Pete Bishop: Hey, Mark, how's it going? Yeah.
Marc Binkley: ⁓ good. Thanks. I'm excited. Thanks for the invite to do this. I'm really excited to chat with about this.
Pete Bishop: Yeah, me too. I feel like it's long overdue and it's nice to, know, I find these episodes kind of a nice way to catch up in general. So even though we're recording it, it is nice to talk to an old friend. And, know, you and I were chatting a little while ago and you brought up this ⁓ MIT study. I was thinking maybe we could start there. ⁓
Marc Binkley: Yeah. Yeah. So the, the big punchline on that one is that of all the companies they looked at that were implementing AI, they found that 95 % of them had zero return. It's crazy.
Pete Bishop: That's crazy. as in they building systems in AI or they're just ⁓ using in their software.
Marc Binkley: Yeah, they like company gets approval to roll out AI and to whatever the thing is that they're going to roll it out into. There's a corporate enterprise wide initiative to start a project. And so it's the projects themselves that they roll out. Very few of them actually show any meaningful return.
Pete Bishop: Right.
Marc Binkley: ⁓ it's just an interesting thing because it's, I think, indicative of where we are now with AI in terms of implementation, in terms of adoption. Like a lot of people are really excited about it, me included. And yet, when we're given the opportunity to do something with it, where most of us are feeling kind of underwhelmed by it. And in some cases, it's because it's not making our lives any easier in the way that people think they will.
Pete Bishop: There was a study we did with Stone-Olafson on this where I think, I can't remember the stats came back, but it was ⁓ similar, whereas there's lots and lots of people using AI, but a lot of people really seeing the benefit or understanding their kind of economics around it, ⁓ for lack a better word, right? And I think there's a lot of pressure.
Marc Binkley: Mm-hmm. Yeah. ⁓ Yeah. Yeah. I read that. It's a good study. I love that you guys did that. It's really good.
Pete Bishop: ⁓ yeah, thanks. take credit, but I had very to do with it. That was all Stone Olves. We're just beating our chests with it. But I did find that it seems universal. There's a of pressure on everybody to be adopting this stuff. And do figure that's coming from social mostly? That's where I'm getting it from.
Marc Binkley: Yeah, I think so. mean, it's kind of the same thing. it's a really common story in general. AI is the death of the old way of doing things. It's the most radical transformation since the last most radical transformation. You know, a whole digital transformation thing that happened a couple of years ago. And then the social revolution before that and the intention economy and there's just... I mean, if you go back even further, TV was going to kill print or radio. mean, there's every time there's a new technology. think there's that adoption curve where there's a lot of hype early on. Um, and it's usually created, I think a lot of times just because it's so new and so exciting and a lot of people generate a lot of buzz around it. mean, certainly the lot of the S and P, like I think I heard yesterday that it was something around 35 % of the the value of the S and P 500 comes from the top seven companies, which are all like Nvidia, Google, Metta, Microsoft, Amazon, giant companies doing a lot of stuff with AI and there's like, they're making a lot of money by selling each other a lot of stuff right now. And so it's, there's a lot of hype around it, but then it gets to you and I, and I mean, do we like you? know you did some really cool projects or implementations trying to make to get to zero inbox as an example. I know it was one of the things.
Pete Bishop: Yeah, I find here's my take on it lately is just Go going little wins rather than trying to overhaul your enterprise is just find those little pieces that have been an annoyance and trying to get them out of way because I you know, I worry that Because of the way things are changing so much it's hard to put your flag in the sand on any one piece of technology at the moment so if you're trying to
Marc Binkley: Totally.
Pete Bishop: let's say rip out your operating system and put in something like this. I'd scared to death to do that just because A, no really got the security figured out and B, there's something two weeks from now that might change everything. So it's just easier to ⁓ get quick And that's where found it been successful. ⁓
Marc Binkley: Totally. Totally. Yeah, it's complicated. I spent, had to have been, ⁓ think I started on April 1st and I spent a solid three weeks, just nights, like obsessing about building agents. After, and I've been using projects a lot in Claude and in ChatGPT and in GeminiGems. And so I was like, okay, I'm gonna go all in on this Claude thing. I leveled up and got the.
Pete Bishop: You
Marc Binkley: the big max subscription and figuring out what this Claude code thing. And then after three weeks, probably 50 or 60 hours worth of work, I asked Claude, I'm like, ⁓ what we doing? Like, it feels like this is not working. It's not doing as much. And then Claude's like, yeah, you're better off going back to what we'd had before. It doesn't make any sense. I'm like, great. So we just start from scratch again.
Pete Bishop: I like how it's so positive about bad news, hey? It's just like, you're right, you're right.
Marc Binkley: Yeah, yeah, you know, you're right. I should have caught that earlier.
Pete Bishop: Yeah, I I wonder because you see the kind of adoption curve that usually happens a big trend and then it levels out. It just feels like this one's taking a little longer because it's such a big thing, I think. But.
Marc Binkley: Mm-hmm. Yeah. Yeah. There's, there's a lot of, a lot of free usage on all the platforms that I've seen. there was a report from, the Ehrenberg Bass Institute on ⁓ they're looking at jeopardy marketing law, not the movie one, on whether, whether like the loyalty is falls ⁓ in the same patterns they've seen another. industries and like from aviation fuel to toothpaste in China. And it does. What was shocking to me about that one is that Claude is so low on the adoption curve. But everything I look at is 100 % Claude. And it's, I'm in my own bubble. Like my, my feeds are all adapting to Claude because I'm using Claude so much. So I think that most people use Claude, but that's just cause it's me seeing that I'm using Claude and everyone I'm talking to is using Claude.
Pete Bishop: Really? Right.
Marc Binkley: And all the people I'm interested in talking to right now are using cloud cowork and they have ideas on it. And so like I just, it's feeding all the algorithms are feeding that back to me. So the reality is most people are just using free tools. Most people are using chat, GBT or Gemini. and not really in space of, of building out agents and all that kind of stuff.
Pete Bishop: Huh, interesting. Hmm. You and I had also talked a little bit about this of sea of sameness. Do you mind just describing that a little bit? ⁓
Marc Binkley: Mm-hmm. Yeah, totally. Like I've been really fascinated in AI and trying to figure out where it helps and why it helps and like to get around this, you know, why 95 % of implementations get zero return. So I've interviewed a lot of people around this, yourself included. And I've interviewed some people from like the big holding company agencies and Roger Martin, I talked to him a couple of guys or Peter Weinberg from Evidenza. there's some, I've had a pretty interesting perspective, like because I'm trying to write these articles for work on AI and what it's doing and the best ways to use it and stuff like that. But the sea of sameness thing kind of came from an aha from Roger Martin. And we all seen this, but like the way he described is that it's optimizing towards the mode. So it's not, it's, this kind of center point of, of the herd of answers that are out there. so tools like and like Claude 4.7 got the highest ever score up until two weeks ago. think it was when chat GPT released their new release and it's the score is on the GDP Val score.
