Mark Ritson on Marketing, B2B, and the Future of AI

Welcome to LabGrone Marketing, podcast rewriting the marketing playbook for the synthetic century I'm John Lombardow. And I'm Peter Weinberg. We're bringing together the latest thinking in marketing with the latest breakthroughs in AI. So if you want to be in the top 1% of marketers, you're in the right place. On today's episode of LabGrone Marketing, we have a very special episode, the season finale where we have the greatest guests that money can't buy. You can pay me if you want. I'm available at PhennyWings. Hey, hey, he could do whatever he wants. Ritsik could do whatever he wants. You know, John and I, through our time at the B2B Institute, we were fortunate enough to work with all the great marketing effectiveness thinkers of our time with Les and Peter and Jenny Ramania, Karen Nelsman, all these geniuses. But I would say ultimately no one has taught us more than Professor Mark Ritson, our guest today. And we're going to be talking about Ritson on marketing. Some of the biggest lessons all marketers need to know. We're going to talk about Ritson on B2B and what makes B2B marketing different or the same as B2C marketing. And then we're going to finish with a thrilling conversation, which is Ritson on AI. What AI can do, what it can't do, what it will be able to do. It's going to be a great episode. Okay. Professor Ritson, Ben Vanute. Bonjorn, no professor. You lab-grown marketing. I'm sure there's a career highlight for you, right? It's up there. It's certainly up there with the Starbucks I had this morning. It's certainly a Sartorial highlight. I mean, look at that beautiful top hat. Look at that hat. I see both of you know. Gorgeous. As both of you know, I spend a lot of time on personal branding. So this is the season finale of the podcast, which is very exciting. Usually, I think you kill a character off in the season finale. We're introducing in the character. Which I was hoping will be John, by the way, I think all the listeners were hoping. I think so. But instead we're adding a new character. One of the greatest characters in the marketing industry, Professor Mark Ritson, thank you for joining us. It's my pleasure, gentlemen. And as our listeners should know, is it listening to you or what's your, what's your both, both from listen to the podcast? As your pod friends may not know, we are very, very old friends from way before this era. We've known each other a long time. We are. Do you remember when we first met? Do you have any memory? You want to walk down memory? I'll give you the specific moment we first interacted. Yes. I rang up, John, as it turns out, because I was trying to think to get linked in to buy the mini MBA, which is still a good idea, by the way. I was drunk at the time. I was drunk in a hotel in Melbourne. It must be more than 10 years ago, right? I was sober as a Catholic priest. It was like six in the morning. We were in LA, actually, I think I remember this. So I rang John out of the blue and I said, hey, I've got this program called the mini MBA and I just wanted to talk to some people in the LinkedIn, B2B Institute about partnership and you know, I was doing the softly, softly. And you went, I know who you are. I've done that course and I'm like, hello, right? We are proud alumni of the mini MBA, which I will, and put the best career in investment. People at LinkedIn are all my friends. Became selling agents afterwards. Yeah, we've got a lot of benefit programs. And I have to return the favor. And then you came and taught my module 11 in AI. Yes, a career highlight for us. Which remains very popular and it gets me out of jail talking about AI. Love it. Love it. It's not that I don't have a point of view as we'll see. I'm sure later, but it's a real pain in the ass to record a class on marketing AI in April. And then it rolls out in October and 80% of it is kind of like, yeah, but what about Gbt7, right? And you're like, ah, okay. Yep. I want to talk about your opinions on the wonderful naming. Yes, we'll get into that. Oh my goodness. So in honor of the mini MBA, wonderful program that it is, we have three segments today, MBA. So the first section is on marketing. The second section is on B2B and the third section is on AI. Yeah, that makes sense. And we are going to start with marketing. I can't even count the number of things we have learned from you over the past decade. Our greatest mentor, but this is sort of a highlight in my mind of like some of the things you have said that live rent free in my brain. And then I want to start with the one that's probably like the least relevant to what you just said. No, I think it's great. So last night we had dinner at El Molino in New York, which is delicious by the way. Fantastic. And we missed Italian establishment. So we just talk about the Ville Parmesan for a moment. It was exceptional. Not as exceptional as your cookie father-in-law. Yes, my father-in-law was very happy. But let's talk about that. I