How AI Is Changing Video Advertising | Scott Salik of Carpe Canum | The Brainiac Blueprint Podcast
- Acy Rodriguez
- Oct 28
- 37 min read
On this episode of The Brainiac Blueprint Podcast by Left Brain AI, we sit down with Scott Salik, Founder and Executive Producer at Carpe Canum, to explore how AI is transforming video advertising - making high-performing, personalized creative accessible for small and mid-sized businesses.
With a career that spans from producing Anchorman and Team America to launching streaming campaigns for local brands, Scott shares how his team is using AI to scale video personalization, automate focus group testing, and keep creative quality high without the bloated costs.
We dive into how high volume personalization (HVP) works, why localizing ads boosts conversions, and how synthetic focus groups are reshaping creative feedback. From legacy film stories to real-world AI workflows, this episode is packed with lessons for marketers, creatives, and media buyers alike.
Whether you’re in paid media, brand strategy, or just want to future-proof your ad creative - this one's for you.
Full transcript below.
🎧 Watch or listen to The Brainiac Blueprint Podcast
Youtube/ Youtube Music: https://youtu.be/nP2hGuGwopU
⏱ In this episode, we discuss:
00:00 – Intro
00:20 – Meet Scott Salik & Carpe Canum
02:00 – From Network TV to Streaming Ads
05:22 – Why CTV Levels the Playing Field
07:57 – Targeting Households by Email List
09:16 – What Makes a Good Ad in 2025
11:53 – High Volume Personalization (HVP) Explained
13:51 – Localizing 240+ Commercial Variants
18:04 – AI Focus Groups & Synthetic Participants
20:49 – Chat-Based Feedback from Virtual Testers
25:24 – Human Empowerment vs. AI Replacement
28:32 – Standing Out in a Crowded Ad Market
32:32 – The Future of Personalized Video
40:02 – AI Tools Scott’s Team Actually Uses
43:51 – Working with Robin Williams
47:45 – Live TV, Mistakes & Hulk Hogan
48:56 – The Risk of Overlooking AI Mistakes
🔗 Scott Salik
Website → https://Carpecanum.com/
LinkedIn → https://www.linkedin.com/in/scottsalik/
📲 Connect with Left Brain AI
Instagram: https://www.instagram.com/left.brain.ai/
X (Twitter): https://x.com/left_brain_ai
📣 Subscribe & Share
If this episode inspired you, taught you something new, or gave you a different lens on AI in video production- share it, leave a comment, or tag us. Let’s help more people stay brilliant.
Know someone who would make a great guest?
Email us at podcast@leftbrainenterprises.biz
Episode Full Transcript
Kyle: All right, welcome everybody back to The Brainiac Blueprint. We have a very special episode today as we're joined by Action Hero partner and, dare I say, should I be so bold to say friend?
Scott: Yeah, we've been doing a lot in the last couple of years. Well, we're 3,000 miles apart, but I still think of you as a friend.
Kyle: I agree - a Slack and Zoom connection kind of friendship here. So we're joined by Scott Salik. Scott, thanks for joining us today.
Scott: I actually might have a shoot in Philadelphia, and I accepted it because you were on the list of people I want to have dinner with.
Kyle: Oh, you have to let me know. That’d be great. Let’s get that scheduled. I’ll take you around to Reading Terminal and all the good spots.
Scott: Perfect. I’d love that. Awesome. So we’ll just have to figure out a date.
Kyle: Love that. All right. So Scott, you are the founder and executive producer at Carpe Canum. People can check you out at carpecanum.com. Per your LinkedIn, you guys do affordable end-to-end streaming TV and video production. A little teaser for everybody - you’ve got a new product that I’m looking forward to diving into. But first, if you don’t mind giving a quick intro about your background and Carpe.
Scott: I’m Scott Salik. I’m a partner in Carpe Canum Media. We do low-cost commercial creative for streaming TV. We build commercials targeted for streaming advertising. Our main distributor is TV Scientific, and we help fill their commercial load for customers who don’t have ads. We can build the commercial and manage the campaign, or just build it and you manage it yourself.
I’ve been in the business a long time and have produced content for every size screen - from old 3×4 TV all the way to high-def 16×9, and even feature films like Anchorman, Beethoven, Fluke, and The Edge.
Kyle: That’s amazing.
Scott: Team America too - that’s another one people love.
Kyle: I was doing a little bit of research. I was scrolling through your LinkedIn, saw that you were like a runner for Disney and all that stuff.
Scott: When I first started it was just ABC Television back then.I ran during the Olympic trials and all the other Olympic events out of college. And I was always on, you know, I guess that started my, my lifelong love of technology. So I was always on the tech units of any kind of shoot.
The biggest one we did, I used to do NFL, USFL football, and I did baseball and some golf, a lot of golf. So we all had all the technology that came with covering those widespread events. But I worked on the US Olympic trials in 1984, I'll date myself. And we had one of the first electric vehicles with a camera mount on it that we had to bounce off a helicopter that was flying off of that was flying above and then we had to bounce it off a tower and then back down to our home base. So it was all the microwave technology, all that kind of stuff. And the funny thing was the cart had a gas engine and a battery.
And when they built and designed, and the idea was that the runners wouldn't have to have exhaust in their face. And when they built the card, they never figured in the weight for the camera and transmission gear. So after about the first five miles, the battery ran out on the car and it was gas the whole rest of the way.