Pete Bishop: Yeah.
Marc Binkley: And what the GDP Val score does is it takes, it scores the ability of these tools to do knowledge work. And they look at, I believe it's 77 different individual professional services, jobs. And they go, how well does it do like, kind of work or how well does it do marketing kind of work? How well does it do copywriting kind of work or legal work or like across the spectrum of 77 different industries. And so it's scores the equivalent of what a 14 year veteran would get in any one of those categories. Which is kind of crazy when you think about it has all of the knowledge, but it to the mode, the middle, because it's trying to be appealing and it's trying to get us to use it. So that's why it says things like, good catch. And I actually did this on three of the different platforms at one point. And I was like, why do you keep saying things like good catch or, know, I should have thought about it that way. You're absolutely right. And then the reason, and if you do this enough times and ask whatever tool you're using.
Pete Bishop: Yeah. Right. Hahaha
Marc Binkley: And eventually will come around to saying, ⁓ it's because I'm optimized to being not correct. And so it picks the mode because it's the closest thing that it thinks it can get to that is an agreeable answer. And exactly why it says things like, if you say to it, any of the LLM tools, guess the number between one and 10. The first number will come up with seven. The second number, I think I'm getting this week and seeing right, but it'll be nine.
Pete Bishop: You
Marc Binkley: and the third number will be three and then it'll be one. And the reason for that is because that's the mode of what most people pick they're, it's not actually a random number, but it's at a very agreeable number. And it sounds like something that's random, but it's not.
Pete Bishop: Ha! Yeah, interesting. You always have to think about at the end of the day, it's a prediction machine, right? And it's trained on ⁓ pile of data that, and if you know where the sources are, right? You get a sense of, I there was a fellow I know that every, everything he was generating at the end of it was kept saying, thanks for tuning in. And he's like, I didn't ask you to do that. Why does that keep coming up? And it's like, I think this was Gemini was it to him. And he's like, He figured it out is because it was trained on YouTube and a lot of the transcripts would end off with totally. So it's just a good reminder. And the other thing I follow this one guy, he's really, he's really funny. He's super deadpan, but he will, he'll just try and trick AI into doing stuff or lying or, or contradicting itself.
Marc Binkley: ⁓ smash that like button.
Pete Bishop: And he made this song where all he did was just make a bunch of fart noises. And he's like, what do think of my song? And it's like, yeah, you've really found like a new genre. You've really uncovered something that's, you know, and niche. And, and every time that I'm thinking of doing strategy work, I just, have that in the back of my head. I'm like, it did that to the fart song. So ⁓ it's like saying I'm super smart right now, I really can't trust ⁓ to degree. ⁓ Yeah.
Marc Binkley: You Mm-hmm. Mm-hmm. Yeah. Yeah, it's true. And I think that's the thing where you get it. The machines themselves are confirming biases first, because that's sounds agreeable. And we like things that agree with us. So it's not like if if you think about their objective, which is to get people to use the tools, because that's how they eventually make money, then they're not going to optimize for providing conflicting evidence or data about who you are. That was actually one of the responses I got if It was something like, I think it was from chat. It was something like if there were two people in a research group and one was, you know, questioning you on the methodology and the results, I'd be the one that would say, yeah, that's great. Keep going. I'd be the one cheering you on. And so it's, just the thing. It's, it's pushing us to, to believe more of the things we, we that sound right to us.
Pete Bishop: Right, which is part of that sameness, right? When you are eventually, if everyone's kind of producing stuff that's agreeable and in the middle, like you would say, then that lead into a little bit more of the value of creativity moving forward, because I think that's it's getting confused now with what creativity when you're using these tools. Do ⁓ you that?
Marc Binkley: Right. ⁓ Totally. ⁓ We had a panel last week for the CMA and like it was, was around stop talking about AI, start making better decisions. And, set up the talk, had, I had created some images, starting with an image of myself, like a bio headshot kind of thing. And that was the first image. And then I just put it into nano banana and said, make me an orangutan and it made turned me in and rang a tang. And then there was another one that was me surfing and then looking Dr. Strange ⁓ when was doing that 14 million choices thing, looking for future and looking for the answers. So there's all kinds of things that you can do with it, but making a piece creative is not necessarily the same thing as having a good piece of creative. And I think there's a whole bunch of things around the strategy behind that, the thinking that needs to be more than ever before, because anybody can make something, but it doesn't mean that you're making something that's good. So I think that's part of our role is like to avoid the CSA and this is to be the tastemakers.
Pete Bishop: Yeah. Yeah, I found that too. This willingness to run with any idea it's got, right? So it'll come up with, here's three campaigns. one do you want to run with? And if you said any of them, it would just start. It would just ⁓ out your ads and your PowerPoints and everything, because it has no sense of this is good or this is bad. And that's the taste is what you're mentioning, ⁓
Marc Binkley: Mm hmm. Yeah. Totally. Yeah, we had ⁓ last year we did a workshop with Carl was Carl from zero to 60 and it was a workshop on we had ish people in the room and so we had we divide the room into three groups. So one was the vibe marketers. So they just we gave everybody same brief and the vibe marketers just ran with it and they just did the things that AI was telling them to do. And at the end of the three hour workshop, everybody's going to produce, creative, you know, ⁓ to serve the create the campaign brief that we, that we had laid out to them. So the vibe markers. They just ran with it. And like a hours later, they got this ton of cool stuff. none of it on brand that they admitted it produced charts and all kinds of stuff. They're like, no idea where this information came from. And then the other one was there's another group that was the traditional marketers. And so they were feeding it with prompts, things like, using Maslow's hierarchy of needs and the funnel and reference Simon Sinek start with why and come up with creative to suit the campaign that we have been to produce a result on. And so theirs was interesting too, because it looked really good at the end of it. They, had a jingle and dance. was for. a fictitious pro-g company. They had dancing pro-g's. It was crazy what they could make. And then there's a third group that was an evidence-based group, and they were there given specific prompts around like category entry points, distinctive brand assets, making sure that they were aligning the distinctive brand assets up with the category entry points that initiated a purchase cycles and triggers. And so there's a whole bunch of stuff around that that were using, and they also produce some great stuff. But at the end of it, When you, if you were blind to, not blind, but didn't know the objective and just were looking at the creative, it would, at the end we were like, that's really impressive that everybody made all these things, um, in a very short window of time, but nobody had to defend their decisions to an executive room or to a client. And so I think that's where a lot of us would fall down because the vibe markers admittedly said they couldn't justify any of the things that were there. They're just, they've made stuff. Right. And then the traditionalists, I think you run into the same situation that markers have been facing all the same for decades, which is they have a hard time justifying the budget and hard to find defending brand, and a really hard time kind of proving any type of ROI on those kinds of investments. Whereas think the evidence-based people would have had a better chance of getting something pushed through in my opinion.