think the less said about him, the better, but the more said about the Ville Parmesan, the better. The better. The thinly pounded, just absolutely delicious. It was. Highly recommend. Divine. And you said something. This actually equip I've never heard from you before that I loved about how you beat Roger Federer. Will you tell the audience how do you beat Roger Federer? With a big stick. Now, we were chatting with your cookie father-in-law brought up. He's treated the appendages of many of the world's most famous men. He has, yes. A sort of a celebrity doctor. But he brought up Roger Federer at one point, which queued my, one of my favorite strategy analogies of all time, which is how do you beat Roger Federer? And the answer is you don't play him a tennis. And it's a beautiful example of a couple of things, right? It's obviously that sense of competitive strategy. It's also a little bit of core competence. You know, don't play someone at their game. And it's also a little bit of kind of differentiation, you know, pick your battles. I always remember when you do really good positioning work at a higher level than the normal dross, you're also thinking about what am I positioning to the market. Yeah. Not just, you know, I'm going to position my abacus against your calculator. But maybe I won't talk about the device. Maybe I'll talk about the system instead or the benefit. You know what I mean, like what do I want? What I will choose the playing field on which I will fight against my competitor because I can do that, right? So it's not just, you know, all your red, I'm blue. Maybe we want to talk about color because color isn't my strong point. I'll choose the playing field. And I think that's a quite a high level strategy thought to kick off the podcast. It is, right? There's segments you should avoid because they're owned by your competitors. There's positions to avoid owned by your competitors. There's distinctive assets owned by your competitors. Don't play Roger Federer in tennis. Also, I love this a lot. I'm not a poker player myself, but a lot of people in the AI world, for example, who are very smart mathematicians or computer scientists, they're very into poker. And a lot of it is, they talk about knowing which table to play at or which table to play. There you go. It's a very, very important question. The thing I find fascinating about Roger Federer is for all of his grace and magnificence and success, he only won 54% of the points he ever played. And so it does also go to show you like poker or tennis or even probably marketing. If you can just be right 54 or 60% of the time, actually with compounding, there's a real advantage over a long period. It's funny this morning because I'm getting up at obviously 2 a.m. because I've got jet lag. I was listening to the podcast about Leeds United beating Everton overnight. And the point that the commentator made was the Leeds fans gave him an extra 2% and in a 50/50 battle, that's a lot of you need. That was a very high end point for a football commentator. You have famously said that if you could give marketers one gift, you would give them the gift of market orientation. Could you talk a little about that gift and why it's such an important gift to bestow upon marketers? Yeah, I mean, it's the start of marketing. So most marketers get into marketing because they're into advertising and what it's done to them. They've got a favorite ad. So they're starting from the tip of the spear and going backwards. The main thing is that most marketers are all about advertising. There's nothing wrong with loving advertising as an entry point in the marketing, but then you have to put down the spear and you have to go right where does marketing begin and what is the origin in order to get to the right place at the end and the answer is market orientation. Otherwise, known as you are not the customer. And it's worse than that. Not only are you not the customer, you are the least qualified person in the world to review your product or service because you made this thing. It's like children. You all have beautiful children of similar ages. Shout out to James Lola who love the podcast. Oh, ladies, friend of the pod. The daughter of the pod. Ladies into it. Roxy, my daughter's into it. Ray, you know, he's just appreciated as much. I think he's slightly behind too. He's a performance marketer. He's behind. Anyway, but we all look at these children and it's very hard not to assume they're not a incredibly attractive and be going to be some kind of really impressive. Right. I understand. And that's how marketers are about their products. Not being blind to the realities of it, you know. And so market orientation is a discipline that you can train in marketers, which is you don't know nothing. You are literally in a dark room with all the lights off when it comes to your price, your web experience, your packaging, the whole you're advertising. You don't know. Even though you think you know, I had a very famous