch them build your actual support team.

First of all, it's, there's tons of different subreddits about almost any topic you could think of. It's very easily searchable. It's, it's easily, it's easy to have your agent search it as well. I just don't know of a better place to find all of it, but I, I think there are, uh, there are other ones and you should, but I, I still wouldn't, you know, discount it. The challenge I've had is like, again, trying to find who our customer is. Like, I don't know that they're hanging out in a sleep subreddit, right? Like, but, and so I'm trying to find where they are and I'm actually trying to find sleep topics in other non-sleep subreddits, if that makes sense. Um,

okay.

Because sleep is such a general thing, but like, I don't think they're hanging out and going, oh, you know, sleep. It's not like they're like a hiker and they're like, you know, like we know our customer is into hiking and then it's easy to go into hiking one. Like no one's passionate about sleep unless you're like a biohacker, which is not our customer. So that's been my challenge with it, but I think we're, we're getting around that.

Yeah, I guess it's true. It's like you got to go into like mom groups or other places that people kind of get together and then sleep will bubble up sometimes as a subject, right? Like, can I have any mom or, you know, the perimenopause thing, like that's probably a topic that you can go hunt around for.

I guess to your point on Reddit, yeah, like use it. But if you see here, right, Amazon reviews, Right. Look at your DTC reviews of other customers, social comments, right? Like find sleep content on Instagram and, you know, do that. Facebook groups are great. Like join some Facebook groups. You know, there's plenty of public Facebook groups that you can join and just look for people having stuff. YouTube comments as well. So yeah, you shouldn't be limited and you shouldn't be doing just one source, but Reddit does make it easy.

How would you advise somebody who doesn't do this now, but is like, they know that there's tons of this stuff on Instagram and TikTok. Like they know that people are making content about this organic content, like not paid. How would you approach mining social media for these insights? Like what would you do? Would you use AI? Would you just manually go on there and start doom scrolling certain topics? Like try to get your feed to serve you more? Like how approach that? How would you approach this or how would you teach your team to approach this too?

Yeah. So, so here was what I did for this and I'll share some other ways I've done it, but I, you know, some people have a document or like in the past, you know, you'd have like a Word doc or you would have like notes, right? And like actually manually take notes. I think there's a ton of value in that, right? It's like the whole like adage of people writing copy. Like you do want to feel it and you want to be like, oh, I feel that. And that's what you miss when, when you're doing it with AI. But like, you also, I want my agents to have access to it. So what I did is I just, I just start telling it like, hey, I'm going to go and do all this research. I'm just going to screenshot it. I don't want you to do anything. Just save it, you know? Mm-hmm. And if I copy something, you know, um, and so that's kind of what I did. And I would just be like, oh, felt irritable and depressed. Boom. Saving that. I'm just screenshotting things that I see. Fatigued. Because you want, you want Their, uh, their emotions. You're looking for their pain points, their desires, what they've tried before. And ideally you have a research document. Um, and I've been building one. I'll, we'll make sure we have it in the show notes and stuff. Um, where you almost have like the questions that you want answered. You know, when you like work with an agency and they send you like an onboarding document? Yep. It's kind of like that. Like you should create a document where it's just like, what, what is everything I need to know to understand this person? What have they tried before? What are their desires? Like, what does their day look like? What are their failed solutions? You know, where, what's their, where, like all of that stuff. And just as you go through, just, just make notes on it or just give it to your agent and be like, hey, screenshotting this, put it where it makes sense. That was kind of how I went about it. The other thing that you can do, and this is more for like ads than like understanding the persona, You can create like an Instagram that, and I had my creative strategy team do this. I'm doing it now for Onyx. That is like you were that person because you should be seeing—

Oh yeah, dude. I had a phone.

Content. Yeah. Right. And cause I had an actual

separate iPhone for

this because where a lot of people start for their creative strategy and their ideas is looking at other ads and, oh, especially, you know, this brand is printing and here's what everyone's doing. Like, You should do that, but that should be to me, to me, like 10% of your ideas and inspo. I think most of it should be—

Oh, that's a great guideline.

Understanding your customer should be like 40%. And then like understanding the content. After all, we're serving social ads, right? And so we need to understand the content that our personas are engaging with on social. So you can even like, and hopefully you have this in your HQ and your agents, like deep persona research. Then you can be like, hey, I want you to I'm gonna create an Instagram, fake Instagram. Tell me who this person would be following, what, you know, and then you can actually, you then have to go and, you know, I'd love to hear how you do it, but train your feed, train your algorithm and go and engage with these kinds of accounts and then just scroll. And like, I would do it at night and like pretend you're that persona and be like, oh, this stopped me. I'm gonna save that to a collection.

Yeah, dude, I'll show you how I do it. I physically, so this is my, phone. That's my iPhone, right? Everybody can see that. This is my persona phone. So I have another phone, another iPhone with another Instagram account that is not mine, 'cause my feed is definitely not our customers. And I try really hard to just doom scroll Instagram as my customer or whoever I'm thinking of now. And I do this, I just have like this thing where like, I don't know what it is, but if it's like multiple accounts on one phone, like multiple Instagram accounts that you switch between, I find for some reason I see the same shit. Um, there's some crossover. So I try to actually have like a separate device. Um, you know what, Pilo, what I did was I actually had an American phone. I'm not kidding. I would like, when I was in the States, I would go buy a whatever, a cheap phone. just to create an Instagram account and on a cheap, on an American network. So it's like, what's the cheapest data plan I could buy? And then I would have that in my office up here in Canada because like we always found that as Canadians, we didn't pick up some of the nuance, the cultural nuance of the people. And 90% of my customers were American.

Yeah.