Kyle: Oh, my God. I wish I could have seen all of this, you know, in motion. I'm sure that was quite the scene
Scott: It was funny, funny, funny stuff. So, yeah. And so, and so, you know, I know you wanted me to ask this question. So it's, I think AI is a blessing and a curse.
Kyle: That's great. Yes, I would agree with that. I think a lot of tools can be used for good and for bad, and I think we're seeing that with AI.
Scott: Well, I think also people have a lot higher expectation as to what it can do than what it can really do right now. I think it'll do all that stuff great. Everybody's seen the Popeye commercial and a lot of the other full AI commercials that have come out. But those are big agencies with big budgets producing those with lots of processing power and data scientists who know how to write the prompts.
Kyle: Yes. Yeah. There's still some kind of quick fix tools and then there's the highly skilled, highly techie kind of ones out there. And you got to know which is which and which is going to help you the best. Yeah.
Awesome. I think that's a great starting spot. I wanted to jump in a little bit about Carpe a little bit more just so people have a better understanding about what you guys do. So, you know, what's the story? What inspired you to launch a performance-based video agency like this? And, you know, what's a little bit about your philosophy when it comes to creative?
Scott: So, you know, there have been two times in my career where I really felt like I could make, there was like a life-changing, amazing technology, amazing opportunity that I was going to be involved with. And the first was like, I worked at a company called Beachbody and they're with a leading provider of fitness videos, Tony Horton, Power90, you know, P90X, yo..
Kyle: I did some P90X back in the day
Scott: Yes, maybe that's why we're friends. And when I came in there, I was hired to build the internet side of it and all the sidecar content. And that was, and at that point, YouTube was a year old.
So we had to build everything out in order to make that work. So all the streaming network, the web interface, everything. But that opportunity to be able to provide fitness information at that level, I thought that was an unprecedented opportunity for me.
So I took that job at Beachbody and we built all that out. And then when an old friend of mine was telling me about what they were building at TV Scientific and the ability to target commercials directly to households and basically make TV advertising available to any size business. I thought that was an amazing opportunity because for the most part, originally, you know, until recently, 500 companies did something like 80% of the TV advertising in this country.
And most small companies couldn't never afford to do it. But with the ability that streaming offers, and targeted advertising offers, any small business now has that same superpower of television to go ahead and use it to grow their business. So I saw that opportunity and I was finishing up at another company and I jumped on it. And it's been an interesting evolution in the last five years of how much that's changed in the last five years. It's crazy.
Kyle: Yeah, I think it's super interesting. I think a lot of people think about this big widening gap that can happen a lot of times with tech and things like that. But I think streaming, similar to AI, has been one of those kind of equalizers that allows the little guy, the local shop, whatever it may be, to get some exposure. And one thing that you and I have talked about a lot that I've really found fascinating is targeting capabilities that are available in there. Do you have anything that you want to share there? Any kind of call outs about the targeting? Again, I think as a paid media guy, it's a little bit different and pretty powerful.
Scott: Well, what we can do is we can, if you're a small business and you have an email list, we can target to the homes on that email list. And hit about a 90% match back to the address, to the physical address, to the email address. And then you can build lookalike audiences that look similar to your customers and then just target those households with that. And the cool part about the TV scientific platform, at least, is you can attribute that impression. So we look at from the point of exposure of the commercial all the way through to the purchase or website visit, wherever you consider a conversion.
And that gives you like an amazing powers of small business, a small or medium sized business, right? And what we do is we, TV scientific came to us and said, can you produce low budget, low cost, not low budget, low cost commercials, so that every business has a chance. And so that's what we do. We create the low cost commercials. And if we bring in the client, or like when you and I work together, we help build out the whole streaming targeting platform of it all, a plan of it all.
Kyle: And one thing I want to kind of clarify is, you know, when you say low cost, it doesn't necessarily mean low quality. It just means cost effective. You know, and, you know, we all get hit by a ton of content and we know immediately what isn't good and what isn't authentic. But I think you guys do a great job of being able to have, you know, creative integrity and have a clear message. So I'm curious what your process is and, you know, how do you ensure like what's is a good ad to you that you want to put out there on these streaming networks?
Scott: Well, in streaming, think of it as a direct response medium. It's not a broad net anymore. It's very specific. You talk directly to your customer. And I think that's what makes a good ad. You talk to that audience that you know you have and you want to reach.
And, you know, from our perspective in creating a commercial, we like to ask the client what they have and we'll work off of those assets, the existing assets to save them money. You know, it's important to not spend so much money on your ad that you can't afford to build out a budget to be able to get the ad to your customers. Otherwise it's useless. You want to have a frequency of around seven before someone will really start recognizing your product. And, you know, in order to do that, you have to have some money to be able to target your audience.
Kyle: Of course. Awesome. So I'm curious, you know, budget obviously is a big thing as well as just, you know, operational standards and all this kind of thing. So I'm curious how you kind of pivot your strategy and your management of clients when it comes to, you know, a small local brand versus, you know, like a large national type of company that we all already know.