Pete Bishop: Mm-hmm. Mm-hmm.
Marc Binkley: And so, but each one of them went in with somewhat of a bias and machines confirmed their bias.
Pete Bishop: Yeah, you're kind of leading the question, right? think our director, Kurt, I think he talked about this a little while ago, which was a good little phrase. was saying he feels his job is to help people make decisions. And in some ways, I think that's true. know, it's companies are messy by nature, which I think is part of the reason AI really has a struggle when it's, you know, it's easy to ⁓
Marc Binkley: Mm-hmm. Totally.
Pete Bishop: set up a company when you're a person of one and you're using Google as your backbone. But if you are, know, a 10 year old company trying to rebrand, and then you need to, like you said, talk in front of a board and talk about, you know, the of the brand and how it's going to last for the next 10 years and all that kind of stuff. And if you just generated everything out of chat, GPT, it would be really hard, I think, to, convince people to that change, or at least, ⁓
Marc Binkley: Mm-hmm. Yeah.
Pete Bishop: stand behind the work and where I think this stuff falls down a little bit. ⁓
Marc Binkley: Yeah. And so like in the vibe marketers case or, you know, there's lots of other examples of this, but there's a great line that should deliver it at the speed of light is still shit. And it's not because we have these tools, it doesn't mean it's helping us solve any problems. It, we could just be doing more of the faster.
Pete Bishop: Hahaha you Yeah. Yeah, I like that. You know, here's the other thought I've been having. I don't know if you've heard this story. It's an old story, but and I wish I knew where it came from and it could be complete fiction, but it was this thought that this group, whatever study group brought in these creative directors from three different areas and they They brought them into this room and gave them a brief. And on the way to this room, they drove them intentionally past a bunch of things like ⁓ the zoo and some billboards and stuff. And then when they got them all together, they gave them the brief and then they all went into separate rooms and to crack it. And they all came back and roughly presented the same creative that seemed very out and whatnot. But it was all based on these like subconscious cues that they intentionally planted.
Marc Binkley: Mm, yeah. Mm-hmm.
Pete Bishop: me wonder. you know, this is getting a little nebulous, but as you getting into creativity is like, are all product of our own inputs, right? The books we've read, the movies we've seen, the we've had in our lives, you are we limited by that? And is this opening that up or are, know, you're talking to computer that has all the books and all the movies and all the campaigns available to it. you know, maybe it down to that ⁓ or whatnot, but
Marc Binkley: Totally. Totally. Yeah.
Pete Bishop: does make me worry a little bit on the creativity side. You know what I mean?
Marc Binkley: Yeah, it does. But I kind of look at it as. Like limitless possibilities. So going back to the GDP VAL score, right? There's 77 different profiles of professionals that they score against. I think I'm pretty sure, I want to say it's 77, maybe it's 44, but anyway, it's a lot of people that they score the knowledge, the ability to deliver knowledge work against. And it's, by the way, it's as good at doing something in seconds that it would take someone seven hours to do. And that's the score. So that was the reason why I had that picture of Dr. Strange up of my head on Dr. Strange. Cause it's that idea of like, there's infinite possibilities. You can run a Monte Carlo simulation in any of these tools to come up with the answer, the one choice, right? helps beat Thanos or whatever. Like you don't have to take the mode answer because there's all this knowledge out there. And so, ⁓ One of the things that I've done, I did early on when I was playing around with, all the different tools is I built in the projects. built, council, a conflict, you can it a conflict council if you want, but a group of people like based on, created personas based off of all kinds of people. Like I put Roger Martin in there. think a lot about him. I put in dead people Steve jobs. I put in Sun Tzu, put in Genghis Khan. This is first my strategy strategic advisory council. but when I do strategy, like I have ideas that I'm talking to Claude about and then my strategic advisory council chimes in on that. so trying to create differences of opinions to make me think differently. their personas are kind of accompanying me in the decision making process. So I'm not just taking Claude's word for it. I'm using some of their input to help me re shift and reevaluate shine lights on some of the dark spots that I haven't thought about.
Pete Bishop: I love that. just so everyone understands, how are you putting a personality in as a counsel into AI?
Marc Binkley: So how I did it originally is a little bit different from how I'm doing it now. And this is my experience with wasting so much time trying to build agents, which isn't really a waste. It's just like, it's just learning. But initially what I did was I, cause in most of these tools you can load up supporting documents. So I had cloddered
Pete Bishop: You Yeah, I get it. Mm-hmm.
Marc Binkley: ChatGPT and Gemini, all kind of the things doing the same type of thing where I would ask them to create basically like a canon of the most famous works from these people in a document. And then I loaded that document into the documents for that project. And then in the instructions, then I would say whenever I'm asking questions about this or that or whatever, however I wanted the pro like I the system or the process that I want to take. I want the, how I want the project interact with me. Then I would ⁓ make that it would reference my strategic council and give me at least three or four quotes or perspectives from their work and my thinking to make me better and make the outcome better.
Pete Bishop: Hmm. Yeah, that's awesome. I've heard of people doing that with boards, right? They, yeah, they take everyone's personality and basically bake it into AI, which I'd also be like, Wow, I wonder what it'd be like if anyone ⁓ read own, you know, personality as written by somebody else. But it's a great way to like, Hey, how is this presentation going to go over? Right? And how do I modify it so that, you know,
Marc Binkley: Yeah, same idea. Yeah.
Pete Bishop: Dan over there doesn't pick it apart as usual, right, or whatever.
Marc Binkley: Right, yeah. Fuck, Dan is always that guy. You
Pete Bishop: Yeah, that's super I like has it when you've been doing that? Has it gotten you to better places?