So I think you, I think I agree with you that you have to try to get like very deep into the mind of your market, your customer. I'm a little old school in that I do like to manually just read things and watch things. I don't know what it is. If there's like a, I think you made a comment on this and I don't know if you meant to say this, but like, to me, when you are researching a customer, it is hard to have AI do that because AI does not feel the way that a person does. So like when I watch a woman talk about a problem, on camera, on a video, I'm going to feel that as much as I'm going to hear the words and see the visuals. Like AI can hear the words and see the visuals, but there's like an emotion part that I think only a human can pick up unless somebody who understands AI knows that the AI can understand emotion. But like, I don't know what it is, man. I just like, I get a different feel for the person if I just manually watch them. So I'm an old school doom scroller. That's kind of how I do customer research is I just go like, I'll sit there for hours.

[Sponsor Content] I wasted 2 hours over the weekend trying to upload ads. And I said, I'm never going to do that again. So I had Runneth do it. It looked at every ad, picked the right landing page. It named all the ads, wrote the headline and copy to match from options I gave it, set the naming conventions, set UTMs, all of it, and pushed it to Ads Manager for me. All I had to do is go into Ads Manager, QA it, hit live. It was done. Runneth is an agent with its own virtual computer. So it never shuts off. It's always on. On, the whole team can talk to it, the same Runneth in Slack. It remembers everything. It has prebuilt skills based on some amazing stuff that the Motion team has built. Here's one example of how I'm using it. Last night I hooked up Tribe, our creator sourcing tool, and anytime a creator video comes in, it gets QA'd against my feedback that I've given it before. We get notified of any new video. I can have it add captions. It watches the video, figures out what it's about, adds naming conventions, edit hooks. I've been doing a lot, building a lot of static ads with Runneth as well. It's pretty awesome and it all comes out of the box. We've been using it on Winks as our main creative strategist, using it to build tons of ads for us. Winks is using it, Ridge is using Runneth, Hexcloud is using it as well. Go to runneth.com. They're giving listeners 500 million free tokens for your first month. Well, you should, you should people watch too. Like, I wasn't doing this on purpose, but right, like the way my mind works, like if I'm right doing ads for Winks, like I'm thinking about that all the time. So I'm at the mall a few weeks ago. It's like a rainy weekend, right? All the we're at the bookstore in the mall and like, I just see parents, like, you know, parents are there, right? Because they're trying to find activities for their kids. And I just see these like moms and dads just like so stressed out and frustrated. And I'm like, oh, like, 'cause you wanna feel that reaction. So like, go post up at a coffee shop or like somewhere where your demo is and like, just don't be creepy, but like watch them and have a notebook and start just like writing some ads.

Yeah, yeah, don't be creepy is great advice. Like, I'm a parent, I can hang out at my kids' school at drop-off and see how everybody is in the morning. Um, if I wasn't at that school, they'd probably call the police on me.

Yeah.

So yeah, don't be a creep. Uh, I— but hang on, I want to double-click on this because like, I don't see enough people in consumer doing this, and I think it is so important. And it doesn't matter if you're selling a supplement or food and bev, like CPG, or you're selling a physical thing, or you're selling beauty, it doesn't matter. You need to get out in the world and see how people interact with things, interact with each other. How do they shop for this stuff? Like, one of my favorite things to do, man, is, uh, I don't know if you do this, but like, when I travel, I go to malls, I go to places that people shop, and I will just get a coffee and I will walk around. I look at how the stores are merchandised. I look at the types of people going into these stores. Like, when I was in LA last, Uh, I went to, Sean took us to this like outdoor mall.

Was it, was it little, little or big?

Oh dude, it's America. Everything's big to me. My thing is like, I wanted to observe the difference between people going into Lululemon and people going into Alo. And that was more just like, it's not like I sell either of those categories, but Lulu's a Vancouver company, big one, very important here. Uh, and I wanted to understand, I'm like, what, who, who is this customer? And it's so funny, like being in person. You can visually see the difference between the two. Who are they serving? Who are the people that are going in? And there were people that would go from Alo into Lulu and Lulu into Alo, but I just found that so valuable. Like if I was selling activewear, like that's exactly what I would be doing. I would just sit on the bench, just watch both stores for a while. So I'm with you, man. I think this is like an under— appreciated and underutilized way of customer research.

It's just like you feel it more and like ultimately that's what we're trying to do is like we're people writing to people. So I, and again, my instinct is like AI, AI, AI. So it's been like, I don't want to say a wake-up call, but like, all right, like let's go more people and let's like really spend time on this and then we apply AI to it.

It's, uh, I think we're in this weird place right now with AI where we're still trying to figure out the current best usages of it in a workflow or like in any workflow in your business. So like we're talking creative supply chain today, and what you and I are actually spending a lot of time talking about is like the human part of this workflow is the customer research part. When, uh, uh, Perplexity's Computer came out, I just immediately started using that. I'm like, oh, you can do research. Like, this is great, but it misses so much nuance.

Mm-hmm.

It doesn't matter how good I, like, it doesn't matter. I've taken so many shots at it. It just misses some nuance. Um, and I think then like other parts of the workflow will be heavy AI.

'Cause we can talk about research all day, but like, let's make this practical now and then we can talk about sourcing it, right? So you have messages. I wanted to turn, take this research into like helping me just visualize and plan. So obviously just built this little, you know, dashboard with Claude. Um, but here are the angles and, you know, I have the research tabs for each of them. And so this is what I'm talking about where I'm like, all right, what's our desire, right? And I just am typing it in as I'm going through. What's, you know, for this, what's the experience they've had? What have they tried? How are they feeling? You know, just trying to fill out as much as I can. Again, the more you can do manually, right? So then when you're going to actually write copy or brief or whatever it is, like you just have it all available, you know? And so that was like step one. Now it's like visualizing it. I'm like, all right, cool. Like, how do I want to show that in ads, right? I think this will be like a minimum viable product testing. But I'm like, all right, like, what is a good way to show that in a static? And maybe you like connect it with something you have in your mind and you're like, all right, you know, I have saved in my asset library because every creative strategist should have that, you know, this us versus them ad. And I think this would be the perfect format to show it here. Like, let me slot it in or let me do that. So it's just like my, like, board for taking these insights and then trying to get a little bit like Practical with it.