Scott: It's all the same, kind of. I mean, except with large national brands, you have a lot more voices, right? Yeah. A lot of times because of our budgets, we have to say one voice.So you guys can take all that feedback, but one person has to transmit that to us. You'll laugh about this, but we use frame for client feedback on the different edits, on the different cuts, right? And the different versions. So we had a client arguing on like they were literally arguing with each other in the notes. And one of the people on the team got fired in the notes.
Basically, it was don't listen to him anymore. And then that person was gone on monday. It's like and then that makes that made us even more reinforce the one voice thing so whether it's a big client or a little client we want one voice to comment in on it so so that helps that that also help manages the budget you know it helps keep costs under control.
Kyle: Awesome. So, Scott, as you know, this is kind of an AI podcast here. You know, I love AI. I think it's very cool, very powerful. And the reason that I wanted to have you on is because of this new product that you guys have been developing and testing and really getting out into the ecosystem. So can you tell us a little bit about it? What is it? And, you know, what problem are you solving with it?
Scott: So, we just launched something called high volume personalization. And what that allows us to do is take a commercial, whether we build it or someone else builds it, and modify it, create a bunch of different versions for each audience segment you have. So whether it's 10 versions or 800 versions, which we have a client talking to us about right now, we can do that at a really cost effective way.
And you can add in AIVOs if you want to. So if you want to have a different VO in each one, we can build out an AIVO. The AI helps manage the process so it's not all manual anymore. When I was younger, we used to have the executive producer would do a commercial and then they'd throw us in an edit bay for a week at night making all the localized versions. Well, now we can do that in 24 hours - with a certain amount of automation. And it really allows, it's another one of those things that allows a small business to have the same power as a major brand like Ford or Lincoln or whoever.
Kyle: What does your QA and your accuracy process look like? Is it kind of like building it out into the system to make sure that's right? Or is there a, you know, I'm sure you guys aren't going through and watching all 800 versions, right?
Scott: We're spot checking them and you can usually see an error if there's an error in the pattern. You know, sometimes it'll flag a version having an error, but we spot check them. You know, we'll look at every 10. I've got one of the team members look at every 10. You know, that might increase or decrease depending on the problems we see. Right now, it's pretty accurate. It's pretty, we haven't really seen a lot of errors. It's pretty reliable.
Kyle: Very cool. Very cool. I know you and I have discussed this, but for some of our viewers, can you elaborate on some of the reasons that people would have so many variations and, you know, why it is so useful?
Scott: Yeah, I mean, let's just do the most basic one. Think of localizations. If you have multiple locations, you can customize the special, the call out, the products that you want to offer in that branch of your store. You can change out the end card so that it calls out the specific address. So you could then target that. You could say, go to your neighborhood action hero store. Right. And in that you could say, change out just the end card that says 16 Main Street, Philadelphia.
Or, you know, just to do that and localization tends to people feel like, okay, this is a local business. I want to support them and they're nearby so it's convenient. I read a study that said like most people won't drive more than five miles. I think it was five miles to a business. You know, that'll be their priority of the distance they're willing to go from their home. I can find the data, but it was, I think that's what I think it was five miles, but if you can make that commercial look like it's a local store, that they know where the location is, they don't have to think about it, they might be more apt to go.
Another thing is you can look at your audience data from your website and you can target products to those specific different audience segments that those people will be more likely to buy.
And so you can customize the commercials like that. You can change out the special, the feature and benefit. I'm laughing at your dog, putting his little face in there. You know, you can change out the promo code, right? And you can do promo codes. What we did for one, we did a run of 240 plus ads for a casino app client. So this client is available in 40 states.
Every state had a different disclaimer line on it they wanted to we did a national spot and then they wanted to localize it because they their data shows that if it's localized for the state there's more app for people to use it because they compete against each other in the game and and people like to compete against people who are in their area so so we created 40 state versions and we changed out the graphics and the vo in each of those 40 and then they wanted to test five offers for each of the states so then we had five versions of each of the 40 spots.
And I think it was like 240 some odd spots 241 some odd spots for that program. So and then they were able to see what offer work best in what market and push that at full speed. And you know, you can get data back and in pretty short form, you know, a week or two, you'll have your data back and what's working, what's not.
Kyle: That's so cool to me. One of the things that I've always enjoyed about paid media, whether that be search or social or anything like that, is the availability of data and the availability of quick testing so that you can learn messaging and offers and all that kind of stuff. And you can kind of carry that to other mediums. And I think, to your point earlier, one of the historical challenges of commercials is the cost prohibitive nature of it. But now that you can kind of again, test and you can learn and see what offers are working the best with these demographics or these geographies. That's just the whole other ballgame. It's very cool to kind of see and hear it in action.
Scott: Yeah, and if you're in something with like a franchise business, I mean, think about what you can do for your franchisees. You can offer them their own commercial package for their market. And then literally target, then you can literally target within five miles of that store all the way around, you know, for the demographic that wants that product. And it's pretty amazing.
And think about how much money you can save by not just throwing this giant net over the entire city of Philadelphia when literally, you know, you know what street, what area that store has customers from.
Kyle: Very cool. Very cool. Anything else you want to share about HVP before I kind of jump into some other AI and creative related stuff?
Scott: No, but this is probably a good bridge. So did I tell you that we're doing this AI focus groups, generative AI focus group testing now? Yeah, we're just launching next week. So the you know, it's kind of funny. I would never launch two products within two months, but it's just timing that we had access. We got access.
Kyle: So you're sleeping well right now.