marketer get very excited about an ad that she sent me on LinkedIn. So we got this ad and we've, one of our users created it, but we like it so much. We're going to put it out there. And you think, and I had to be quite hard with her. I'm like, I'm not your target market. And neither of you, I would test this. Go ask a customer. This has been tested once I want to say I believe Aaron Bergbast did a little study around. They did. Yeah. Like whether marketers could judge whether I think an ad would be successful or not, average marketer no better than a coin flip. You're nothing. I think the insights department maybe knew like a shade more, but I mean, probably within margin of error. You're right. Nobody knows anything and it's been empirically tested. They updated it. They updated the distinctive brand assets stuff, right? So they looked at how consumers thought about distinctive brand assets kind of confirmed Jenny Romney's work. And then they checked the brand managers who owned the distinctive brand. and that says, and they were way off the basis of consumers. So yeah, the core lesson of market orientation is you don't know anything. And that sounds negative and redundant, but what it does in a market as head is it breaks them down. You know what I mean? So you break the marketer down to realize you don't know anything, and then suddenly data becomes useful. Because again, I mean, the analogy I use is business school professors, who I've worked with for an excruciating long period of time, we all get at the end of our MBA classes feedback. You know, giant pad of feedback, where each student who spent $5,000 on this course says, "Now, yeah, let me tell you about this course, "right, I'm gonna fill out this stuff." We all get that, and the difference between good and bad teachers of business school is not natural teaching quality, 'cause we all start out pretty bad. It's whether you read that shit, and you change. Do something about it. And my point there is it's not market research that makes a good marketer. It's first the ability to listen, absorb, and respond, which comes out of market orientation. So in the marketing process, market orientation precedes market research, otherwise you're pouring water in the wrong hole. It's not gonna go in there. This is a fast and any point, especially for our business, which is a market research business, it's like the marketer's just ignored the market research if it doesn't tell them what they already want to believe, right? So you have to start with this point of ignorance and humility and truly you have to be open to listening. Before you've commissioned, right? - A broader idea, a lot called the Inside View and the Outside View, that I'm familiar with through the finance world in Skyme-Lykel, Mabison, but it is true that the Inside View is not representative of anything. You're not the customer, you're a worse position to know it, and you ought to always pair that with the outside view where possible to really understand the market and then make decisions, right? And so you can do that qualitatively, you can do it quantitatively, you can do it with experts, there's lots of ways probably to get at it, but people are mostly operating off of the Inside View, especially in B2B where they have very little access to the outside view. - Well, and they may not ever get it without data, right? I mean, it comes back to our famous American Eagle campaign, right? I mean, it's been an excruciating three weeks, watching people, flagulate themselves over what you're called again, the one that's called the Sydney Swini-Ed. And it's the end of work advertising, if you're the right wing or what you see is, the end of woeckiness, and if you're a left winger, you see general sort of white supremacy and eugenics, right? So the answer to that was very simple, it was like, it doesn't matter what you think, yeah? Most of you don't look to me like you're buying jeans any time soon on the high street, and the answer isn't one or the other or neither. It's, let's find out what the target market think. - Yeah, and also advertising. - By the way, there's a fun story about market orientation. So I was talking to my wife about, she brought it up actually, that's often I get the customer's perspective from Alex, my wife. And she was like, did you see that Levi's campaign with Sydney Swini? And I was like, there we go. Nobody in the marketing industry talking about how it actually violated one of the number one rules marketing to his-- - Wasn't very distinctive. - As Ritz and I, it says, first they must know it's you. - I agree with that. - I agree with that. - You attributed it to the wrong brand. - Oh, I agree with that too. - Advertising is also, is a, people forget is a very weak force. So people attribute, I think way too much, kind of impact to advertising. Does it have an effect? Yes, the stock price actually went up, although I'm not sure what it's done since then, but I think people overthink the advertising stuff, especially