Marc Binkley: ⁓ yeah, for sure. Yeah. Cause it, I mean, it gets to the point where I'm like, try and pick the divergent thinkers. So they're not, they're all the same thinkers. but yeah, like more and more, I kind reconsider who I would want to have in the board. And I kind of reshuffled around based on some of the responses I'm getting. but like, Einstein is in one of mine. I've got, ⁓ you know, in a creative uh, project that I've got, George Lois is in there. Um, you know, I don't know, just people that I think, uh, Picasso was in one of the, the creative one as well, like just different people with different perspectives that I'm like dead or alive. It doesn't matter. Cause a lot of these people have published ideas and literature and stuff that it knows about. And I'm just formalizing some of the things that I want it to reference when it, when it challenges me. this that I the most recent article was writing, there was thing I found. There a survey that was done asking 1500 executives what their biggest challenges were. So what came back was bouncing short and long-term collaborating cross silos, protecting margins in commoditizing industries. These sound familiar, probably, right? The crazy part is that that survey was done in 2008.
Pete Bishop: Yeah, yep.
Marc Binkley: So the big problems that we've faced our business are still the big problems that have been around for decades. And I think when you look at the usage of things like chat, GBT, there's a study I found that showed that the number one usage at work for chat GBT is copywriting. So I think some of the things that we've been talking about, like challenge groups or advisory boards or whatever you want to call it. And then things like you have access to.
Pete Bishop: Right.
Marc Binkley: the world's information pretty much you can run Monte Carlo simulations like we don't have to trade off solving big challenges wicked problems with delivering shit today like it can be both and I think that's kind of an interesting thing
Pete Bishop: You Yeah, I think that makes sense. you find like I've heard this many times is how content is the number one. Do you figure is or writing anyways, creative writing. It just felt like that was the very first example of AI when they first, I remember seeing, I don't know how old this was. It's got to be over 10 years of seeing an example of AI writing a blog post and they'd show it side by side and everyone, everyone's jaws dropped. It just feels like the very first initial application of a text-based AI. And then all this other stuff's relatively new. Like you think of over the last year, agents just ⁓ I think within the last
Marc Binkley: Mm-hmm. Totally. Yeah.
Pete Bishop: year at least in the public realm as far as I remember right so all of these updates and stuff which is going at an exponential rate like the ability for us to adopt seems so hard I read I do you read Sapiens ⁓ I can't remember who it good book but it was ⁓ his point was is that
Marc Binkley: Yeah
Pete Bishop: from just an evolution point of view, we are still adopting to agriculture. That's how slowly we're getting into things versus tech is so far beyond us. And the rate that this is coming out right now where they used to take us like TVs came out. It took us like 40 years to really get TVs going in our Now we have something coming out almost every day. I just don't believe we can.
Marc Binkley: Mm-hmm. Totally. Yeah.
Pete Bishop: To your point around, this isn't an effective tool for a lot of people. I just don't think we know how to get into this stuff that quickly. It's just so fast.
Marc Binkley: Yeah, I agree. And I think if you look at the digital landscape, I would say 2012, ⁓ just mark a time, a moment in time, that's when Facebook went public. That's when I really got heavily involved in digital media, social media, probably around that time, a little bit before that, or actually, but anyway, that's just a good marker. And so ⁓ here are, help me with the math, but I think 15 ish plus years later. And it's now taken over digital has now taken over traditional in terms of total ad spend. Like if you look at, e-marketer charts and that kind of stuff, right? Like the, the dollar spent like for every dollar, more than half is now going to digital platforms, but that took 15 years ⁓ least.
Pete Bishop: Right. Mm-hmm. Right, I remember that. Yeah, I remember it creeping up and they were like, it's never going to be more. Right? Yeah.
Marc Binkley: Yeah. So, you know, for all the hype though around how everyone's doing social, you got to be on social and like I was at conferences where that was a thing and, true. And that's, and it's not to say that it's not important to understand. I think AI going to be amazing for a lot of companies to do this, to do dangerous, but also amazing, like same kind of thing, double-edged sword, but, but it's still going to take people are people. It's still going to take a long time to get. used to it and figure out how to use it. And especially organizations, think agents are so hard, my experience with, don't I'm that unique. Maybe I, I'd say a sock, that's, but maybe everyone figured out on the first time, but it's really hard to explain what we do. Like, I think we take that for granted and it's like, we wanted to do all these things so quickly, but you realize quickly that there's a lot more nuance and a lot more.
Pete Bishop: ⁓ yeah.
Marc Binkley: comes into what we do. And then you have to think about what do in a way that is not ⁓ natural, we're just so used to doing it.
Pete Bishop: Yeah, that's been my experience. This agent building is relatively simple, but it's coming up with all the business rules and the the strategy around what you're trying to build and how it's going to work with your own messy background that makes it tough, right? So, yeah, I get it. And think that's the disconnect right now that's going on. It's just everything's simple and should be automated and you're falling behind and you're missing out. And then there's reality where everyone's like, yeah, this doesn't for me. My company doesn't work like that. We've got a million other variables here, but we're all buying subscriptions and everyone's working on it.
Marc Binkley: Yeah. Totally.
Pete Bishop: Yeah, it's a bit of chaos.
Marc Binkley: Yeah. And I think where it gets even more complicated is if you start thinking about the organizational way you do things, like kind of thing, that becomes even more difficult to codify. Cause that can, that ⁓ change one department to the next or one manager to the next. And so if you're trying to produce something that's ⁓ consistently good,
Pete Bishop: Yeah.
Marc Binkley: you could do that at an individual level, maybe, but then at a departmental level or a corporate level, it becomes harder. And so I think there's a whole bunch of thinking around decoding how. Businesses function that. Well, we'll face a lot of opposition because then you're ⁓ people to give up their secrets. How do you do stuff? And then you'll end up finding that a lot of people won't want to share that because that's the special sauce that.
Pete Bishop: right.
Marc Binkley: is what makes them expendable.
Pete Bishop: Right. Yeah. How do you, can you break down what you do exactly and put it into a document so that we can get rid of you? Wait a minute. Yeah, I hear you. Okay. Well, Mark, I appreciate you coming on. I feel like we could probably chat for a couple hours at least. In fact, the next one I booked you in for about nine hours. But
Marc Binkley: Yeah, right. I've seen this movie before. Wait a sec- Yeah. No!
Pete Bishop: This one, unfortunately, we've got to wrap up. So thank you. And I just wanted you also have your own podcast, Sleeping Barber, right? So I'll stick that in the show notes as a link.
Marc Binkley: Yep. Yep. that'd be great. Yeah, ⁓ going to be going back to Cannes this year, actually, as media to cover the festival. So be a lot of fun. We had a lot of great interviews lined up and we just have actually published a lot of great interviews this year. yeah, it's a lot of good things happening with that one too.
Pete Bishop: ⁓ cool. Okay. ⁓ Well, thanks again, Mark. we'll have to get you back. Thanks again.
Marc Binkley: Thanks, man. Yeah, that was great. That was lot of fun.