You know what would be great here is, are you building this in the Wings OS?

Uh, I will. I can.

Yeah. 'Cause I think, dude, it would be awesome if you could just like drag in examples of each of these from other brands. So like, if it just think of like, think of how to, how do we scale this up then to a team? Right. 'Cause like, I, I think there's going to be people in all, in every brand still. Right. And as somebody, like one of the downsides to the way that I do this, like that I think of creative is like, I'm just constantly online, but it's always on my phone. So it's like, how do I get, how do I get that swipe file? Like, oh, this is the, this is the ad. This is an example. I want to save this in here.

I have so many screenshots just sitting in my photo library.

Yeah, I've got like so many like screenshots in my photo library. And in the past, I would just like drop them into my creative director and just be like, here, make this, right? And then she would just take it and run with it and put it through our supply chain. Um, but now I'm just thinking like with something like this, it would be amazing at a brand level to say like for the cortisol sort of like funnel, here are the statics that I really like as examples for this style of like this format or this approach.

We can get our agents to do that. I feel like pretty, pretty easily we can definitely do that. I think if we build this in and then train them on it, Then we could just be like, oh, like we could send it to Runneth and be like, hey, here's some inspo. And Runneth could be like, hey, I think this would work best for this angle or whatever.

Right. Yeah. So you just like drop it in Slack. That's my favorite thing about Runneth. I just like that it lives in Slack. I don't need to learn a new tool.

Yeah. All right. Well, I'll work on that. That'll be a fun one. So from there, where are we going?

[Sponsor Content] Well, okay. So I think what you just showed is like, we've got research, we have core sort of like pipes. Thanks, CTC. CTC, I actually like that term. Um, And I always called them like streams or, or, you know, uh, paths. So we have those laid out, right? That's like in our, in our creative supply chain. So like just for people's reference, that's like persona message. Okay. Like that's sort of like column 1, pillar 1 to a supply chain. Then we need to talk about what do I do with this? So like, where do I start making ads? Right. And what you see on screen here is you've got tribe creators, there's seeding, gifting, UGC, there's partnership ads, like larger creators, there's founder content, there's like, there's tons, there's AI statics, there's AI animation. Like, I literally, you cannot put all of the variations like on source. Like, where can you get these things, right? Like, I can use AI to make something, I can use a person to make something, I can use an agency to make something. I think Like for me, best approach is, is like to combine source and format and start with static image. And I don't know what your, what your take is on this, but like there's varying costs to different types of creatives. Like AI makes statics, carousels, I guess even some video, like some slop, pretty cheap. But let's talk about how we have been thinking about sourcing creative before we talk about, uh, formats. You have been listening to us talk for years about building 9-figure durable goods brands, HexClad, Ridge, Simple Modern, Pela. When we sat down to build Winks, there was never a question. It had to be a subscription brand. And being the veterans we are, we could work with anyone, but we decided to work with Skio. To build wins. Working with the team over at Skio has been awesome. We get to see all of the stuff that the fastest growing CPG brands are actually doing to acquire more subscribers and keep them longer. My favorite example of this has been incentivizing new customers to go from a monthly subscription to a quarterly subscription right after their first purchase, but before they get their first order. The team at Skio has been sharing insights on how much in-portal flexibility to give subscribers, when to change frequency, when to offer product swaps. When to do gifting, what rewards should look like, what is cancel flow optimization? The list is freaking endless and they have shared it all. They have been incredible to work with. They are bringing so much value to this partnership. It almost feels a little unfair. If you run a subscription brand, you should be talking with the team over at Skio. They know their stuff.

I'll hit them high level and then we can go into them. So our main sources and formats, so for video, we have Tribe, and we'll talk about our Tribe program. We have more standard UGC. This is kind of briefed or scripted UGC creators, and there's definitely some overlap with Tribe, but there's definitely separate pathways for that. Partnership ads. These are generally influencers, right? So larger followings. It doesn't have to be huge, but I would call it a different style of ad. It's a little bit less direct response and a little bit more like follower based, um, you know, then, so those are primary video, obviously you have AI videos and we haven't done a ton of these, but you do have various different forms. Obviously you have your AI song ads, you have your, you know, AI, yeah, AI VSLs. We're not doing it, but you have AI UGC, like, um, primarily using like Higgsfield MCP for that. You know, we're not doing Founder Ads actually. You have one ad in our account that's doing pretty well. Um, so definitely, I should make more. Yeah,

I, yeah, I will, I will make more for you.

We'll show that one.

But yeah, it's so funny. I was the number one, like I made so many ads for Pela and Lomi too, and they always worked. People love founder ads, founder ads crush for a certain audience.

Um, absolutely. Everyone should be, I would say. I see this one guy on X and I think you might have dinner with him at Beanstalk, this guy Ben. I don't know him. I don't know his brand he runs, but I just, he posts these like, um, screenshots of his camera roll and it's just like hundreds of founder ads. I have like tons of respect for that. Yeah.

He's a nice, he's a really nice guy. I hope he's listening to this. He like, uh, I met him for the first time in New York, uh, fellow Canadian. So shout out, shout out Ben. Um, dude, he's a great marketer, like really good, like student of the game.

But yeah, I'll see his camera roll and it's just like hundreds of ads and I'm like crushing it.

Okay. So those are your sources. What did we start at? Like, so walk us through like where we started. So we're like, what, 6, 7 weeks in right now to the brand being live. Where did we start? Where are we today? Like, walk us through the evolution of like how we sourced creative.

So we started very heavy AI, partly by design, partly we hadn't had the supply chain set up. And so we didn't have UGC creators. Like, we're starting from scratch. We had no UGC creators, no influencers, no partnership. You know, and so we had obviously AI statics, whether that was stuff we made directly in Claude, Codex, or with Runneth. So that was a big one. We didn't have a designer on our team. We now have one designer who does ads as well. Yep. You know, we had probably AI animated stuff, like, 'cause we just, it was what we had. It was what we had available to us, right?

