
Marketing Mindset Podcast · 2026-03-02 · 34 min
Key moments - from our scoring
Substance score
36 / 100
Five dimensions, 20 points each
The episode explores the fundamental shift in creative advertising strategy over the past few years. Harry, a creative strategist who previously worked at MHI Media and now runs his own shop focused on front-end ads and pre-sell copy, breaks down why the proliferation of tools like Atria and competitor-copying has made traditional USP-driven static ads increasingly ineffective. The conversation centers on what actually separates winning agencies today: the ability to create organic-feeling content with direct response elements, exemplified by creators like Rosabella Moringa. Harry explains two tactical approaches gaining traction in 2025 - partnership ads (running content from influencer pages) and whitelisting (creating relevant Facebook pages like 'Living with X Problem' to run ads without paying creators). He emphasizes that page naming matters significantly, as does thoughtful targeting around mindsets and emotions rather than blanket approaches. The discussion also addresses the volume-versus-quality debate, with both agreeing that replicating proven formats and angles across clients at scale, while reserving human effort for strategic swings and script development, represents the ideal operating model. Critically, Harry and the host identify a significant human advantage over AI: contextual learning from past campaigns - understanding that esteem works well in fashion with certain demographics - which AI struggles to apply across different scenarios. The episode challenges the tendency to over-automate creative work, arguing that the 10% extra effort humans put into research and thinking creates a sustainable moat against commoditization.
No - you can rip ads from a freshly created whitelisted page on day one, though some run a cheap like/follow campaign for a few dollars over 5 days. The page name and logo matter most; naming it around the problem or demographic (e.g., 'Living with X Problem') has shown better performance than generic names even without warmup.
Partnership ads involve paying creators or influencers to run ads from their pages, which works well but costs money. Whitelisting is free - you create your own relevant Facebook/Instagram page and run ads from it, which achieves similar lift in CTR and CPA without the influencer fee.
Deploy proven formats and angles at scale across similar clients (replicating winners across your bank of past campaigns), then concentrate human effort on strategic new tests, finding quality partners, and writing strong scripts - treating AI as a research and data-collection tool rather than a replacement for thinking.
The brand should own the asset whenever possible, so they retain optionality to use it elsewhere if the agency relationship ends, rather than ceding leverage to the agency.
Contextual pattern recognition from past work - understanding that esteem works in fashion with certain demographics and applying that insight across different scenarios - which AI cannot do effectively because it lacks the human ability to synthesize specific client histories and emotional triggers.
Our reviewer’s read on each dimension, with quotes from the episode.
There are a handful of tactical nuggets - naming a whitelisting page after the ICP problem, using AI to scrape Reddit then manually reviewing output, structuring Claude projects by asset type with modular prompts - but the episode is heavily padded with hedging, filler, and meandering tangents. The opening product pitch and much of the closing small-talk further dilute density.
I literally just changed the name to Living with X Problem and then that out the gate. And that performed, like, instantly better than that, um, than the other, like, whitelist page.
I do it a lot with like Reddit right now where I get it just to scrape Reddit for me based on keywords. But then I still read through the reports that it gives me.
The whitelisting page naming insight and the argument that brands - not agencies - should own volume production while agencies focus on high-difficulty creative are mildly contrarian, but the rest recycles standard post-iOS 14 DTC discourse and broad AI takes that circulate everywhere in the space.
I actually think brands should be responsible for most of their volume because yeah, if you can copy and paste like the formats that you know, work well and you can produce that on mass volume, like surely you'd want like the experts to come in to make that harder content
people who can make content that people actually enjoy watching... focusing on the organic feel with like a hint of like direct response
Harry is a junior-to-mid-level practitioner just launching his first shop, with prior time at a creative agency and a brief stint as head of growth at an eight-figure brand. He has relevant hands-on experience but has not operated at meaningful scale and is still establishing his track record.
left there, uh, then worked as like head of growth for like an eight figure brand for a few months and now have just opened up my own, my own shop.
I haven't announced it yet at all. Uh, fun facts about how that name came up. Um, it was actually Andrew Foxwell's daughter Nora who came up with that name.