Scott: Yeah, I know. It's like not that's that, that's why I almost canceled this today. Yeah. The, um, but I'm in the Producers Guild of America and because of that I had some information about a platform that originally had been created for feature films and TV shows to AI focus group test scripts and budgets and concepts, right?
And they recently launched a brand module. And so they were kind of in a beta test of the brand module. So we were able to get in on that early. And M&M Mars is also on the platform as one of their big people, one of their big clients. And they're parallel testing the generative AI platform creative focus group testing and the real audience focus testing. So synthetic audience and human audience. And they're getting pretty parallel results on the focus group testing at a fraction of the cost.
Kyle: I have to say I'm a little disappointed that I'm not involved in that test area. You know, there's those focus groups. If you ask my family, they'll they'll confirm this. I was obsessed with M&Ms, so I could have been a really good voice, you know, to provide some feedback there.
Scott: I'll see we cross paths with the folks who are doing that and I'll let them know.
Kyle: Perfect. Perfect.
Scott: It's funny because did you ever do dial testing? Do you remember dial testing?
Kyle: I don't think so. No.
Scott: So I used to when I was doing stuff for HGTV and DIY networks and Food Network, I would sit on the dial test for the shows I was working on. And basically, you had a bunch of people sitting in a room turning a dial is what they liked and they didn't like in throughout the show.
And then you'd get all these scores. You'd see all these meters going up and down on the screen and you'd get kind of a copula of score. And then, and then you would then interview the people, uh, about the, about what they like, why they said what they do and stuff like that. So that was like a good kind of background for me on this. And this is a very similar process. We can talk, we can look at the answers from and the scores from the synthetic participants. And that's so weird to say that. Synthetic participants. And then we can actually chat with them.
And ask why they, yeah, it's crazy. And like we just went through a test we did for college, a small women's college, and the buying score came out low, but the how likely are you to go visit the website score came out really high. And that seemed counterintuitive to me. So when we went and we interviewed, when we went and chatted with the synthetic participants, we found that they considered buying to be enrolling.
And all of the commercial was asking for was to learn more. And 90% of those people scored on a scale of 10, 9 out of 10 that they would go to the website. You can't. I mean, that's like amazing numbers. And on other stuff we've seen, we've gotten really we haven't scored that. That's the highest we've scored on anything.
But it's been a good learning to see how the system works. And by the ability to chat with the synthetic participants, they've allowed us to get information on why these answers that intuitively seem wrong to us.
Kyle: So interesting. So, I mean, the concept of interviewing a, and just the term synthetic participant is, is wild to me. That is very cool.
Scott: And the cool thing about this is, is that they did during the pandemic, they did like thousands of paper surveys with real participants to build the generative AI database. Is that the learnings, the database, whatever you call it. And that's how they were able to build out this product. And it's really, and you can see it getting better each week. You literally right now, it's very strong. And our clients love it. Like, they're loving it. It allows clients to go to their bosses and say, here's why we should do this. Right.
They don't care that. It's a fraction of the cost of what a real focus group would do because it's small business.
Kyle: I'm sure. What are we calling this product?
Scott: I haven't called it anything yet. But it's a generative AI focus group testing. Hey, you're marketing. I'm terrible at marketing ourselves.
Kyle: I mean, I think we need to go to a focus group to figure out the...
Scott: That's what we should do. We can actually do that. We can actually test. We have this guy, one team member named Emmett, runs, manages that platform. And I'll ask him to do that.
Kyle: I love that. You have to let me know how it goes. Is this product available for everybody?
Scott: Yeah, starting next week it'll be available. We're kind of light releasing it right now because we're heavy into this high volume personalization and talking to people about that. But yeah, anybody can ask for it. We're talking about it.
Kyle: So this feels like a conscious effort to be leaning into AI. Do you and the Carpe team kind of have, you know, strategy sessions saying like, how can we get more efficient or offer more services like this? I'm curious, you know, two products in, like you said, two months isn't necessarily a coincidence. So I'm curious how this has all come about.
Scott: Well, as they're presented and you kind of have to look and see how they fit into your model and what would benefit your organization. So that's kind of how it's happened with us.
Kyle: So is this like a solution has come to you and said, like, we have this technology and we can integrate into your-
Scott: Yeah, kind of in this case, yeah. And on the high volume personalization we we did a test a few years back for tv scientific and the and the and the the platform never really did what it was supposed to do so it got killed off so I knew from that that there was a demand for it because we were really trying to make it work maybe two years ago and now yeah it's so funny it was off a platform now it's now we can do it on local on local uh on local workstations we can do it you know with with some computing power in the background.
But you know we kind of you know have a problem to for a solution you know it's like we're not in a rush to do any ai we have an ai commercial building low-cost commercial building platform that a lot of people are using but we see a lot of those commercials because we know what the templates look like but what we do is we lay a human layer on top of it and most of them most of the ones we've seen it's just the templated commercial and you know I can tell you from experience a lot of those templated commercials have a pretty low success rate compared with the ones that you have some more human polishing on. We always add that polish layer. And I think that that's an important thing in any AI integration now.
Kyle: So you literally read my mind there in terms of, you know, where I wanted to go next. So one of, I think, an interesting, you know, concept or perception that needs to be overcome when it comes to AI is people thinking about, you know, the replacing of jobs and just replacing of humans with AI. And that's certainly happening. I'm not trying to say that it isn't.