Yep.

And so now over time, we were quickly like, let's get more people in there, right? So You shot founder ads. You know, we'll do more. We got some of them with, you know, Dr. Todd Bourne. So getting more people in there. We built out our tribe. And so we'll talk about that in a little bit. But we had tribe creators. We also— I've been working on building a roster of UGC creators and you obviously posted on X and did some recruitment stuff because I think that's like a big part of it is like, it's like a— Did you see that video? I haven't actually watched the video, but that guy from Whisperflow went viral for his like UGC program. Did you see that at all?

No.

Yeah. The guy from Whisperflow, and then he was like the second marketing hire.

His name is Hector. Yeah. Yeah. Yeah. How they ran their creator thing. Yep. Sorry. I saw another guy from Whisperflow doing his media buying thing today. I was confusing the two.

Yeah.

They're doubling down on that format.

Yeah. I didn't watch it, but some of the comments, I want to watch it, but some of the comments I saw was like, hey, it's really like you're building a sales team. And like, that's really what it is. Right. It's like, and so that's how I've been thinking about like, the recruitment aspect even, it's like you kind of constantly have to be recruiting creators, whether they're Tribe creators on a percentage of commission, whether it's flat, like you just need, it's a big asset to your company to just have people that you could turn to that you know can create really good content. It's just your supply chain. So working on that, also partnership ads, right? Like we knew that was gonna be a big one. So—

Yep.

We probably launched only a handful of partners. We've kind of slowed it a little bit as we've testing other stuff, but Definitely, we're very bullish on partnership ads and can be a big fan. So yeah, we just push, pushing out. Like right now, we have been very light on statics lately. I think we were very heavy on static. Oh, interesting. I don't know that it's by design. I think I've just been, well, look, we're a 2-man show and so it's more of just more of bandwidth. But I think especially for this like funnel testing, we'll do that a lot. I mean, statics are so cheap, they're so easy. They can be done with AI really easily now and pretty good. So like everyone should be doing it, but obviously I think like it's hard to replicate good video assets.

Oh dude, yeah. I mean, it doesn't matter. I've tried, I've been trying so hard to like make a great piece of UGC content. It just always feels off, always. Yeah, I come back to it every 3 months just to see, 'cause these models improve all the time. And then I'm like, okay, great, new model, it's gonna get better. And they keep getting better and better. And there's a point where I'm sure it's gonna be like indistinguishable. But there's just something that feels off about it. Um, and I just assume that that's like the viewer, it feels off for them. What, where would you say we've had the most success in sourcing? So like what performs best for us?

Partnership ads, definitely.

Yep.

[Sponsor Content] Um, so yeah, we've probably got like 3 or 4 partners live. Definitely. They've gotten, you know, most spend, uh, most of, most efficiency. Obviously they're more expensive, but it's, it's a worthwhile investment. I think also because you get, Obviously a new brand, you get the credibility of the creator, but you also get some meta signal on the partnership ads. And I think that you're just starved of signal in the beginning, so it can be really valuable. So big, big partnership ad fan. I'm getting back into the weeds. One of the things I'm probably the rustiest on is email. I'm super excited to be using Omnisend. They're using them on Winks. We're AI-pilled. We are using it to do everything. I'm running most of our email program from Claude and Codex using the MCP, and it's awesome. We also have email, SMS, and popups all in one place. So we have all of our flows, welcome series, it's all talking to each other. I love the platform. The analytics are really good. So highly recommend it, especially if you're in the position that Wings is in, new brand starting out. But if you're on a different platform and you're paying too much and you want to switch, they will do the migration completely for you. No lost data, no lost list. It takes only 5 days start to finish. And they are giving our listeners 30% off the first 3 months. Go to omnisend.com/9operators to try it out for yourself.

What is your take on, um, so let's talk partnership ads here because I've seen a bunch of stuff on X lately and just in conversations at Beanstalk. Do you think that partnership ads is like, is it more effective to have larger creators or smaller, do both work? Like, how would you advise somebody to think about this?

I would think of it like a portfolio allocation strategy.

Okay.

And so what I mean by that is like, you need to have enough creators where you have a portfolio and you have like a hypothesis that you're going for, right? It's very hard to do, like, you know, some people will be like, and I've seen some of the commentary like, oh, partnership ads don't work. Well, it's like, you ran like one creator or something like that, right? Like,

you

need to have enough budget to run like 10 and you need to find what works for you. If you have, you know, most brands listening to this, you might have $20K to start with, right? It's kind of hard to do with less than that. Do you go, you could go with one creator that's $20K. I don't recommend that. You could go with 20 that are $1,000, right? Like maybe, but maybe you go with one. Like if I had $20K, right? I would go with one creator who's $7,500 a month, right? And that's a lot, but like there is a good chance of them working, you know, or it's worth the bet if they're the right fit.

It's a higher probability shot 'cause they're like, they're better. That's why they cost more.

And then I'd probably have, yeah, they have larger following, more credibility, whatever it is. Then I'd probably have a handful that are in like the $3K range. And then I'd probably have, you know, fill it out with as many as you can in like the $1,000 range. And usually that's like per 30 days of content. And that allows you to try a few different creators, a few different bets. And so there's probably always a, you know, an allocation of that where it's like, 60% of the budget is on like your, your, your, your long tail, you're kind of cheaper, but then you use 40% of the budget on some more expensive ones.

These bigger swings.

Yeah. More concentrated. That, that's how I would go about it.

Is that what you did at Jones Road too? It's like just allocation. It's like we want a mix of people. And did you find that the larger ones typically outperformed?

Yeah.

Just because of their size of audience. And also like somebody who has a large audience, typically that means that they're pretty damn good at making content. So like, they're also very good at making these things.