A few brand names are dropped (Rosabella Moringa, The Ridge, Hexclad) and some rough numbers appear (two to three dollars a day warm-up, eight Claude modules), but there are no concrete performance figures, timelines, or client case studies with real data. Most evidence is anecdotal and approximate.
the Rosabella Moringa people, like that's like their, their forte and they've like absolutely done like insane numbers in such a short amount of time.
you can just do like a like or follow thing for like mhm. Literally like you know, two, three dollars a day and just do that for like, you know, five days
The host asks a few genuinely useful follow-up questions about page ownership and social-proof requirements for whitelisted accounts, but the episode opens with an extended self-promotional monologue, many questions are leading or softball, and there is no meaningful pushback or challenge to any claim the guest makes.
When you're doing whitelisted ads, do you um, do you as the agency own that page or do you get the brand to own it?
Do you feel like you need followers on the page? Do you need previous content? Like do you need any social proof? Or can you just rip it with ads?
Computed from the transcript - who did the talking, and the words that came up most.
Creative has never been easier to produce, and that’s exactly why so much of it stops working. In this episode, we sit down with Harry to unpack what’s really happening inside modern ad accounts and agencies. From the rise of creative volume and AI-driven workflows to why copying competitors now leads to “good enough” results, we explore what actually differentiates brands and agencies today. We talk about: Why average ads used to work, and why they don’t anymore How agencies balance volume, quality, and early performance pressure Where AI helps creative teams (and where it clearly doesn’t) Why judgment, pattern recognition, and deeper thinking still matter And why quality-focused creative may be making a comeback An honest conversation about scaling creative without losing what makes it work in the first place.
Transcribed and scored by The B2B Podcast Index.
Speaker A: If you know me, or if you've been listening to the podcast for a while, which I appreciate, you know that I run a performance marketing agency called BrighterClick and like a lot of agency owners that listen to this, we specialize in creative strategy. Now, when I started BrighterClik, I had been freelancing and transitioned into building out an agency because demand started to grow and I had and still have a deep obsession for marketing. That's what made me go into the field of marketing, to freelance and then build out an agency. But what I've learned over the past six years of this agency is that I also have an equal passion or focus on operations and process building. And as creative strategy became a newer service for agencies over the past few years I've really focused on building out, uh, processes not only to plan and to execute creative strategy, but also to monitor performance and optimize it as well. And through all of that, one thing that has become painfully honest or painfully clear is that too many hours get wasted by big brain people with great pay and great capabilities doing repetitive menial tasks, pulling data reports, going through spreadsheets, looking for data, and they don't get to spend enough time doing what they're hired for, which is using their capabilities in their big brain to make sense of the data and and then plan out next steps that will ultimately improve performance. When I focused on solving this, I started out with a Google Sheet setup and we'd use that to aggregate data and it worked. But there's limitations with Google Sheets. And so over the past two years we've taken that setup that we have as an agency, modified it a bit and worked to turn it into a software that could be used by any agency to streamline their processes. And I'm very happy to share after two, what seems like very long years working behind the scenes to get this built out, data Ally is officially in beta testing. We are fully bootstrapped. So we're rolling this out with a purpose. And I mean that in two areas. One is we want to make sure that we keep agencies as the forefront. So we're looking for agency owners or agency head of operations, whoever it may be, that are interested in trying out a new tool for free and are willing to give us transparent, clear and honest feedback on how we can improve this so we can make sure from day one we're building an agency focused tool to help them streamline their processes, improve client retention and rapport, and improve their margins. But we're also bootstrapped in the sense that we are doing this with a limited uh, budget. So we're going one feature at a time. We're also not going to be accepting a ton of agencies for the beta testing because we want to make sure that we can provide great uh, service and communication to the people that we do work with in this free beta period. So if you want early access, please, uh, go to Data Ally AI, sign up for beta and I will personally reach out to you with next steps so that we can get you into the tool and into the Slack community so you can start testing it out. Awesome. O' Harry, glad to have you on today. I know we first connected in person, which is rare for the D2C space to meet in person for the first time versus Twitter or LinkedIn or something. But uh, first met m in person at a Foxwell founders event in San Diego. Was uh, that last year? That was that this year?
Speaker B: No, I think it was January. Right.
Speaker A: January of this year. Okay. Um, so I'm glad to have you on. I know we talked about uh, having you on a while back, but glad to have you on. If you could just introduce yourself uh, for the show and then we'll kind of go from there.
Speaker B: Yeah, thanks for having me on. I'm Harry. Uh, don't know what to introduce myself as, you know, worked at a, you know, a creative agency before called MHI Media. I was a part of there, uh, left there, uh, then worked as like head of growth for like an eight figure brand for a few months and now have just opened up my own, my own shop.