Scott: I'd hate to be a voiceover artist right now. You know, I respect that they do amazing work. But, you know, when you're in the low budget land, you know, a certain amount of our work now is done on AIVOs. And they do a really good job. And that's something that gets better every week.
Kyle: Absolutely. Absolutely. So, you know, the way that I try to frame it is thinking about it as human empowerment, you know, versus complete and total replacement. So, you know, it seems like you're on the same page there.
You know, I guess how do you kind of like draw that line of, you know, this is too much AI and we need to get more humans involved? Or, you know, do you kind of have a process there? Or is it just like you said, like, there's a problem, we have the solution, let's test it and kind of figure out, you know, where to go from here?
Scott: It's like the song goes, day by day, oh dear Lord, three things I say. There you go, there you go. It's changing so fast, you can't do that. You have to kind of say, this is what we are able to digest as a company. And my business partner, he is on the post side. He manages like a lot, all the posts and he is using AI to clean up stuff and clean up audio. And he's an audio, his background is audio. So for him to jump on the AI stuff, it's like, you know, he went, he didn't go begrudgingly, but he sees the value of the tools and the speed that it picks up.
So we're, you know, using it as we have a challenge. And if we see an interesting thing like the AI focus group testing and the like, I spoke on another webinar, and that's how I found out about the product that allowed us to do the high volume personalization, but no one else was doing it on a CTV level.
Kyle: Oh interesting
Scott: There's a company that does there's a big company that does it on the commercial level that does it all with ai and it's it's a major you know big time backers all that kind of stuff well you know we're trying to help small and medium-sized businesses we could do it for a big company i'm not I don't push that business away but you know we we're trying to make it cost effective where we can make some money and they can grow their business.
Kyle: Yep just more accessible to everybody yeah
Scott: I think that's the beauty of AI is it makes stuff available to people.
Kyle: I agree. Yeah. I agree. Side note, can you hear the thunderstorm happening? I feel like, you know, we've got an interesting, like this, okay, this podcast is doomed right now. I've got a thunderstorm and rain coming.
Scott: There's change in the air. That's what it's saying.
Kyle: There's always change. There's always change.
All right, cool. So I want to jump into ad agencies in the AI area. So I'm curious how you think about Carpe and how you guys fit best in this very oversaturated landscape. There's a million ad agencies, myself included, and a million creative agencies. So what is your positioning? How do you guys fit best? What is your kind of selling point that you guys continue to push out there?
Scott: We're a production company that understands streaming TV and targeting. Having been an early investor in TV scientific, we understand the goals of streaming. I was with the CEO yesterday talking about the high volume personalization and also the focus group testing yesterday. I was actually teaching him about them, what we were doing. So we have a lot of conversations about the streaming, the audience targeting, how their new product, which is a guaranteed outcomes product where you pay for your outcomes and how that works and how that fits into it and how creative integrates into that. So we have that relationship. We work with other platforms. We've supplied commercial for Mountain, we've supplied commercials for Ribeye, some of the other ones.
I don't want to diss Vibe, but I'm going to diss Vibe because that's a self-serve platform with a self-serve commercial. And we get customers who have been on it and have gotten no results out of it. And it goes into that, well, you need to have, you can't just, everything can't be just automated. It's not to that point yet. And we get a lot of those and then a lot of those people will come to us and say, well, why should we go with you? Because we didn't have any results on Vibe.
I think that you get what you kind of pay for. And again, Vibe does a great job for a lot of people. Don't get me wrong. But for, you know, when everything is automated, I think that the consumer realizes that and they're not interested in that. That commercial doesn't grab their attention.
Kyle: Again, it comes to that quality, that authenticity. There's still a need for that human touch to make sure that there is good quality and everything involved.
Scott: Yeah. And I think there'll always be a need for some human interaction with that to make it better, to improve the app. And unless we become the singular personality, then all bets are off.
Kyle: So I'm curious, Scott, I think we talked about this a while ago, so I think I know the answer, but streaming is obviously huge and it's only getting bigger. Are you guys planning on just staying there or do you have aspirations to branch out to network commercials or YouTube video production or whatever, maybe anything like that?
Scott: Yeah, we do a lot of stuff for YouTube. Everything we do now, we also do social versions of it. So in any package that we deliver, we deliver either a nine by 16 or a one by one or whatever format the client requests for their social channels. So we do support all that. You know we can film, you know, YouTube is just another streaming platform really. So we don't care, you know, we're commercial, but you know, I've done network commercials in the past and, you know, I've done local commercials.
And originally when I was interning out of college, I did supermarket commercials and that's really where I learned about, you know, high volume personalization basically that's that was my first exposure to that so we don't, we don't care.
You know I would say there's kind of a cap on how much we want budgets we want we're set up to manage but then we would bring in partners so we have relationships with other big production companies where we can we act as the agency with and then they produce a commercial just because they're more skilled at it you know and I've done big shows. I've worked on big live shows. I've worked on features, you know, all that kind of stuff. And it's just, you know, I don't want that stress anymore.
Kyle: Yeah, I feel that. Yeah. Cool, man.
The storm is coming down. So cool.