Yeah. So like we would spend between like $100K to $200K a month on, you know, influence, like whitelisting.

Yep.

And so we would, we would test these different allocations. We'd test personas. Like we would always be looking at our portfolio and being like, hey, if we go, you know, 40% moms, 30% wellness, how's that performance compared to vice versa? But also on follower and audience size, we would look at that. The biggest bets we took were like $30,000 to $40,000. And they always worked. We always felt like we got it, but we were like, we only did 'em a few times. Like, I wish we went bigger. And we were very selective about trying to make sure they were the right fit. We knew that they were like a known performer for other brands.

Got it.

And the way we would look at it, we shot for a 10x content efficiency. So if we were paying $10,000 a month for somebody's creative, we need to spend $100,000.

You wanna spend $100,000.

Yeah, that was our target.

Okay. Yeah, we had a similar target at Pela. I think there's also like a, in a new account, partnership ads are kind of breaking open signal, right? Like you hit on this, that when you're new or you're smaller, like if you're not spending a lot of money on Meta, you are at a signal disadvantage to somebody who is. And partnership ads are a way to kind of like hack the system a little bit and get more signal going. Like we, we immediately saw it on Wings. It's like you fired up one of our partners. I won't say who, but we immediately saw like the order flow shift to a certain demographic, like right away. We're like, oh, that's the algo. Look at it. Like it's just doing its damn thing. It's delivering us customers. And when you're small and it's like the order volumes are small, you can see it so clearly. Um, so like that to me was like a, a wow aha moment. 'Cause at Peel, it's like, you know, thousands of orders a day. Like I just didn't, it's it's much harder to see it in that stream. But I just wanna think for the audience, like it would be great if you could unpack how we think about compensating Tribe creators and how that contrasts to what we just talked about with partnership ads. 'Cause partnership ads, like there's some kind of retainer, they deliver a certain amount of content, we're allowed to whitelist it for a certain period of time, right? Yeah. That's like, that's partnership ads and you're negotiating those retainers. I do want you to unpack how we negotiate those. Like, I think you've got some good systems there, but first start with Tribe Creators because they're kind of on the other end over here. It's still people making content for us, but we're paying them differently. How did we set that up? Because we had a call with the founders from Tribe and they kind of gave us some guidance on how we should do this. So maybe just walk people through how we set our Tribe program up, because there's lots of ways to do this too, by the way.

We're not Tribe experts at all. We obviously set it up. We had a call with Lucas and the founding team that was extremely helpful. So we'll share a little bit about what they taught us and I think what we've learned. I will say, like, I was super bullish on Tribe and I'm a huge fan of it. We've not had the success that I was hoping and expecting, and I have some thoughts as to why. But one of the things that Lucas cautioned against was like, everyone hears the Hudson Method, everyone hears, you know, 17,000 critters in a Discord or whatever it is. and they just go volume and they just go, hey, this is a, we're gonna pay a percentage of spend, no flat fee, and we're just gonna ship samples out and do like TikTok shop. Yep. And that doesn't seem like that works for most brands. And, you know, Lucas very clearly like cautions against that. And so that is one style of program. And I actually think you should have that. You should have a program where people apply, they don't get an upfront retainer or cost per video and they just get a, you can either do a percentage of spend or a percentage of sales. We're doing a percentage of sales.

Yep.

Getting 15% of GMV, which is using the Tribe attribution. And so we have that. The challenge that I've noticed is the content is just very hit or miss. And you can say that's just to be expected and it's a volume play because you're just paying a percentage of spend. And maybe you just need more volume. But I also think there's ways around that and you can do it. So the other program that we have set up is a retainer program. And so these are incentivizing creators to create content for you, especially for a new brand that we're not spending that much. They're not gonna make that much on a percentage of spend, right? Yeah. We started this in the beginning. And so, you know, we initially went a few thousand a month for 60 videos, you know, per creator. And it was a lot. And I think it was too much for us. A, we had a lot of volume.

It was sized wrong for our size of spend. Yep.

Yeah, we got a lot of volume. A lot of the videos are the same. And then we didn't at the time have the right systems and maybe we still don't for coaching and investing in the creators. And that's, I think, the thing that I have learned is most important is the actual knowledge of what makes really good content. As well as incentivizing them, um, are probably the 2 biggest things.

[Sponsor Content] Every CPG founder thinks they know their margin until the Amazon fees, the chargebacks, the COGS fluctuations in the month, and the ad spend, and every other little fee you never see quietly eats into it. And then the real number, well, it shows up a month later when your accounting team finally closes the books and maybe has the bravery to tell you. Iris Finance connects your Shopify, your Amazon, your QuickBooks, and, and your ad accounts all in one click, and it gives you a live P&L by channel, by SKU. you, not a spreadsheet from last quarter. Brands like Jones Road, Caden Lane, Jolie, Stars and Honey, and now Winks use it to catch margin leaks in real time before they turn into a real, real problem. If you're running a CPG brand and you're past $5 million in revenue, go to irisfinance.co/operators for a free demo and see your real numbers. Just for people, uh, who are following along in this episode, we are absolutely coaching Tribe creators. Like you are getting on, you are showing them what good ads look like. And I think you sent me a message, it's like, it is working. Like they are making better content, like the ones that are listening and they're engaged and they're in, like they're making better content.

So I wanna go to those creators and, and you can start with people just on retainer or like we've had a few creators that are not on retainer that are showing up consistently, producing great content. I'm like, hey, you're clearly serious about this. Like we'd love to invest in you. Can we go put you on a retainer? We'd love to work closely with you and scale this together. So that's what I would recommend. And like, even if it's 5 people, like this is what I'm recommending to like the Jones Road team now. Like if you have budget, like put 5 creators on retainer, but just like weekly calls, give them feedback, coach them. Highly recommend it. So I think that's the playbook there. Once you find winners, then your other tribe creators are gonna see that and they're kind of gonna learn from that slash knock those ads off. And so I think that's where it's helpful, but you gotta get Some of those winners first.