Pete Bishop: Okay, how does that feel? Seem okay?
Marc Binkley: ⁓ good. Thanks. I'm excited. Thanks for the invite to do this. I'm really excited to chat with about this.
Pete Bishop: Yeah, me too. I feel like it's long overdue and it's nice to, know, I find these episodes kind of a nice way to catch up in general. So even though we're recording it, it is nice to talk to an old friend. And, know, you and I were chatting a little while ago and you brought up this ⁓ MIT study. I was thinking maybe we could start there. ⁓
Marc Binkley: Yeah. Yeah. So the, the big punchline on that one is that of all the companies they looked at that were implementing AI, they found that 95 % of them had zero return. It's crazy.
Pete Bishop: That's crazy. as in they building systems in AI or they're just ⁓ using in their software.
Marc Binkley: Yeah, they like company gets approval to roll out AI and to whatever the thing is that they're going to roll it out into. There's a corporate enterprise wide initiative to start a project. And so it's the projects themselves that they roll out. Very few of them actually show any meaningful return.
Pete Bishop: Right.
Marc Binkley: ⁓ it's just an interesting thing because it's, I think, indicative of where we are now with AI in terms of implementation, in terms of adoption. Like a lot of people are really excited about it, me included. And yet, when we're given the opportunity to do something with it, where most of us are feeling kind of underwhelmed by it. And in some cases, it's because it's not making our lives any easier in the way that people think they will.
Pete Bishop: There was a study we did with Stone-Olafson on this where I think, I can't remember the stats came back, but it was ⁓ similar, whereas there's lots and lots of people using AI, but a lot of people really seeing the benefit or understanding their kind of economics around it, ⁓ for lack a better word, right? And I think there's a lot of pressure.
Marc Binkley: Mm-hmm. Yeah. ⁓ Yeah. Yeah. I read that. It's a good study. I love that you guys did that. It's really good.
Pete Bishop: ⁓ yeah, thanks. take credit, but I had very to do with it. That was all Stone Olves. We're just beating our chests with it. But I did find that it seems universal. There's a of pressure on everybody to be adopting this stuff. And do figure that's coming from social mostly? That's where I'm getting it from.
Marc Binkley: Yeah, I think so. mean, it's kind of the same thing. it's a really common story in general. AI is the death of the old way of doing things. It's the most radical transformation since the last most radical transformation. You know, a whole digital transformation thing that happened a couple of years ago. And then the social revolution before that and the intention economy and there's just... I mean, if you go back even further, TV was going to kill print or radio. mean, there's every time there's a new technology. think there's that adoption curve where there's a lot of hype early on. Um, and it's usually created, I think a lot of times just because it's so new and so exciting and a lot of people generate a lot of buzz around it. mean, certainly the lot of the S and P, like I think I heard yesterday that it was something around 35 % of the the value of the S and P 500 comes from the top seven companies, which are all like Nvidia, Google, Metta, Microsoft, Amazon, giant companies doing a lot of stuff with AI and there's like, they're making a lot of money by selling each other a lot of stuff right now. And so it's, there's a lot of hype around it, but then it gets to you and I, and I mean, do we like you? know you did some really cool projects or implementations trying to make to get to zero inbox as an example. I know it was one of the things.
Pete Bishop: Yeah, I find here's my take on it lately is just Go going little wins rather than trying to overhaul your enterprise is just find those little pieces that have been an annoyance and trying to get them out of way because I you know, I worry that Because of the way things are changing so much it's hard to put your flag in the sand on any one piece of technology at the moment so if you're trying to
Marc Binkley: Totally.
Pete Bishop: let's say rip out your operating system and put in something like this. I'd scared to death to do that just because A, no really got the security figured out and B, there's something two weeks from now that might change everything. So it's just easier to ⁓ get quick And that's where found it been successful. ⁓
Marc Binkley: Totally. Totally. Yeah, it's complicated. I spent, had to have been, ⁓ think I started on April 1st and I spent a solid three weeks, just nights, like obsessing about building agents. After, and I've been using projects a lot in Claude and in ChatGPT and in GeminiGems. And so I was like, okay, I'm gonna go all in on this Claude thing. I leveled up and got the.
Pete Bishop: You
Marc Binkley: the big max subscription and figuring out what this Claude code thing. And then after three weeks, probably 50 or 60 hours worth of work, I asked Claude, I'm like, ⁓ what we doing? Like, it feels like this is not working. It's not doing as much. And then Claude's like, yeah, you're better off going back to what we'd had before. It doesn't make any sense. I'm like, great. So we just start from scratch again.
Pete Bishop: I like how it's so positive about bad news, hey? It's just like, you're right, you're right.
Marc Binkley: Yeah, yeah, you know, you're right. I should have caught that earlier.
Pete Bishop: Yeah, I I wonder because you see the kind of adoption curve that usually happens a big trend and then it levels out. It just feels like this one's taking a little longer because it's such a big thing, I think. But.
Marc Binkley: Mm-hmm. Yeah. Yeah. There's, there's a lot of, a lot of free usage on all the platforms that I've seen. there was a report from, the Ehrenberg Bass Institute on ⁓ they're looking at jeopardy marketing law, not the movie one, on whether, whether like the loyalty is falls ⁓ in the same patterns they've seen another. industries and like from aviation fuel to toothpaste in China. And it does. What was shocking to me about that one is that Claude is so low on the adoption curve. But everything I look at is 100 % Claude. And it's, I'm in my own bubble. Like my, my feeds are all adapting to Claude because I'm using Claude so much. So I think that most people use Claude, but that's just cause it's me seeing that I'm using Claude and everyone I'm talking to is using Claude.
Pete Bishop: Really? Right.
Marc Binkley: And all the people I'm interested in talking to right now are using cloud cowork and they have ideas on it. And so like I just, it's feeding all the algorithms are feeding that back to me. So the reality is most people are just using free tools. Most people are using chat, GBT or Gemini. and not really in space of, of building out agents and all that kind of stuff.
Pete Bishop: Huh, interesting. Hmm. You and I had also talked a little bit about this of sea of sameness. Do you mind just describing that a little bit? ⁓
Marc Binkley: Mm-hmm. Yeah, totally. Like I've been really fascinated in AI and trying to figure out where it helps and why it helps and like to get around this, you know, why 95 % of implementations get zero return. So I've interviewed a lot of people around this, yourself included. And I've interviewed some people from like the big holding company agencies and Roger Martin, I talked to him a couple of guys or Peter Weinberg from Evidenza. there's some, I've had a pretty interesting perspective, like because I'm trying to write these articles for work on AI and what it's doing and the best ways to use it and stuff like that. But the sea of sameness thing kind of came from an aha from Roger Martin. And we all seen this, but like the way he described is that it's optimizing towards the mode. So it's not, it's, this kind of center point of, of the herd of answers that are out there. so tools like and like Claude 4.7 got the highest ever score up until two weeks ago. think it was when chat GPT released their new release and it's the score is on the GDP Val score.