I wanted to kind of get into, you know, some of the speculative and prediction part of the conversation here. So I'm curious where you think the future of video personalization is going to go. You can, again, talk in, in general, you can talk about streaming and CTV. You know, if you think there's something else kind of headed, you know, down the horizon, what do you think is going to happen?
Scott: Well, what's interesting is a lot of these platforms now are getting into a position where they can buy broadcast and streaming at the same time and model in the same way and look at, look at, you know what's crazy is I don't know I mean you must you probably know this that now they have ways of judging metrics within store you know of commercial what kind of in-store traffic they get there's all these firms taking you know jumping on the the attribution chain because that's what everybody wants to say sometimes I think we over we overlook an attribution you know and sometimes it's just real it's really just it's really just in a way it's awareness which which builds attribution um but like you know there you know there are firms that specialize in foot traffic and there's and I just think that the future is more specific attribution.
You know a media mix with broadcast and and streaming and then be able to track all that I think from a production standpoint there's gonna be more and more tools out there that you're gonna be able to do lower cost production, more automated but you're still gonna have to have some kind of creative differentiation.
In order for it to stand, it's because they're all going to look alike. You're starting to see stuff be similar already. Talk to any professor who's working at a university getting papers in from kids, and they can tell that the papers were all generated by AI because they're all the same.
Kyle: Yeah, exactly. And like, it's all like the same punctuation and things like that. Yeah.
Scott: But you know, the good thing is ChatGPT just released the learning version. So the kids, so when you're writing a paper, you can learn while you're writing it. I can tell you how many, I don't know how many students are, you know, 20 somethings are going to be using that, you know?
Kyle: Oh, come on now.
Scott: Cloud AI also came out with something like that about four months ago, and now ChatGPT just released it.
Kyle: I guess that makes sense. I'm so torn with that all because I think in the simplest version of it, we go back to the days of even just having a calculator, and we had to learn all these complicated timetables and all that kind of stuff. And it's like, why do we have to do that? We have a calculator.
And so it's like, well, you know, why do I need to be able to write a 20 page essay? I have chat GPT.
Scott: You know, sometimes you need some of that knowledge to understand the logic behind that stuff. You know. I would say, you know, we, you know, I learned about abacuses, but we never used an abacus, but they just learned about it. How, how the abacus led to the how the abacus led to the calculator basically in the and how all that and you when you see the kind of how that works and the foundation of the technology you can use the technology better and if you have an understanding about.
I mean what sucks is is that people that that whoa moment is going to go away where you like kind of discover something and that clicks in your head as humans I think that that makes our brains grow and I think that you know I think that that's missing from a lot of I get laughed at by the younger people who work for us because I'll say, well, this is why that happens.
Do you know why that happens? Do you know why when you're editing audio, they say, put a hair or put some air in there and put a hair? And they're like, no idea. And the reason is because when you used to hand cut audio tape, you used to raise it and put it on a block, put the tape on a block, cut it where you wanted the edit to go, the amount of distance between the two pieces of tape was a hair or an ear or air some air or a couple beats it was a physical visual tactile experience and that's where those those things come from and when you know that when you understand that you can you can make your products better
Kyle: I mean, that's a great point. That's a great point. I'm now picturing you sitting around using an abacus, but I think that's a great point.
Scott: If I wasn't in my virtual office, which is, by the way, this is actually what the office looks like. I just couldn't get there today because I'm waiting for a delivery. But on the shelf back there, I still have three quarter inch tapes and I have a quarter inch audio tape
And I had that stuff to just kind of remind me of where it came. I have a dial phone from a prop from a show I did once. And I used to have it on my desk in my office when I was at one of the companies I worked for, not Beachbody, but the one after that. And the kids would come up and go, well, how did you text on this?
My favorite one was, well, wait a second. If you wanted to dial 1-800, you had to wait for the dial to go all the way around? Yeah. They said, well, that took so much time. I said, yeah. I said, but that gave you a chance to think before you yelled at the guy on the other phone because he hit speed dial or you told this guy you're an asshole on text. It gave you a chance to think and digest the thought.
And I think that's I just think that's there's a certain importance to that that we don't have anymore because we just rely on the technology. I'm just as guilty for that.
Kyle: I mean, it's too easy to do. You know, you just get into it becomes so integrated into our lives. So it's too easy to do. I totally get it.
Do you think that campaigns by default will soon have like 50 to 100 variations or even just like, you think so?
Scott: I think based on the targeting, yeah. I think that maybe not this year, but in the next couple of years, you'll see the power of, you know, and it's funny when I speak with the TV scientific team, I said micro audience and they go, what is it? What do you mean by that?
Define that, you know, and it's like, I think you can look at your website data. If you're a Ford dealer, a local Ford dealer, right? You can look at the IP traffic from your website for the week and see this many people looked at Explorers, this many people looked at F-150s, and this many people looked at electric Mustangs, right?
And then you can literally on Sunday say, okay, we had a thousand people at this, a thousand people at this, a thousand people at this, and we could build out a commercial with the Mustang inventory you have on the lot, you just send us the still pictures and we'll just build a VO behind it. And you could target those people who are interested in the electric Mustangs or the F-150s or the Explorers. I think there's so much competition for advertising. I think that'll give people a better chance at selling to their customer.
Kyle: It's the age old thing, you know, concept when it comes to marketing, be there with the information they need when they need it. So that's a perfect example right there.