Speaker A: And your focus solely on creative at your shop?
Speaker B: Uh, predominantly I am, yeah. But like I also started it with my, my partner who I think you met in San Diego as well. So she's on like the retention side and I'm just doing more like. And she does like a bit of ops and stuff on my side as well because I'm super, super unorganized and that's never been my like strong suit. But yeah, she's like focused on retention and that kind of stuff and then I'm more doing like yeah, uh, front end ads and then also like uh, like pre sell page, like copy. I don't like build it but like you know, just like listicles advertorials, that kind of jazz.
Speaker A: Awesome. And, and so um, creative, you know, probably going to be a big part of our talking points today. I know I told you before we got started, I went through and looked at some tweets you've put out, uh, for anybody that's listening that doesn't follow Harry on Twitter. Definitely go give him a follow. He puts out a ton of great, like, documents and PDFs to share with your team, but also just good information. Um, so I have a lot of. A lot of good things to go off here. And I know one thing you said in a tweet was just talking about how the ad games got more sophisticated. And it's. It's a lot easier for brands now, or not brands, but agencies brands now to make ads that are just average now that could have popped off two or three years ago. Personally, from my perspective, I think, um, copying competitors is a very common thing. And I think AI is just kind of perpetuated that, especially as certain like, um, creative board tools and stuff open up API and just kind of promote copying more and more versus creating net new. Uh. What do you think separates agencies from being able to create successful ads now versus just average ads?
Speaker B: Yeah, it's a good question. I think, like, you know, kind of like touching on like the sophistication kind of thing because, like, I started kind of like getting involved in like, ads, like I'd say like post iOS 14 or whatever. It was like, you know, know that one. Uh, but I remember people telling me like before that you could just like grab like a product image, put it on Facebook and it would just like rip at like a 5, 5 Ross, you know what I mean? Or whatever. Insane numbers. And then I feel like that, like forced creative to like, you had to be better then, right? But then it's like, you know, you could just do like. And then if you're doing like a simple three reasons why or whatever, just like simple USP call out statics and obviously they still work now. But like, back then that was just like, holy shit. Like, this is so new. This is like. And then obviously it's just become more and more, uh, sophisticated. But yeah, I mean, like, obviously all the, uh, you know, you've got like, Atria foreplay, all that stuff. And I do think that like really promotes like a copy and paste. And you hear about like, all these DTC brands that are like, you know, you know, the ridges, the hexclads, um, and you're like, wow, they're nine figures. They're crushing it. And I think all these kind of like, um, these apps and software is really like, promoted people to like, copy that and like, works to, to a point. So I think what really makes a difference in creative agencies now is like, people who can make content that people actually enjoy watching. Like when I Go into like ad accounts or whatever. It's like you see these like USP static headlines and stuff like that. But like it will be like the ugliest static ever. That's actually like carrying the account or whatever. Like not saying those don't work, but it's just stuff about like making content that really does show up in people's feeds because it's like if it's saturated for us, it's saturated for them. Is my kind of thought process is people can, that can like look at like, you know, organic tick tock content, YouTube, short stuff, and then like recreate that stuff. That's like the people that I see like doing, doing the best, focusing on the organic feel with like a hint of like direct response. So you see like the, you know, the Rosabella Moringa people, like that's like their, their forte and they've like absolutely done like insane numbers in such a short amount of time.
Speaker A: Yeah. Honestly, some of the best performing content we're seeing in accounts right now, um, is taking organic posts that did well and just putting them in the ad account. So just graduating in there. Um, and then also partnership ads. And I know I've seen some people in the Twitter space talking about partnership ads and you've. So I think I've seen you talk about that. But also you've been talking about uh, whitelisting ads recently. So like partnership being running from an influencer's page. But if I define it the same way as you whitelisting, like creating a page, like if you're selling um, wall art, then create like a home decor something Facebook page or Instagram page and just running it from that.