in the advertising industry. - I like Les Binett's point about in these 30 years, he's never seen an ad campaign ever damage the sales of a brand, right? So whatever it does, it's not gonna damage it. - So, John and I have always been very passionate about finance and marketing, and there's a whole sort of subculture and marketing effectiveness on finance and marketing alignment. You, I think, have a very fun take on this. So tell us your thoughts on that discussion. - All right, I'll go there if you want. - Please do. - I'm sort of bored with everyone that says, marketers need to understand the language of the boardroom, marketers need to understand finance, marketers need to understand the C-suite and CFOs, right? My point is really very simple. I'm sorry, I don't think marketers need to know the language of the boardroom. What they need to know is fucking marketing. And the biggest problem we face in marketing in every major economy are marketers that don't know their trade, the end. And the last thing I want are these quite intelligent, well-meaning people with not enough operational skills, going off on a course on fucking corporate finance. - Right, gap. It's like teaching someone that can't swim, how to life save. Hey, great, but let's just concentrate on you keeping your head above the water. Most marketers, you had the star, original. It runs about 25% in America, right? And it's always hard to define where is the formal line of training, right? It sure is shit isn't watching YouTube videos of Gary V. in your underpants, which is what a lot of people think it is. A formal training doesn't have to be a degree or an MBA, but you've done a course in marketing at a graduate level if you're a marketer. It doesn't seem like an unreasonable requirement. - How much, I will ask you a question about this. How much financial training is in a marketing degree? - Zero. - Zero. - Even at Wharton today, if you go majoring, you got an MBA in marketing as your major, it's gotta be some financial training. - I would bet, and I don't know the water, I mean, it's 30 years since I was a warden, but I'll bet you that there's an elective in marketing and finance. But it's an elective of the corner of the main marketing. - You also made a really interesting point, dinner, which I had not considered on the finance front, which is that many professors of marketing actually get their undergrad in math or an econ or in psychology. - So we've given marketers a bunch of shit for not having formal training in marketing. That's one thing. I do have some sympathy for them, because if you look at the formal trainings in marketing, they're by and large, not very good. If you look at places like the University of Chicago, let's call out one of the finest institutions in the world. What you'll see across their faculty are wonderfully smart people that have zero experience of working in marketing and have, for the most part, no undergraduate training in marketing and for the most part in their PhD in graduate studies, also no training in any marketing relay topic. They worked in economics, they worked in quantitative analysis, but not in marketing. I've seen how these guys do it. Again, they're very bright, but they know nothing of the thing they're teaching to MBA students who've come to learn about the thing they're not teaching. They say things like, I've had a direct experience of this, the first class in market research taught by someone that has never worked in market research at a very top business school is to say, market research is fundamentally about statistical analysis. And then let's do nine and a half weeks of statistical analysis. Teaching MBAs things that Excel can do in point two seconds. - Yeah, well, I love this. If it's statistics, kind of previous degree for the professor, market research is just statistics. It's a thing I'm from psychology. - It's just a cycle of thinking. - It's a cycle of thinking. - It's really just a thing in economics. - Of course. - Of course. - Marketing's really just option theory. Here's nine and a half weeks of option theory. That's our problem at the moment. We're a very, again, it sounds like I'm boasting and I'm not. I won the teaching prize at MIT seven weeks after I got there. But that's a reflection of a wonderful school with wonderful people that didn't have practical experience in the field to supplement the academic world. In my opinion, there are people there that have that, right? But there are also too many people that don't. And if you contrast it, we did it yesterday with your father-in-law who's a very well-known practitioner and a professor in the medical field, the idea that you would be a professor of surgery but not actually do any surgery at the moment or have ever done any surgery is hilarious, right? - I mean, I think the bottom thing, bottom boom, is that stuff out there for marketers, they have no formal training in marketing and if they were to go get a formal training in going to one of