Pete Bishop: Yeah.
Marc Binkley: And what the GDP Val score does is it takes, it scores the ability of these tools to do knowledge work. And they look at, I believe it's 77 different individual professional services, jobs. And they go, how well does it do like, kind of work or how well does it do marketing kind of work? How well does it do copywriting kind of work or legal work or like across the spectrum of 77 different industries. And so it's scores the equivalent of what a 14 year veteran would get in any one of those categories. Which is kind of crazy when you think about it has all of the knowledge, but it to the mode, the middle, because it's trying to be appealing and it's trying to get us to use it. So that's why it says things like, good catch. And I actually did this on three of the different platforms at one point. And I was like, why do you keep saying things like good catch or, know, I should have thought about it that way. You're absolutely right. And then the reason, and if you do this enough times and ask whatever tool you're using.
Pete Bishop: Yeah. Right. Hahaha
Marc Binkley: And eventually will come around to saying, ⁓ it's because I'm optimized to being not correct. And so it picks the mode because it's the closest thing that it thinks it can get to that is an agreeable answer. And exactly why it says things like, if you say to it, any of the LLM tools, guess the number between one and 10. The first number will come up with seven. The second number, I think I'm getting this week and seeing right, but it'll be nine.
Pete Bishop: You
Marc Binkley: and the third number will be three and then it'll be one. And the reason for that is because that's the mode of what most people pick they're, it's not actually a random number, but it's at a very agreeable number. And it sounds like something that's random, but it's not.
Pete Bishop: Ha! Yeah, interesting. You always have to think about at the end of the day, it's a prediction machine, right? And it's trained on ⁓ pile of data that, and if you know where the sources are, right? You get a sense of, I there was a fellow I know that every, everything he was generating at the end of it was kept saying, thanks for tuning in. And he's like, I didn't ask you to do that. Why does that keep coming up? And it's like, I think this was Gemini was it to him. And he's like, He figured it out is because it was trained on YouTube and a lot of the transcripts would end off with totally. So it's just a good reminder. And the other thing I follow this one guy, he's really, he's really funny. He's super deadpan, but he will, he'll just try and trick AI into doing stuff or lying or, or contradicting itself.
Marc Binkley: ⁓ smash that like button.
Pete Bishop: And he made this song where all he did was just make a bunch of fart noises. And he's like, what do think of my song? And it's like, yeah, you've really found like a new genre. You've really uncovered something that's, you know, and niche. And, and every time that I'm thinking of doing strategy work, I just, have that in the back of my head. I'm like, it did that to the fart song. So ⁓ it's like saying I'm super smart right now, I really can't trust ⁓ to degree. ⁓ Yeah.
Marc Binkley: You Mm-hmm. Mm-hmm. Yeah. Yeah, it's true. And I think that's the thing where you get it. The machines themselves are confirming biases first, because that's sounds agreeable. And we like things that agree with us. So it's not like if if you think about their objective, which is to get people to use the tools, because that's how they eventually make money, then they're not going to optimize for providing conflicting evidence or data about who you are. That was actually one of the responses I got if It was something like, I think it was from chat. It was something like if there were two people in a research group and one was, you know, questioning you on the methodology and the results, I'd be the one that would say, yeah, that's great. Keep going. I'd be the one cheering you on. And so it's, just the thing. It's, it's pushing us to, to believe more of the things we, we that sound right to us.
Pete Bishop: Right, which is part of that sameness, right? When you are eventually, if everyone's kind of producing stuff that's agreeable and in the middle, like you would say, then that lead into a little bit more of the value of creativity moving forward, because I think that's it's getting confused now with what creativity when you're using these tools. Do ⁓ you that?
Marc Binkley: Right. ⁓ Totally. ⁓ We had a panel last week for the CMA and like it was, was around stop talking about AI, start making better decisions. And, set up the talk, had, I had created some images, starting with an image of myself, like a bio headshot kind of thing. And that was the first image. And then I just put it into nano banana and said, make me an orangutan and it made turned me in and rang a tang. And then there was another one that was me surfing and then looking Dr. Strange ⁓ when was doing that 14 million choices thing, looking for future and looking for the answers. So there's all kinds of things that you can do with it, but making a piece creative is not necessarily the same thing as having a good piece of creative. And I think there's a whole bunch of things around the strategy behind that, the thinking that needs to be more than ever before, because anybody can make something, but it doesn't mean that you're making something that's good. So I think that's part of our role is like to avoid the CSA and this is to be the tastemakers.
Pete Bishop: Yeah. Yeah, I found that too. This willingness to run with any idea it's got, right? So it'll come up with, here's three campaigns. one do you want to run with? And if you said any of them, it would just start. It would just ⁓ out your ads and your PowerPoints and everything, because it has no sense of this is good or this is bad. And that's the taste is what you're mentioning, ⁓
Marc Binkley: Mm hmm. Yeah. Totally. Yeah, we had ⁓ last year we did a workshop with Carl was Carl from zero to 60 and it was a workshop on we had ish people in the room and so we had we divide the room into three groups. So one was the vibe marketers. So they just we gave everybody same brief and the vibe marketers just ran with it and they just did the things that AI was telling them to do. And at the end of the three hour workshop, everybody's going to produce, creative, you know, ⁓ to serve the create the campaign brief that we, that we had laid out to them. So the vibe markers. They just ran with it. And like a hours later, they got this ton of cool stuff. none of it on brand that they admitted it produced charts and all kinds of stuff. They're like, no idea where this information came from. And then the other one was there's another group that was the traditional marketers. And so they were feeding it with prompts, things like, using Maslow's hierarchy of needs and the funnel and reference Simon Sinek start with why and come up with creative to suit the campaign that we have been to produce a result on. And so theirs was interesting too, because it looked really good at the end of it. They, had a jingle and dance. was for. a fictitious pro-g company. They had dancing pro-g's. It was crazy what they could make. And then there's a third group that was an evidence-based group, and they were there given specific prompts around like category entry points, distinctive brand assets, making sure that they were aligning the distinctive brand assets up with the category entry points that initiated a purchase cycles and triggers. And so there's a whole bunch of stuff around that that were using, and they also produce some great stuff. But at the end of it, When you, if you were blind to, not blind, but didn't know the objective and just were looking at the creative, it would, at the end we were like, that's really impressive that everybody made all these things, um, in a very short window of time, but nobody had to defend their decisions to an executive room or to a client. And so I think that's where a lot of us would fall down because the vibe markers admittedly said they couldn't justify any of the things that were there. They're just, they've made stuff. Right. And then the traditionalists, I think you run into the same situation that markers have been facing all the same for decades, which is they have a hard time justifying the budget and hard to find defending brand, and a really hard time kind of proving any type of ROI on those kinds of investments. Whereas think the evidence-based people would have had a better chance of getting something pushed through in my opinion.