Speaker B: Yeah. I think obviously partnership ads are great and like definitely recommend everyone to use them. But like it comes at a cost, right? Yeah, like, you know, you got to pay these like creators or these influencers to do that. And I think there's like a lot of brands that maybe like can't do that or they wanted to kind of like they see the stuff and they want to dip their toes in. So that's why I kind of mentioned that page. It's just like a, it's a free way you can do it. You literally go into your business manager, create the page, make it relevant, create like a little logo. And even still that has like a massive impact on just like, like what I found. Like, you know, especially like, I know it doesn't matter so much but like click through rate cost per click. Um, and then ultimately CPA as well. And it takes like two minutes and you don't have to like pay someone. So that's why I kind of like started posting about that because I think it's like very easy to go, oh yeah, go get a partnership one, spend like an extra like whatever. I mean like, because the prices vary on it so much as well. Like some people like really reasonable, some people like charge like, you know, they're trying to ask for like thousands and like sometimes the value is there. But yeah, so that's why I can't post about that because like, you know, enough people are saying that I'm like, yeah, you can do it for free as well.
Speaker A: When you're doing whitelisted ads, do you um, do you as the agency own that page or do you get the brand to own it?
Speaker B: Yeah, it's tough sometimes like you will encourage a brand too. Because I never like to own assets over a brand, you know, I mean like, you know, like, you hear like agencies who like own the ad account and stuff like that, which is like, I don't even know how that still happens. It's insane. So like I'll always encourage the brand um, to like make it themselves. But if they won't at the end of this, like, okay, I'll just go make it for you and we'll just do it our end and then you know, run it. But like, I would always prefer the brands own their own assets, like especially if it does really well. And like, you know, for whatever reason they wanted to go like elsewhere where they can still use that page rather than us being like, no, we have that now and have leverage. So yeah, ah, I always try and convince them to, but as I'm sure you know as well, sometimes brand owners are just like, they don't see the value in it because it's like uh, whatever, I'm not interested.
Speaker A: Yeah, no, that completely makes sense. And for an agency or for a brand that wants to try that, it's easy to hear an idea and think like, oh, well, I could have to do this and this and this and that could take a long time. And um, do you feel like. So you said make a page, put a profile image. Do you feel like you need followers on the page? Do you need previous content? Like do you need any social proof? Or can you just rip it with ads?
Speaker B: I mean some people say like need to like warm it up. You can warm it up with like, you can just do like a like or follow thing for like mhm. Literally like you know, two, three dollars a day and just do that for like, you know, five days, but I've just ripped it before.
Speaker A: Okay.
Speaker B: And it's just like, from day one, like, because it's really in the name as well, you know, I mean, like, the name and the logo. So there was like, one we did where we had, like, I think, like, warmed it up maybe for like a week or so. And then I just made another one as well because I was just like, I don't know. I think I could do a better name than that that would, like, resonate with people more because, like, they're one, like the name that they had. So I think this is really important as well, is, like, it wasn't super over. Ah, to like, what the ad would be or, like, the demographic we're trying to call out. So I literally just changed the name to Living with X Problem and then that out the gate. And that performed, like, instantly better. Like, we didn't warm it up and it instantly performed better than that, um, than the other, like, whitelist page.
Speaker A: So ideally, the name of the page kind of relates to either the problem or the icp.
Speaker B: Yeah, it can do. I think there's so much nuance to that stuff. You know, if I went in and tested it more, I might see different stuff. Like, you know, I see loads of people do, like, you know, I obviously, like, wouldn't just recommend this to put out there, but they do, like, fake doctor, uh, like pages, you know, I mean, like, you know, and they're like, do like an AI, like woman or whatever. Like, you know, I'm not going to recommend people do that because I'm pretty sure that's like, fraud. Yeah, yeah. But, like, you know, I see people do that as well. And then, like, you know, obviously, obviously it crushes.
Speaker A: But.
Speaker B: So, yeah, I think there's, like, nuance to it. But I think, like, yeah, I would start with calling out some kind of, like, demographic. You want, like, you just want them to see this and be like, oh, yeah, this is for me. Like, the same way you'd make an ad where you want your cool. Yeah, this is for me. You see the name and you go like, okay, yeah, this relates to me. I want to know more about this. Or, like, I can at least relate to this pain or, you know, whatever.
Speaker A: Yeah, ask, because. So we're working with a drug rehab to try and get people who need to get off drugs into the rehab. So I'm just. We've been looking at ways that we can improve lead quality, um, and getting people in. And I think whitelisting could be A good opportunity. We, as an agency, we've done partnership ads. We haven't done. Excuse me, white listing here, uh, as of late, recently. So, yeah, I'm just thinking through how we could do that. But that's interesting.