We went from, you know, AI statics, we did some AI animation, then we did tribe, then we did partnership ads. And I would actually say that that's probably like the correct sequence, right? Like you wanna take a message, you wanna test that message funnel, whatever we're calling it, like the whole funnel, you wanna test it as cheaply as possible to get some signal. Then you go to the more expensive creative. formats, right? You go to Tribe, then you go to like partnership ads being the more expensive one. And that's kind of like the process you're gonna run in sort of supply chain. Would you agree with that? With what I just said?

Yeah, I agree with that.

Okay. And then, uh, when you get to partnership ads, how do you think about negotiating with the creators their fees? Because I, I'll be honest, I've struggled with this, like, How to like put guardrails up and what do we know we want to pay? And because it seems like it's a feel thing, right? Like you just have to have a feel for it, but you clearly have like some science behind this. Like this is just the math that I'm willing to work with.

We worked with Superbloom at Jones Road, and so they've taught me a ton, but I've kind of obviously now we don't have budget for them, so we're doing ourselves. And so I have just also just done a little myself and, you know, adapted it. First, first reach out is just, you know, you're just emailing, hey, this is, you know, Cody, I love your content. Would love to work with you. If you're interested, can you please send us your stats? And so this is super important is you want to get their stats. You have to anchor it in something. So you want them to show a screenshot of their 30-day Instagram story views, engagement, and link clicks.

So not just stories, like you wanna see reels, you're like, I wanna see like what your data, like inside your Instagram app, like send me that.

Yeah. So it's like, can't be faked, right? It's not a follower metric or something like that. 'Cause part of what you're doing is you're anchoring either to a CPM or to a cost per link click. I'll ask them for like, the quote is an Instagram story, 3-frame, you know, video plus 30 days of usage. I don't expect to be profitable on that, but I find that the quotes for that are sometimes the same or less than if you just got whitelisting. So that's at least what I found. And then, you know, what Superbloom did a great job of is they never paid extra for whitelisting. They just had that in their contract and deal. And so I prefer to ask for the rates of that, but only doing it once you have that. They had a really great spreadsheet where they would essentially have a calculator that would create an anchor rate or based on the stats, based on the follower size, you know, based on maybe expected conversion rate, just like, hey, here's what we're willing to pay. Yeah. Right. And so you get a quote and so you might, let's say that in our calculator based on somebody's stats, it's, we can pay $5K for that deliverable of, you know, the Instagram story post plus 30 days whitelisting. They're probably gonna come at you at $10K first. Right. And I, you know, often it's with the manager, sometimes not. And there's some trial and error of how you get to the rate that you want. Sometimes you don't and they're just gonna push, but you'll get some crazy high rates. You'll get some and you gotta push, you know, and you're not really gonna upset people. Like, that's what I worried about. Like, I was always really good at like negotiating with like a vendor, like a Klaviyo, 'cause I'm like, I don't care, it's B2B, but like, I don't wanna piss off an influencer. Like, I don't think that I have. The managers just know the game and the managers can be really tough and, you know, you just gotta do it, but do it respectfully. And so the better you can anchor it in data, right? the better it is. Like, hey, based on these numbers, here's the most we can do. And you just have to go back and forth a few times. Sometimes you gotta like leave 'em on read, ghost 'em a little bit, they'll follow up. You know, you get a little bit of leverage from that. Um, one of the things that I've, I've noticed is they also have a, you know, depending on how the, the conversation's going, they have a number in their mind, right? They don't think about gross margin and stuff like that. No. So like, let's say they want $10K. I'm like, hey, we can do $5K. They're like, $7,500. I'm like, well, great, we can do that if you can also add this in. And so your like cost per post goes down. So you just gotta play it out and go, you know, back and forth. But that's it. And just know in your mind like what you're willing to spend for that.

Is there like price bands that you've noticed with like this many followers is typically gonna cost you like X, this many followers gonna cost you Y? Like, or has that been all over the place because there's just too much variance in the quality of the creator?

It's a good question. Probably. But there's a ton of variance as well. Yeah. You know, definitely you're, if we're just talking whitelisting, you're like $0 to $10K, like you should be paying, you know, $500 to $750, $1,000 at most. $25K to $50K, you're probably paying like a few grand. You know, you get into like $150Ks, you're, I mean, maybe a little bit more than there. And then when you get up to millions, you're, you know, it really depends anything between $3K if they're like engagement and stats are pretty terrible.

Yeah, it's terrible.

Yeah. You know, to, I mean, again, I've, we paid $30K, $40K. So.

I think the biggest one we did was like maybe $15,000 to $20,000 a month as their like retainer. But their content was incredible. Like they weren't a huge, huge, huge influencer creator person, but their content was like just top shelf and their engagement was amazing and they were super likable. And like we sort of went through like, here's the numbers. But then also here is the, just the, like the vibe in the, like they're brand aligned. You know, this is the kind of stuff that we want to produce. They had influence. Like you could just tell that like their audience really loved everything that they did and said. Like I, we know, I won't name drop, but like we know of creators where like it doesn't matter what they tell their audience to buy, their audience buys it. Like they have rabid fans. So like those people are gonna cost more 'cause the— and those people typically know that they're gonna cost more because they're very, very good at what they do. Um, you know, so, and we all want them, like every brand wants them.

And if they're more selective also, because a lot of creators are like, hey, I'm not gonna take on every deal. My audience knows that. And so I have to, yeah, I agree with that.

I think the thing to remember with large creators is like you are working with another brand. Like it's not just some like random person. Um, for me, it's actually the mental frame there is a little tricky with like, do you got Tribe creators or TikTok Shops creators, which are like, these are your salespeople. I think they're closer to employee than not, right? And like, you have to train them and coach them and like, whatever. And then on the other end of the spectrum, it's like large influencers are brands in and of themselves, right? And that you need to work with a brand like you would work with any other brand. They expect—

I mean, that makes sense. They have a team. They have a team.