Pete Bishop: Mm-hmm. Mm-hmm.
Marc Binkley: And so, but each one of them went in with somewhat of a bias and machines confirmed their bias.
Pete Bishop: Yeah, you're kind of leading the question, right? think our director, Kurt, I think he talked about this a little while ago, which was a good little phrase. was saying he feels his job is to help people make decisions. And in some ways, I think that's true. know, it's companies are messy by nature, which I think is part of the reason AI really has a struggle when it's, you know, it's easy to ⁓
Marc Binkley: Mm-hmm. Totally.
Pete Bishop: set up a company when you're a person of one and you're using Google as your backbone. But if you are, know, a 10 year old company trying to rebrand, and then you need to, like you said, talk in front of a board and talk about, you know, the of the brand and how it's going to last for the next 10 years and all that kind of stuff. And if you just generated everything out of chat, GPT, it would be really hard, I think, to, convince people to that change, or at least, ⁓
Marc Binkley: Mm-hmm. Yeah.
Pete Bishop: stand behind the work and where I think this stuff falls down a little bit. ⁓
Marc Binkley: Yeah. And so like in the vibe marketers case or, you know, there's lots of other examples of this, but there's a great line that should deliver it at the speed of light is still shit. And it's not because we have these tools, it doesn't mean it's helping us solve any problems. It, we could just be doing more of the faster.
Pete Bishop: Hahaha you Yeah. Yeah, I like that. You know, here's the other thought I've been having. I don't know if you've heard this story. It's an old story, but and I wish I knew where it came from and it could be complete fiction, but it was this thought that this group, whatever study group brought in these creative directors from three different areas and they They brought them into this room and gave them a brief. And on the way to this room, they drove them intentionally past a bunch of things like ⁓ the zoo and some billboards and stuff. And then when they got them all together, they gave them the brief and then they all went into separate rooms and to crack it. And they all came back and roughly presented the same creative that seemed very out and whatnot. But it was all based on these like subconscious cues that they intentionally planted.
Marc Binkley: Mm, yeah. Mm-hmm.
Pete Bishop: me wonder. you know, this is getting a little nebulous, but as you getting into creativity is like, are all product of our own inputs, right? The books we've read, the movies we've seen, the we've had in our lives, you are we limited by that? And is this opening that up or are, know, you're talking to computer that has all the books and all the movies and all the campaigns available to it. you know, maybe it down to that ⁓ or whatnot, but
Marc Binkley: Totally. Totally. Yeah.
Pete Bishop: does make me worry a little bit on the creativity side. You know what I mean?
Marc Binkley: Yeah, it does. But I kind of look at it as. Like limitless possibilities. So going back to the GDP VAL score, right? There's 77 different profiles of professionals that they score against. I think I'm pretty sure, I want to say it's 77, maybe it's 44, but anyway, it's a lot of people that they score the knowledge, the ability to deliver knowledge work against. And it's, by the way, it's as good at doing something in seconds that it would take someone seven hours to do. And that's the score. So that was the reason why I had that picture of Dr. Strange up of my head on Dr. Strange. Cause it's that idea of like, there's infinite possibilities. You can run a Monte Carlo simulation in any of these tools to come up with the answer, the one choice, right? helps beat Thanos or whatever. Like you don't have to take the mode answer because there's all this knowledge out there. And so, ⁓ One of the things that I've done, I did early on when I was playing around with, all the different tools is I built in the projects. built, council, a conflict, you can it a conflict council if you want, but a group of people like based on, created personas based off of all kinds of people. Like I put Roger Martin in there. think a lot about him. I put in dead people Steve jobs. I put in Sun Tzu, put in Genghis Khan. This is first my strategy strategic advisory council. but when I do strategy, like I have ideas that I'm talking to Claude about and then my strategic advisory council chimes in on that. so trying to create differences of opinions to make me think differently. their personas are kind of accompanying me in the decision making process. So I'm not just taking Claude's word for it. I'm using some of their input to help me re shift and reevaluate shine lights on some of the dark spots that I haven't thought about.
Pete Bishop: I love that. just so everyone understands, how are you putting a personality in as a counsel into AI?
Marc Binkley: So how I did it originally is a little bit different from how I'm doing it now. And this is my experience with wasting so much time trying to build agents, which isn't really a waste. It's just like, it's just learning. But initially what I did was I, cause in most of these tools you can load up supporting documents. So I had cloddered
Pete Bishop: You Yeah, I get it. Mm-hmm.
Marc Binkley: ChatGPT and Gemini, all kind of the things doing the same type of thing where I would ask them to create basically like a canon of the most famous works from these people in a document. And then I loaded that document into the documents for that project. And then in the instructions, then I would say whenever I'm asking questions about this or that or whatever, however I wanted the pro like I the system or the process that I want to take. I want the, how I want the project interact with me. Then I would ⁓ make that it would reference my strategic council and give me at least three or four quotes or perspectives from their work and my thinking to make me better and make the outcome better.
Pete Bishop: Hmm. Yeah, that's awesome. I've heard of people doing that with boards, right? They, yeah, they take everyone's personality and basically bake it into AI, which I'd also be like, Wow, I wonder what it'd be like if anyone ⁓ read own, you know, personality as written by somebody else. But it's a great way to like, Hey, how is this presentation going to go over? Right? And how do I modify it so that, you know,
Marc Binkley: Yeah, same idea. Yeah.
Pete Bishop: Dan over there doesn't pick it apart as usual, right, or whatever.
Marc Binkley: Right, yeah. Fuck, Dan is always that guy. You
Pete Bishop: Yeah, that's super I like has it when you've been doing that? Has it gotten you to better places?