Speaker B: I guess that could be, like. That would be like, a tricky, tricky one. Like, I feel like. I mean, I have no context about that, but, like, like, the client or whatever. But I imagine people with, like, rehab or whatever you would, like, want to create, like, a whitelisted page where it doesn't come across, like, preachy or anything like that. And it feels like, oh, this is a group full of people like me. And it's more like, you know, rather than being like, you need to get better, it's just like, oh, this sounds like a group of people that I've been through. I've been through. And this is just them trying to, like, help rather than, like, blame or whatever.
Speaker A: Yeah, my thought was it would be like, veterans, like, something related to veterans and, like, the experience. Or like, parents with kids that are dealing, uh, with addiction. Those types of things. Um, no, that's interesting. Well, that's something. I'm going to definitely share this with our team, so we could see what we could do there. Um, creative volume is the term for 2025. I'm interested to see what 2026 term will be, but creative volume has been the term for this year. Where do you stand on volume versus quality? Can you have both? As an agency working with a new client, uh, do you feel like one takes priority over the other?
Speaker B: Yeah, I've got, like. So obviously if I had to pick one, I would pick to do something, like, better. Um, you know, I'd always pick. Like, if I had to pick, I'd be like, well, I'd rather have less, but it'd be, like, really high quality. But if you can do quality at scale, like, that's also in my head, like, the ideal situation. I think, like, what you probably want to do is, like, an agency or brand is just like, have, like, set formats and, like, messages and, like, you know, say if you're an agency onboarding a brand, you've got, like, a big list of, like, formats, angles that have worked well for you.
Speaker A: Right.
Speaker B: That you can, like, replicate across certain, uh, like, different niches. Like, say, okay, like, okay, this ad format has, like, worked well across every. Every niche I've done. Okay, so, like, when I onboard song, I'm just going to go do that. And you can do that on scale, uh, at scale, and, like, do a Lot of volume and like, have good, like, you know, you're betting it's going to work in like, you know, seven times out of 10 it will, like get decent results. So then you can spend more time on doing like the big net new swings, trying something new that the account hasn't seen before. Because like, ultimately as an agency, depending on how you position yourself, like, you can position yourself as like full net new big swings. If it works, great. If it doesn't, you know, you pay it for whatever. But like, for like most of us doing performance, it's like you need to come in, get decent results from the get go or you're gone. So I think. And that's still beneficial for brands because you still can bring in like, you know, because like so many different agencies have like different playbooks or like, you know, they have like different ads that they deploy that have like different looks and feels and you can do that scale. Um, but then you can just spend more time doing like, you know, writing like those banger scripts maybe and like, you know, spending that like extra hour or two, like trying to find like a really good authority to partner with and like writing really like good prompts and scripts for them. So I think you want to do both, but I think like a lot of people just fall into this kind of like slop at scale. Um, whereas I think the stuff you want to be doing at scale is just like multiplying your previous winners across all clients and.
Speaker A: Yeah, do you, you uh, say multiplying previous winners across clients. Do you ever, like, say you work with three skincare brand, you bring on a new skincare brand as part of your onboarding or audit. Do you look at other accounts that have similar ICP or similar product type and try to see what you could replicate for quick wins?
Speaker B: Yeah, yeah, yeah. I mean, you always have like a bank of like your winners. Right? But it might even be like different. Like, you know, I think I first got put onto this by like, um, Sarah Levenger. Whereas it's like, you know, like targeting like mind states or whatever, um, or like, you know, you know, there's like stuff like, you know, steam and whatnot. And I just noticed like, you know, for example, steam worked really well in like fashion ads. So like, okay, we're gonna come on. When I come on, we're gonna try and find clips that really talk about how confident they feel all the time and then you're gonna make a mashup around that. So it's like, you know, it's like even Stuff you can do of that. Like what emotions work really well in our niches and do I have like a playbook around that? Okay, cool. I know people feeling confidence and like obviously in clothing that makes a lot of sense. Then you just kind of do that across all and I guess you could do that in like beauty. Um, but then, yeah, I think it's like you also got to be careful. It's like if the brand has like the same like target audience, um, and they're like competitors, it's like tough to like, you know, just I'm do that ad for both. Um, but if it's like different products and like, you know, or different target audience and yeah, just going to replicate it across.