Yes.

They care about their brand. They have a business they're running. I think that makes a lot of sense.

And you kind of have to size and scope and allocate your own resources accordingly. That was how do you source creative? How do you then think about layering in like all of the different formats and approaches? This is where I find all content on X and on Instagram. is here. It's in the format. It's like, hey, you gotta try this style of video. How much time, how do we, like, how are you actually thinking about this? Like these, like steps 3, 4, you know, media format, creative approach. It's like, there's just so much variety or variance here. And I just, I always have a hard time sizing effort.

Yeah, I think it's important, especially at later stages. Like again, the persona and the messaging is the biggest factor for diversity, but format is. important. Like you're gonna reach different people in different placements with different formats, right? So for example, and again, I don't know the latest, but like longer form, slower VSLs, you're probably gonna reach older people and on Facebook feed, shorter, quicker cuts, you're gonna reach younger people on Reels. And so if you're trying to maximize your delivery to these placements, you gotta understand the type of content that works on those placements, but also of From the demos. Um, you know, same thing with setting stuff like that, but because we're so new, like we don't really need to think about diversity right now. Like we just need one, right? Like we just need to figure out like, what is our best bet? And like, I think YAP or UGCs are probably our best thing, right? Like that's just like,

yeah,

it's like you have a best

bet.

Yeah. Can you have a creator who just is good at talking, follow a good script, hit all your direct response principles? Like that's Probably my bet. And then once you get something that works, like then, all right, cool. How do we take the same message? How do we put this into a song ad? How do we wrap some statics around this? All that stuff. That's how I think about it at least.

Yeah, 'cause you could take that, that's just it. You could take a Tribe creator who has a hit or a partnership ads, and then you could say like, all right, I wanna turn that into a 3-minute to 5-minute explainer video. So you can take like a piece of UGC, then it works, you then turn it into like, okay, great, now I'm gonna make a 3 to 5-minute VSL-style video. I could take that talking head, talking to camera UGC from a Tribe creator, and then like now I want 50 other Tribe creators to make it, right? Mm-hmm. And I want different sizes and I want some of them to be faster. I want some of them to be slower. Um, and that'll kind of let then like optimize to different placements. I pulled up the supply chain again because like it shows that you can sort of take like a short-form video, um, and then like have a whole bunch of different approaches, right? Like you can do ingredient explainers, Q&A, unboxing, objection handling, like all of those are just like approaches to a short-form video from TribeCreator is how I like think of supply chain. It's like, okay, step 2 is like, okay, it's TribeCreator. Step 3 is what are you making? Short-form, long-form. Some of them make static, some of them are doing all kinds of stuff for us. Then it's like step 4, creative approach. Like what style of ad is it? And then 5, I wanna spend some time here. This was actually a pretty serious unlock for us at Pela was we started thinking of creative around stage of awareness. I guess like funnel, right? Like top of funnel, bottom funnel, middle funnel. Um, and I don't know what your take is on this, but like we actually really had to put a lot of effort into making things that we had a strong thesis on, like where in the various stages of awareness was this piece of creative living? What is your take here?

I think it's extremely important. I think you want full funnel creative in the account, especially with like the changes on Meta. And again, this is probably more important at larger brand sizes, but like, you know, with, with GEM and stuff like that and sequence learning, like you, people aren't gonna convert off one piece of creative and you want to just have a bunch of different things in your account that can reach them at different stages, right? Yeah. You might have your top of funnel video ad that like the attribution looks terrible on, but it's getting a lot of spend and then you clean up with your bottom funnel static. And then same thing with landers as well. Like now it extends beyond the ad. It's just for us, right? Like we don't need that much 'cause our account is so new. So I think when in doubt, go problem aware, right? Like, that's probably who you want to be. But I don't know, especially because when you're new, you also like, you need to go up funnel and you need to reach more people, right? If you're just talking bottom of funnel. So I think we could probably go a little bit more bottom funnel. Like I did one of those like comparison pages and blogs, like we got some purchases off that. Like, you know, one of the funnels that we're testing is this new thing is melatonin. Like that's I'll call that a middle funnel thing. We're not gonna reach people and be like, hey, are you having trouble sleeping? That's gonna be like, hey, are you having trouble with melatonin? So like, that's inherently middle funnel. Yeah, you definitely need to think about that and understand where people are at in the funnel.

Yeah, I guess it's a good call. It's like some messages are going to be inherently a different stage of the funnel, right? And then some creative formats also lend themselves well to different stages of the funnel, like longer Like we had a lot of success doing longer videos for top of funnel, like just stories, things that like had just wider reach. And

we were

going after like more unaware people.

Also based on your product as well, right? Like how much of a, you know, again, demand creation versus capture, like market are you in?

Yeah, that's totally it. I think like, you know, Meta is this incredible targeting machine, but it is most attuned to going after people that are in market for the problem that you are trying to solve, right? Like, it's incredible at it, dude. I, uh, it was Canadian Thanksgiving here this weekend when we're recording this. And, um, I don't know, I was sitting on the couch just like doom scrolling Instagram and I saw this like duffel bag thing that had it. Look, it was, they built this duffel bag to also carry like suit jackets without getting wrinkled. And I'm like, oh, that's novel. Like, that's a, that's a neat thing. I frigging watched the video. For the next 2 days, I just got hammered with freaking duffel bag ads. And I'm like, I'm not in my, like, that's how, that is how their algorithm works. It's like you watched a 2-minute video going through the design of this bag. I did it because I just like products. Uh, and Meta's algorithm was like, oh, you must be wanting to buy this. I'm going to show you every bag company on the freaking planet all at once. And my feed just switched to like that. It was instant. That's how little signal it took to, but I think that's like, it's an important example of like, this is how people's feeds are working. And when you're making ads, like, I think you said something at the beginning, which is really important, which is you are still making content, right? So like, you have to think like, I need to make content that fits into a person's feed. And that content also has to be an ad and it sits in a stage of the person's Sort of awareness, right? And Meta is best at people who are at a stage of awareness where like, not only are they problem aware and solution aware, like they're probably your brand and product aware, and that's who you're going after. All this said, Cody, we've gone through like these sort of the supply chain, but I think there's still gonna be somebody out there saying, okay, but how do you edit the videos? How do you design the statics? Like, is this AI? Is this people? Like, how does that fit into this? entire supply chain? How do we do it right now?