Marc Binkley: ⁓ yeah, for sure. Yeah. Cause it, I mean, it gets to the point where I'm like, try and pick the divergent thinkers. So they're not, they're all the same thinkers. but yeah, like more and more, I kind reconsider who I would want to have in the board. And I kind of reshuffled around based on some of the responses I'm getting. but like, Einstein is in one of mine. I've got, ⁓ you know, in a creative uh, project that I've got, George Lois is in there. Um, you know, I don't know, just people that I think, uh, Picasso was in one of the, the creative one as well, like just different people with different perspectives that I'm like dead or alive. It doesn't matter. Cause a lot of these people have published ideas and literature and stuff that it knows about. And I'm just formalizing some of the things that I want it to reference when it, when it challenges me. this that I the most recent article was writing, there was thing I found. There a survey that was done asking 1500 executives what their biggest challenges were. So what came back was bouncing short and long-term collaborating cross silos, protecting margins in commoditizing industries. These sound familiar, probably, right? The crazy part is that that survey was done in 2008.
Pete Bishop: Yeah, yep.
Marc Binkley: So the big problems that we've faced our business are still the big problems that have been around for decades. And I think when you look at the usage of things like chat, GBT, there's a study I found that showed that the number one usage at work for chat GBT is copywriting. So I think some of the things that we've been talking about, like challenge groups or advisory boards or whatever you want to call it. And then things like you have access to.
Pete Bishop: Right.
Marc Binkley: the world's information pretty much you can run Monte Carlo simulations like we don't have to trade off solving big challenges wicked problems with delivering shit today like it can be both and I think that's kind of an interesting thing
Pete Bishop: You Yeah, I think that makes sense. you find like I've heard this many times is how content is the number one. Do you figure is or writing anyways, creative writing. It just felt like that was the very first example of AI when they first, I remember seeing, I don't know how old this was. It's got to be over 10 years of seeing an example of AI writing a blog post and they'd show it side by side and everyone, everyone's jaws dropped. It just feels like the very first initial application of a text-based AI. And then all this other stuff's relatively new. Like you think of over the last year, agents just ⁓ I think within the last
Marc Binkley: Mm-hmm. Totally. Yeah.
Pete Bishop: year at least in the public realm as far as I remember right so all of these updates and stuff which is going at an exponential rate like the ability for us to adopt seems so hard I read I do you read Sapiens ⁓ I can't remember who it good book but it was ⁓ his point was is that
Marc Binkley: Yeah
Pete Bishop: from just an evolution point of view, we are still adopting to agriculture. That's how slowly we're getting into things versus tech is so far beyond us. And the rate that this is coming out right now where they used to take us like TVs came out. It took us like 40 years to really get TVs going in our Now we have something coming out almost every day. I just don't believe we can.
Marc Binkley: Mm-hmm. Totally. Yeah.
Pete Bishop: To your point around, this isn't an effective tool for a lot of people. I just don't think we know how to get into this stuff that quickly. It's just so fast.
Marc Binkley: Yeah, I agree. And I think if you look at the digital landscape, I would say 2012, ⁓ just mark a time, a moment in time, that's when Facebook went public. That's when I really got heavily involved in digital media, social media, probably around that time, a little bit before that, or actually, but anyway, that's just a good marker. And so ⁓ here are, help me with the math, but I think 15 ish plus years later. And it's now taken over digital has now taken over traditional in terms of total ad spend. Like if you look at, e-marketer charts and that kind of stuff, right? Like the, the dollar spent like for every dollar, more than half is now going to digital platforms, but that took 15 years ⁓ least.
Pete Bishop: Right. Mm-hmm. Right, I remember that. Yeah, I remember it creeping up and they were like, it's never going to be more. Right? Yeah.
Marc Binkley: Yeah. So, you know, for all the hype though around how everyone's doing social, you got to be on social and like I was at conferences where that was a thing and, true. And that's, and it's not to say that it's not important to understand. I think AI going to be amazing for a lot of companies to do this, to do dangerous, but also amazing, like same kind of thing, double-edged sword, but, but it's still going to take people are people. It's still going to take a long time to get. used to it and figure out how to use it. And especially organizations, think agents are so hard, my experience with, don't I'm that unique. Maybe I, I'd say a sock, that's, but maybe everyone figured out on the first time, but it's really hard to explain what we do. Like, I think we take that for granted and it's like, we wanted to do all these things so quickly, but you realize quickly that there's a lot more nuance and a lot more.
Pete Bishop: ⁓ yeah.
Marc Binkley: comes into what we do. And then you have to think about what do in a way that is not ⁓ natural, we're just so used to doing it.
Pete Bishop: Yeah, that's been my experience. This agent building is relatively simple, but it's coming up with all the business rules and the the strategy around what you're trying to build and how it's going to work with your own messy background that makes it tough, right? So, yeah, I get it. And think that's the disconnect right now that's going on. It's just everything's simple and should be automated and you're falling behind and you're missing out. And then there's reality where everyone's like, yeah, this doesn't for me. My company doesn't work like that. We've got a million other variables here, but we're all buying subscriptions and everyone's working on it.
Marc Binkley: Yeah. Totally.
Pete Bishop: Yeah, it's a bit of chaos.
Marc Binkley: Yeah. And I think where it gets even more complicated is if you start thinking about the organizational way you do things, like kind of thing, that becomes even more difficult to codify. Cause that can, that ⁓ change one department to the next or one manager to the next. And so if you're trying to produce something that's ⁓ consistently good,
Pete Bishop: Yeah.
Marc Binkley: you could do that at an individual level, maybe, but then at a departmental level or a corporate level, it becomes harder. And so I think there's a whole bunch of thinking around decoding how. Businesses function that. Well, we'll face a lot of opposition because then you're ⁓ people to give up their secrets. How do you do stuff? And then you'll end up finding that a lot of people won't want to share that because that's the special sauce that.
Pete Bishop: right.
Marc Binkley: is what makes them expendable.
Pete Bishop: Right. Yeah. How do you, can you break down what you do exactly and put it into a document so that we can get rid of you? Wait a minute. Yeah, I hear you. Okay. Well, Mark, I appreciate you coming on. I feel like we could probably chat for a couple hours at least. In fact, the next one I booked you in for about nine hours. But
Marc Binkley: Yeah, right. I've seen this movie before. Wait a sec- Yeah. No!
Pete Bishop: This one, unfortunately, we've got to wrap up. So thank you. And I just wanted you also have your own podcast, Sleeping Barber, right? So I'll stick that in the show notes as a link.
Marc Binkley: Yep. Yep. that'd be great. Yeah, ⁓ going to be going back to Cannes this year, actually, as media to cover the festival. So be a lot of fun. We had a lot of great interviews lined up and we just have actually published a lot of great interviews this year. yeah, it's a lot of good things happening with that one too.
Pete Bishop: ⁓ cool. Okay. ⁓ Well, thanks again, Mark. we'll have to get you back. Thanks again.
Marc Binkley: Thanks, man. Yeah, that was great. That was lot of fun.
Pete Bishop: Okay, how does that feel? Seem okay?