Speaker A: You just described exactly what humans have as an advantage over AI, which I think needs to be talked about more, which is like AI is very good at large data set interpretation. When I say large data set, I mean like hundreds of millions of rows of data, like finding pattern recognition and then calling that out. Right. Humans are good at. I worked with this one brand or I worked with these two brands in the past. I learned this from it Esteem worked well with fashion and I learned that it worked well with this gender and this age range and this product type, et cetera. I'm going to try to do that for this brand as well and see how it performs. That's what AI is very bad at. And that's why a lot of times you'll get like overgeneralized recommendations for strategy from AI versus like what a human can actually output.
Speaker B: Yeah, I mean AI, ah, things like, interesting because it's just like, it's just such like swings and roundabouts. People like one week are like, it's gonna come for all of you. And then the next week just like, oh yeah, like every creative agency is like doing ah, like AI slop or like whatever. Um, but yeah, no, I agree. I think, and I don't know, I think people use, in my opinion, like AI kind of, kind of wrong in that sense as well. Like, you know, I think they try and use it to kind of like create everything rather than just be like. So like, for example, like what I'm working on it alone now as well is like, how can I get it to do loads of like data collection for like research for me rather than. So like, for example, like, you know, I do it a lot with like Reddit right now where I get it just to scrape Reddit for me based on keywords. But then I still read through the reports that it gives me. So it's like you still save, like, you know, 45 minutes, like, depending on how far, like the rabbit hole you're going to go. But you're still like, doing the manual parts, but you're just trying to shave, like, half an hour of blocks off or, like our blocks off, like, you know, get it to, like, scrape, like, organic tik toks in, like, the niche or whatever so I can find out, like, what content. So rather than me, it's like searching up, uh, screen, recording it, putting it through Gemini or whatever. Like, you can just get it to do all that manually, but then you, like, uh, automatically. But then you still have, like, that kind of part where you're doing things like manual. But I feel like everyone's just trying to outsource, like, everything to it. And I. Going back to your question, this is like the biggest moat I think, that agencies, like, have right now, especially as creative, is just like, spending like, 10 more time on something or putting in 10 extra effort. Because, like, naturally, as humans, we want to just, like, outsource all of our thinking to, like, AI. Because thinking is painful. It causes, like, discomfort. Especially if you're like, oh, I can't figure this out. So you would just take this thing that sounds okay and run with it because, like, that's less painful than doing the extra 10%. So I think, like, obviously that's going back to there and kind of like derailed, maybe what you asked. But, yeah, I don't know. I've got lots of thoughts of, like, advertising and AI and think, like, a lot of people like, using it. Right. But there's also a lot of people that like trying to outsource, like, too much, and then you can see that in the results they're getting. And like, you know, I think like, the general kind of like, dissatisfied. Dissatisfied. I can't even say that word. Um, just like how dissatisfied a lot of brands seem to be with creative agencies at the moment.
Speaker A: Yeah, I think, uh, the way I always frame it is try to figure out what the junior level of a role would do. So junior creative strategist, junior media buyer, AI can replace most of that right now. So, like, data collection, aggregation, um, um, like putting together reports to a degree. Uh, but when it comes to actually thinking through next steps and strategy and stuff like that, that still needs to be in a person's court. Like, that still needs to be something a person is doing.
Speaker B: Yeah, for sure. And like, you know, I guess there's like, a lot of people talk about copy as well. I Think for me, the thing I struggle with it the most is probably like, I don't know if you've seen this, but it's like headlines like, I can get it to write like, Claude to write a pretty, pretty good script in like, you know, not very much time at all. But. And it really depends on like, the context you give it. So like, you know, I think like, you've got to be making like, you know, custom Claude projects. Like, kind of like how, how I do it and know how other people do it is like you make one per like asset type. So like, I'll have one that's like a script writer for like, you know, like AI voiceovers, say. And then like, I give it like loads of different modules to then like potentially like run through. So I think it's like one of them I played around is like, has like eight modules, but I only use like three at, uh, a time to not like overload it. Like, you know, you put all like the CSVs in there. You give it like, Persona, um, context and then kind of like, you know, give it some like, winners or whatever just so it can like do that. And I think like, if you do that, you can get like a pretty good script in like 10 minutes after that. And then it's just a case of like, you know, being like, oh, ah, make the unique mechanism, metaphor or like, you know, working on the hook. And then that's pretty much you can just do that and spend your. Spend your time there. Um, so, yeah, I don't know what you found, but like, I found like, still with like, the copywriting is like so many people like, oh, yeah, it can't produce good copywriting. I think if that's your firm belief, then you're probably not giving it the right context. But then the only one I still can't seem to crack is like the headlines.