Scott: Yeah. So it'll be interesting to see this product's going to have to evolve as the targeting abilities and the streaming targeting involves. But yeah we're willing to do that and invest in that evolution because I think it's a good, strong product and I think it'll help a lot of businesses.
Kyle: Cool. Very cool. So I'm just curious if there's any other trends that you're kind of monitoring closely that you think might have an impact on you guys in the next year, year to two years?
Scott: Well, obviously the AI footage, the video generation. And we've tested out a few things. Right now, the most useful tool that we have is a client can send us a vertical video from a phone, right? And then as long as like if someone pans the scene and they send us a vertical video, the AI software will build out the wings and put it into a 16 by 9 frame.
Kyle: Oh, interesting.
Scott: Pretty effective. It's pretty good. Yeah. You have to watch it, but you need three seconds. You need three seconds. You can get what you need out of it. So that's kind of a thing we're using right now. And it's going to come down to people knowing how to write prompts in order to get that image the way they want it.
Kyle: Is that what it is? You upload the video and you kind of engineer a prompt that says, "I want you to fill in the background."
Scott: Yeah, you tell it to fill in the left and right side and base it on the pan. If you pan, it does a really good job. If it's just a static shot, it does, it depends on the scene, how well it does. But it guesses, obviously. from pattern recognition, what is what goes there. But that's when you need a three second shot and you're stuck, it works really well.
Kyle: Very interesting. Very interesting.
Scott: But I think, I'm sure in two years, I’ll have someone say, “OK, I need these scenes,” and instead of going to a stock footage library like we do now, they’ll just create it.
I’m still concerned about copyright. You know, the early internet days of copyright battles - I’ve been sued a couple of times by the NFL for stuff we thought we had the rights to. So I’m a little copyright hesitant, shall we say.
That’s what concerns me. All this intellectual property is being scanned and used. You know, like for the Screen Actors Guild - part of the strike was about AI-generated characters, synthetic people basically. And they came up with - I don’t know the details of the agreement - but I know they decided that if you combine Person A and Person B, both of the original people get a residual on that synthetic person.
But how many generations down the line before you lose your rights to that thing that’s a part of you, you know? In the future, how many Tom Cruise noses are we going to see on people and go, “Hey, that’s Tom Cruise’s nose,” you know?
Kyle: That's interesting I i knew about the strike I didn't go that deep into it that's a very interesting frontier I guess you can call it to be you're kind of marching down to figure out how to navigate that
Scott: Yeah and the studios are dealing you know I mean think about you know cost savings and the thinning of the dollar. There's still the same amount of dollars out there basically, but it just goes to a lot more, a lot more different sources of entertainment.
It's crazy in Los Angeles. Cause you know, I've known people who've been in the business for years who no longer have jobs because of the evolution of technology and you know, just streaming versus broadcast versus, all that kind of stuff. My saving grace is in what's kept my career going is always a thirst for technology and how to integrate technology what i'm doing so I was lucky. My parents gave me that so you know. My dad was the first one on the block to have a microwave oven and the people used to tell us we'd be we're all going to be dying, We're all going to be dead by the end of the month you know
Kyle: Oh my god amazing Awesome. Awesome. I got a couple of rapid fire questions here for you, Scott. So one, if you could snap your fingers right now and have one workflow or one AI tool or one automation, just boom, implemented into your day, what would that be?
Scott: I don't know, really. There's so many potential answers. I mean, some kind of project management tool, I guess, that would help us manage all the pieces with a lot of different editors all over the world and different participants. We deal with a lot of different verticals, so something that would make sure that everything was there so that we didn't lose all this time having to track stuff down. I think that would be the biggest help right now.
Kyle: What about the tools that you have in place right now? What's one tool that you can't live without?
Scott: ChatGPT. I have dyslexia, so my business partner calls it Scottish, and sometimes they'll say, let me translate Scottish. So I'll write what I want, and then I'll have ChatGBT help me to clarify it and make sure I'm clearer and more effective in my words. And then also Grammarly. I love that product.
Kyle: Okay, great. Yeah, yeah. I would love to see the custom GPT that's solving Scottish.
Scott: Even that shakes its head.
Kyle: Yeah, yeah. Do you have a creative project in your entire career that you're most proud of?
Scott: No, a lot of them. There's a lot of different things. It's funny… It's funny because when I get together with friends or someone will ask me something, I'll say, oh, yeah, yeah, I did work with that person. And I'll forget about it. You'll forget about them.
The coolest thing was probably working with Robin Williams. We did this thing. You know this is where you know ai is screwed you know if you if you could ai his brain I mean so like we did this thing for cbs about keeping kids in schools and so he was supposed to do something and he hated the idea right he said just find me a library so we went to the library at the high school where we were shooting this thing and he walked in and he said he walked through the stacks and then he pulled out a bunch of books just a smidge out of the stacks he said okay i'm ready to go and then he walked through the stack he walked through and talked about the importance of learning and reading and became the characters in all these books and then he said okay I want to do it again and then he went and adjusted the books again and then he went through and did it again just to see this thing come out of nothing
was the most one of the most brilliant things i've ever seen you know and you know i've been one of a whole ton of celebrities but that was probably the most amazing amazing thing I ever saw.
Kyle: He is one of the most talented people that for sure he has one of my favorite stand-ups of all time. One of his specials in the early 2000s. Does that live out on youtube anywhere?