It's crazy how quickly it's changing. So for, for, we have one designer/editor on our, on our team, uh, that we source through Propel. We, and again, you always need humans, but I think there's, AI can do, you know, so much right now. It's like crazy, like weekly it's changing. Statics, we're doing a lot with Runneth, and that's essentially running on like, you know, APIs of different things, but you can use Higgsfield for it. You could use, you know, Melius, you could use directly OpenAI API. Um, you can do that. I still think that there, You can still tell they're AI, but they're really good. I still think there's value in having human-designed statics as well. And so highly recommend both of them, but they're easy to produce, cheap. You could test quickly. And then for video, so almost, I would say Tribe stuff doesn't need editing. Creators edit themselves. What we do, it's actually pretty cool, is we have Runneth watch them, check them for safe zones, give feedback. So it's like trained on all of the feedback that we've given. And so it'll look and be like, hey, there's not a great hook, or you said that, or, you know, there's a compliant thing. And then we will review it and manually send it. But that's one thing that we do. There's just a lot of back and forth in that partnership ads and standard. So partnership ads, we have to edit, we get the raw videos.

Yep.

And so I would say our editing approach has evolved every week for the last 3 or 4 weeks, I would say, starting in September. Right. But prior to that, it was all human editing. Yeah. You know, and then Astra came out and I saw a lot of people using like Astra with DaVinci Resolve MCP. And so I started doing that, built some skills and has been, you know, for basic edits, like just as good, trained it on the captions. Again, not out of the box. You got to train the skills. And then last week I saw people using Opus 5.5. Without anything external. And so that's been editing. So like editing is very solved, which is pretty wild. And so now it's easier for me to just go into, you know, an agent and have it edit it instead of need to like draft it to a person. So it's actually pretty incredible and quite a time to be alive.

And the workflow there is like you basically give it the video and then you could use like WhisperFlow to describe kind of the edit you want and the style you want and And then it just goes and does it and then you can kind of iterate on it.

Yeah. Well, now I have skills. So like if it's partnership ad, I have one skill with like different workflows in it. And so if I say it's a partnership ad, it'll find the creator's Instagram, find how they normally talk about things and the font that they normally use on their Instagram and use that.

Dude, that's smart.

It's pretty cool. And then I hand it off to Runneth who adds naming conventions, uploads it. So like there's a lot of those manual parts of the supply chain that like, you don't have to be doing? Because I would consider that like, I remember a year ago at Jones Road, like we had this big spreadsheet where like naming conventions, every ad upload, like it took a person. It took—

Dude, the creative key, right? Like remember years ago, like that was Connor McDonald's thing. It's like, you need this creative key and this is how we met. I'm like, I don't know anymore, man. I think it's just like, let the AI figure that out.

It took a person, like literally their job was just a low-level person tagging things, doing all of that stuff. Now it's done. Now the thing is you have to have that key, that setup, right? Where you have your taxonomy, but then your agent just can do it all.

And you're doing, so your workflow is like, okay, I got a video from a creator. You drop it, you open up HQ with like Opus 5.5 or whatever, Astra with DaVinci, and you drop it in. It's like, hey, this is, go edit, like run the skills, right? And then from that, you give it to Runneth. And then Runneth takes care of the meta upload, the naming conventions, all that stuff in our account.

Yep. Yep.

It's, it's freaking wild that this is where we are.

It is.

Like it actually, like it gives, it makes me so bullish on everything marketing because like we can just focus on the ideas and the content and a little less of this, like the production. Um, keep saying it, man. Like AI is just taking out the, like there's slowly, it's taking out blue collar internet work. Right. And, you know, you get to go do all the creative stuff, which is so fun.

And that's where I'm currently at with it, where especially as like, you don't have to be technical because all of these agents are,

you

know, you don't have to use terminal and they just come out of the box with all this stuff. It's like, I know it's like super cliché and everyone's like, the only thing left is taste, but like, it is true. Like I was at this Runneth event and Dara, who's awesome, I didn't know she was really in AI, but she has like a creative strategy agency and she shared some AI ads that her team made. And they were like, I'll send you the ads. They were the best ads I've seen. Yeah, flip 'em over. Like, you and I spent a lot of time looking at this stuff in AI. I couldn't tell they were AI at first. Like, because I think like sometimes when I design stuff, I'm like, hey, I actually don't think that this is a Claude issue. I think this is like a Cody issue. And that's where I felt like her team knew how to prompt and brief in a way that like, I don't know. I've tried to like recreate it and I haven't yet. where like they were able to do stuff like it was really impressive, like in terms of how they trained the reference for the person and like they cut at 1.5 seconds all the time. They had specific prompts about what to do with the hands and the face. That was like you kind of had to be like very skilled to be able to do it. So my point is, I think creatives that apply this stuff are actually in like the best position possible.

Yeah, sure. Because they have really great taste and a good eye. It's fun to see something like a real example take shape. Um, yeah, Dara's awesome. Uh, very cool. All right, dude, I think that's the pod. I think that's like, we've mapped out our entire creative supply chain, how we're thinking of things. Cody has shown you some of the tools we're using, some of the workflows. Um, we will share as much of this as we can in the newsletter. I said at the beginning of the episode, it is free. It's crazy if you're in consumer and you're not on our newsletter. Like it's probably the best piece of written content that goes out every week in the space. Um, but I think that's the pod, dude.

I think it was a good one.

All right, we're done.