Speaker A: It just. Yeah, I do think, I think it's, um, easier to correct someone else's work than to start from a blank page. Which I think is one benefit of AI is because it does help regardless of if you're using. It's a hundred percent complete end result for like, headlines or whatever. It is beneficial to use that as a jumping off point. Have you found that that's helpful to at least like get the headlines or get the copy and then like, you go through and change it versus, like trying to start completely fresh or do you start fresh?
Speaker B: I think it depends on like, depends on the day, honestly. Like, I go through like weeks where like, I always try and make some effort to, like, write ads completely from. From scratch. Um, I don't know, maybe people say, like, that's. That's stupid or whatever, but, like, you know, like, at least, like, you know, once per. Once per week, I'll just have a day where I'm, like, I'm writing all the headlines myself. I'm writing all the scripts myself, um, just to keep myself, like, you know, keep. You know, keep my mind sharp. Yeah. Because it's like, then I know how to, like, copy chief Claude or whatever. You know what I mean? Uh, and if I can write like a. Like a banger without AI, then I know I still got it, you know, I mean, so, like, um. Yeah, sometimes, like, a lot of the times I use it just like. Like you said, like, it's like a springboard. And then I'll just, like, give it, like, very minor tweaks. Because, like, honestly, I think for, like, AI voiceover stuff, it's not hard to, like, get it pretty much, like, you know, an 8 out of 10 instantly if the. The context is right. Um, but then with static headlines, I still don't. Still don't use it a ton, just because, like, I don't know. It's just the results aren't tough to do. Yeah.
Speaker A: Well, I have to check when this episode is going to come out. We're currently recording it September 19th. Um, we said creative volume was the keyword or the buzzword for 2025. What do you think is going to be the industry's buzzword or focus for 2026?
Speaker B: I, uh, don't know, man. I don't think I've ever made a prediction that was right. So I'm not going to 2026. I think it's just going to come back. I don't know. Maybe I'm wrong, but it's going to come back to quality, because everyone's talking about quality now. Like, I feel like every tweet I see or, like, whatever, like, oh, I think it's like, you know, obviously the algorithm kind of, like, favors just, you know, blanket statements, but I think, you know, with the kind of like. And this is what kind of, like, prompted me to start a creative shop as well. It's just, like, it seems very saturated, but, like, with the amount of, like, complaints I think people have, I also think it's kind of brands that have done this as well because they've asked for so much volume too. Like, you've kind of, like. And I think I made a tweet about this. I was like, I actually think brands should be responsible for most of their volume because yeah, if you can copy and paste like the formats that you know, work well and you can produce that on mass volume, like surely you'd want like the experts to come in to make that harder content that could unlock that scale. Like, because I think basically brands like, and I'm not even like you know, talking, talking on brands because that's the people that hire me. So that's, that's not what I'm doing. But like it doesn't make sense to me to like favor uh, getting people to like mass produce stuff and like you can already do of them into like middle of funnel, bottom of funnel, like kind of like ad concepts. Because like I'm sure you've gone into some ad accounts and you'll see like three, four agencies in there and they're all just doing like bottom of funnel content on like mass scale. Like whether it's like an offer about like a voiceover or offer on like a um, like static, like because that's really easy to produce at volume. But then as soon as you start trying to do like the real diverse creative stuff that takes four and like maybe you're trying to like unlock a new Persona and you're like, you know, trying to like write a banger script that's going to like, you know, spend like hundreds of thousands. Like that's, I think that's really hard to do at scale. Um, I don't know if many people are doing it. Maybe, maybe they are but I feel like you need some like, you know, really solid operations for that. So I think like, yeah, I think brands and agencies are much to blame for that kind of stuff. So I think it's going to come back around next year and then like it'll be the case. Like, and I think it really depends on like how AI develops as well because I'd be really interested to see like brand's going to be asking or like agency is going to be like going back more into human content or are we going to see the absolute, you know, dive into like AI content? Because there are a lot of people doing AI creators at scale.
Speaker A: Mhm.
Speaker B: But it's. Yeah, I think personally you're going to see like brands wanting agencies to be producing higher quality work and like more fallout and you know, but I could be wrong but that's basically me being biased because that's what I'm trying to do.