these like top MBA schools, they'd often be, you know, no better off. - Well, it's worse than that. You will be told as I was as a junior professor, you're not allowed to do consulting. You're not allowed to work in marketing. You have to concentrate on your inane research. - Right, then nobody reads. - That nobody reads. - And which is based on, you know, an echo of an echo of an echo of an echo of an actual problem. So it's a real problem, it's a real problem. - So maybe the last question in our marketing section 'cause we gotta get to B2B and AI, could you give us your riff on price versus pricing? I think this is a very subtle thing that I come to appreciate. - Well, I've enjoyed your personal evolution, Pete, because many years ago when I would tell you pricing is the most important thing. - Yeah, I did not get it. - You looked at me with that polite kind of, yeah. - Yeah, I'm sure it's in smart, but I have no idea. - You must be drunk again, right? - Yes, must be drunk again. - And yet I've watched this beautiful evolution as you've run your own business to go, oh my God, pricing is everything, right? - Who know, who know? - I didn't know, now I know. - So what do I mean by that? So obviously everything in marketing has a price. Yeah, you can't avoid a price, you know, the sticker. Pricing the verb is a bigger thing, which includes the price, the sticker in the middle, but before we get to the price, if we're doing pricing properly, we've got the research that informs the price. And once we've got the price, we have the more important thing of how we manage, communicate, frame, change, emphasize, de-emphasize the price. And it sounds bizarre, but it's absolutely true that if you look at the three elements of pricing, the research, the price, and then the communications of the price, the price itself is the least important factor in that equation of consumer perception, right? If we do research into price, we usually will charge a lot more because suddenly costs are not the main driving force and our dumb competitors are not the main driving force in the price we set, we go to the source, which is the consumer, and we understand what's the value, what's their situation. You know, I used the example yesterday, of in my hotel where I'm staying, it costs you $7 to get a shirt laundered, right? Why? Why? It's a $0.40 process, right? Because the competition for your laundering of your shirt in a hotel is to buy another shirt, which costs you $50, right? So the research has to be done because pricing is about value far more than it's about cost, first of all. And then the best bit of all, which freaks everyone out, is the way you present a price is far more important than the price itself. Nobody knows what a bag of apples is going to cost, but how I present it. What do I present it next to? What is my anchor price? And how have I framed it? It's super important. This has been to me, maybe the most enlightening journey of our startup experience, which is pricing. Right? Prices a number, pricing is much more dynamic. How you present the number? How you present the number again, what you compare it to, how you think about value not just cost. And then also, how much of it in the software business actually ends up being just dropping to your bottom line, your profitability. It's just an amazing thing once you learn it. I used to talk about Buffet, pricing power is the most important part of any business. The big quote. You have absolutely no idea what that means. Just explain more. And then you go through it yourself, then you're like, okay, now I really start to understand it. We really struggled with it. We were in new brand, introducing a new product that we're all dead never seen before, synthetic marker research. How do you price that? And another thing I think you said to us over the years that I took a long time for me to appreciate is how pricing and strategy in general is set of integrated choices, right? That's right. That's an idea that like when we decided on the price, all of a sudden, first of all, you have to choose the right category entry point because different category entry points, like I'm entering a new market or I'm launching a new product, come with different willingness to pay. All of a sudden, your targeting choices get revisited of, well, you know, if our average deal size is $60,000, you're not going to target startups for whom that's a lot of money. You're going to target enterprise brands for whom that's, you know, 10,000. The only caveat on that is there's a danger with a lot of marketers where they sort of price starts to align everything else around it. In reality, that's what you're with. Yeah, your strategy should have done that. So price like distribution and product design and comms should have been coming out of your core strategy. Sid amine, yeah, and you would have aligned it that way. But maybe let's just do a wrap of what we just talked about. There's lots of very interesting stuff in here. Number