Scott: I haven't been able to find it. I haven't been able to find it so
Kyle: okay okay yeah
Scott: You know, I've been very fortunate. I've done a lot of really interesting projects. I've shot all over the world, and have friendships with people all over the world. My very first real job was when Richard Belzer got knocked out live on TV by Hulk Hogan and dropped on the floor. Me and the guy next to me were doing cue cards, and we had to run out and hold him up so he didn’t collapse. Then I called 911 backstage.
So, the reason I mention that is because early in my career, I learned that anything can happen in live TV, and you have to be prepared for it. The producer at the time was a guy named John Gilroy, who was Dick Cavett's producer - really dating myself here. He had all this live experience, and we still had 20 minutes left in a live show with about 400 kids in the audience. He just got up, sat in the host chair, and finished the show.
And that was kind of a cool thing. But, you know - Yeah, I could go on.
Kyle: Awesome. That’s a good lesson to learn early in your career.
Scott: Yeah. Well, I've had late, late, late cool stuff too. My experiences with Beachbody were amazing. I've been really fortunate. I've done very interesting things.
Kyle: Very cool. Very cool. A couple more - just real quick, rapid fire. If there's one thing that you want marketers and creatives out there to remember, or always keep in the back of their mind about AI-powered video, what would it be?
Scott: Watch it carefully because you’ll never know what it’s screwed up on. It can make your product, your commercial, or your media a joke because of what it messed up. One of the things we run into is people give us social clips to use in commercials, and they go, “It’s great, it’s great!” They don’t realize that with streaming TV, you’re dealing with a 65-inch screen where every detail can be seen.
There are some things you don’t want to see in your commercial. We had a food commercial where there was grease and crumbs all over the table. We had another one for women’s slip shorts where it left nothing to the imagination. They worked great on social, but not on a big screen. Watch your stuff and realize that you’re in multi-format displays - you’ve got to consider all of them.I mean, I've seen broadcast commercials where I thought, “How did that get in there?”
Kyle: That’s a great point. Bigger screen means bigger visuals.
Scott: Yeah.
Kyle: Cool. And what does your late-night edit process look like? Are you pounding coffee and tequila, or what’s it look like?
Scott: I’m calling a guy in Spain if it’s late night and we have to work with him. On certain projects, I still edit myself. I’ll do the rough because I like telling the story and finding what works. Sometimes I’ll have a scotch, and I usually have my dog sitting next to my feet, reminding me he’s down there and that I should be paying attention.
Then I’ll always look at it the next day to see if it’s as good as I thought it was before sharing it with the client. Never share late-night work with a client.
Kyle: Yeah, absolutely. Amazing.
This is great, Scott. I wanted to give you a second here for open forum. Is there anything that you're passionate about or excited about, that you just wanted to share?
Scott: No, not really. My thing would be with younger people coming into the business - foster their talent. Help them not rely on technology. Help them use technology to enhance their talent and take that time to grow. That’s kind of the most fun part for me now: helping bring these kids alive.
Even on the client side - I deal with a lot of 20- and 30-something clients. Don’t be afraid to explain to them why something is done. Help them understand so they can do better in the future. That makes a better industry.
Kyle: That’s a beautiful sentiment. Love that. Well, awesome, Scott. Just for anybody watching or listening, you guys can again check out Carpe at carpecanum.com. Scott, you’re on LinkedIn. Do you have any other handles you want to plug or share?
Scott: You know, I got off social media because it wasn’t good for my mental health. About six months ago, I got off everything, and I’m enjoying that. Cooper J. Dog is my dog, and he does have an account, but I haven’t posted recently. I took away his iPad because he was on it too many hours on social media himself.
Kyle: Put that timer on for him.
Scott: Yeah, I tried to. But like most kids, they figure out how to get around the safeties.
No, not really - just go to carpecanum.com. C-A-R-P-E-C-A-N-U-M dot com. “Seize the dog” is roughly what it means. That comes from the idea of making the best out of anything. A “dog” is like a bad script.
I created the company at a bad time - I had just lost my job and was in a rough spot. I said, “I’m going to make the best of this situation and grow from it.” So, Carpe Canum - seize the dog, make the best of anything. We’ll work with just about anybody to get them good content, and hopefully have fun doing it.
Kyle: Fun - I mean, why do it if it’s not fun, right?
Scott: Well, sometimes for the money.
Kyle: Yeah, of course. I mean, Cooper needs his Greenies, you know?
Scott: Yeah, exactly.
Kyle: Awesome. Scott, thank you again for joining me on The Brainiac Blueprint. I really appreciate it.
Scott: Thanks for having me. I appreciate it. I always like hanging out with you - whether it’s digital or hopefully soon in person. We’ve never met in two and a half years.
Kyle: We haven’t. I’m really hoping we can make that dinner date happen whenever you’re ready.
Scott: Or get your passport and come out to California.
Kyle: I’ll be out there at some point soon. I was just having some conversations about my next trip, so hopefully sometime soon for sure. All right, well, thank you again.
Kyle: Do me a favor - look in the camera and just say, “Stay brilliant, Brainiacs.”
Scott: Stay brilliant, Brainiacs.
Kyle: Awesome. Thanks, Scott.
Scott: All right. See you soon.
Comments