Speaker A: I was just talking with two agency owners in a group chat yesterday and it was interesting. They Were talking about how they're going to test shifting from a monthly retainer to focus on like here's the creative amount we'll give in three months. Like this is just focus on quantity of creative over a period of time. And what you just said made me kind of piece that together with um, what the chat was yesterday of almost. It feels like the monthly retainer model might limit the ability for agencies to come in and focus on achieving a goal of unlocking a new winning um, concept or unlock a new winning icp. Because it's very much about like what did you do for us this month in terms of performance and spend towards ads and stuff like that. So I'll be interested to see if agencies also start to pivot towards like more goal based partnerships versus like monthly retainer. If that makes sense.
Speaker B: Yeah, I think, and I think that probably comes back to the thing that I was thinking about earlier is like do the like, do the stuff that you know, can like work well. Like and uh, you've got like proven stuff now and it's like easy to make do that at scale because like I said, when you come on and you're like hedging your bets to like produce some like ads that are going, yeah, quickly you can do that. But then like you've got to make sure you spend time on like the top of funnel content. Like you can't, I don't think it's like one or the other. You've got to make sure you do both. And then that way like still the retainer model can work really well or like, you know, and I, I always do like a monthly minimum but like everything I do is like percentage of spend like memory saying that performance based because like I want to come in and be incentivized to make like you know, a six figure banger. You want me to do that. But also like I'm not really incentivized to do that if I'm only going to get like five grand a month or whatever, you know, like I'd want to get, I want to get paid like 10k plus for that as it like scales up. Because if it unlocks that scale then I think you should pay the person who makes it. Um, and to be fair, pretty much every brand I spoke to is absolutely fine with that. So I think, I don't know, I don't know how many because yeah, what do you do? Do you do like a percentage spend model as well? You just Flat retainer.
Speaker A: We have not, we've done flat retainer. After talk with you in San, uh, Diego, I thought about it. I haven't put action to it, which we're almost a year so I'm kind of lagging on there. But it, it's an interesting thing to test out that I still think we should on our end.
Speaker B: Yeah. Because I think it's like I'm always like about like setting intentions and stuff like that. And like, I think like, even though it sounds like maybe a bit like woohoo or whatever, but it's just like if you go into it going like, okay, if I make a six figure banger, I'm gonna get paid for that. Uh, then you're going in with the incentive to do that rather than like if you go in, you go like, okay, if I make a six figure, seven figure banger or whatever, spend banger, like I'm just gonna get paid the same. And I think like, whether we realize it or not, like just having that intention is going to change how you view the energy. Because like, um, I'll work on a Saturday or like do whatever like in the accounts. Because it's just like if I unlock that next big thing, like I could get paid like a lot for it and that's like still beneficial for both of us. So like that's kind of why I do it because I feel like it just sets the intention of just like I'm gonna make a banger.
Speaker A: Mhm. No, I like that. Um, well, what's the name of your new shop? What's the name of your agency? How do people get in contact with you and what type of brands you working with?
Speaker B: Yeah, so it's actually called uh, Roly Poly Digital. I haven't announced it yet at all. Uh, fun facts about how that name came up. Um, it was actually Andrew Foxwell's daughter Nora who came up with that name.
Speaker A: Oh really?
Speaker B: Uh, Shout Out Nora, man. Uh, yeah, she just came up. It's like, yeah, I'll call it that. M. How's getting contact with me? I guess like Twitter. Um, I don't know when this will like air. Ah. But like as of right now, I don't unfortunately like have any space. Uh, but that might change when this, this comes out. That's why I'm not posting at the moment because I'm just like, I don't, I can't like take anything on and like, I feel like the last thing I'd want the brands to see who just started working with me is just like posting all the time, trying to get leads rather than them being like, why is this guy in our account doing work, man? Um, so, yeah, again, like, I don't know how, uh, when this will end. That. That might be different by then because I'm like, you know, working on onboarding some new stuff and whatnot. But, yeah, just. Just Twitter. Uh, I don't know. I don't even know what my handle is on Twitter. I think it's Harry Delmage. Um, there's a photo of me and my cat. So that's how you recognize it, I guess.
Speaker A: Well, Harry, I appreciate you coming on, sharing all this insights. I walked away definitely with something we can do for white listing for that one client. Uh, but I'm sure listeners walked away with a lot as well, so I appreciate you coming on and look forward to staying connected.
Speaker B: Yeah, no, good, man. Um, it's been a pleasure. Have a great rest of your day and I'll talk to you very soon.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.