one, competition is for losers. Don't compete with Roger Federer on his terms at his game. Avoid competition, right? Avoid competition. Market orientation is a gift and it's because you are not the customer. You are the worst possible person to think like the customer. You're incapable of thinking like this. Yes. And if you start with market orientation, you very rarely go wrong. So who's going to do setting a price? Probably finance. And finance ops sales. Anyone except marketing. The minute you understand pricing, who's going to do the research, only marketing. Who's going to do the comms around the price, only marketing. That's where marketing should be playing a role in where it currently is. I like it. Marketing only thinks about its contribution to volume. It doesn't think about its contribution to pricing power and pricing it. It doesn't. It's just phenomenally more in the price. Yeah, arguably where the real value is created. Yep. So now let's move on to B2B. Many have had the pleasure of hearing Ritz and talk about marketing. Today we'd like to treat everyone to Ritz and on B2B. Nobody likes to talk about B2B. No, no, thank you. So we're sorry to have to do this to you. But going back to why you don't play Roger Federer, tennis. John and I don't try to play marketers at B2C. We try to play B2B because it's less people playing there. You've said a number of insightful things to us over the years about B2B. Let's start with the age old question of is B2B any different from B2C? Is it a false dichotomy? Yes, it is. The answer to the question is no. No, for a very simple, important reason that everybody misses. Whenever you get B2B marketers getting all precious about B2B, like you two lose us sometimes. Yeah, yeah. Asseless. B2B is this. B2B is this. And it's different from B2C. Blah, blah, blah, blah. But the minute you created that bundle of shit known as the B2C thing, you're missing everything. That generic bundle of B2C things is trombones, insurance, sex tools, banking services, boat rides, airplane tickets. It's incredibly heterogeneous. You can't say B2B is different from all of that because all of that isn't the same that. And the minute you see that, you go, oh, well, I am. The whole thing is kind of B2B isn't that uniform either. So I think what we're really saying is most businesses have different dynamics. Now, the answer is also, yes, it is different. There are a couple of things that make B2B in general different in general than B2C. I would say they're really at two, right? The first is the role of the buying committee, the plurality of people that we're often having to convince. That's to use marketing parlance. That's a giant pain in the ass. Because now we have seven people rather than half a person to worry about, right? The other thing is the emphasis on the sales force. So there's no point being a marketer in B2B if you can't have traction with the sales force. You might as well go home, right? If we really simplify it, the long and the short of it in B2B is basically the short of it is the sales force, right? I mean, there's a few things you can do with comms and tactics, but it's mostly the sales team and where they go and how they call and what they do. A marketer in B2B has to be channeled into the sales team and I saw it my whole life in farmer and medical. I would happily take a technical marketer who was six out of ten, over one who was ten out of ten, because I would also multiply that score by how intertwined and respected are they with the sales force. And a six times a five would score 30 in B2B versus a ten times a two, which would score 20, you know what I mean? So you have to be able to work with the sales team to manipulate them to a certain degree, to give you information and then respond. This is a good segue to the second conversation around B2B segmentation because there's one, we've been having this conversation, but part of it is that segmentation is generally, how does the sales team go to market? Right. And I'm going to do my segmentation in that way in B2B, but there's another world in which you just understand what are the customer needs, the category entry points, the jobs to be done. In fact, we don't segment based on kind of industry or company size, we segment based on need. There's two very different ways to do it. What's your take on the pros and cons of strengths and weaknesses of those two? I've done this a lot. I mean, I spent a lot of my career working in medical businesses, the variety of them. So it's classic B2B territory. And I know, exaggeration, I've sat in on, I've built and I've reviewed more than a thousand B2B segmentation. So I really know this shit, right? There's a couple of things you learn, 300, 400 segmentations in, right? The first thing is defining the total market is a really important thing to do. Like who's in this market we're going after? If you take John Dawes's 955 rule, which is brilliant, before we get to that, who's the 100? Yeah? Because there are some people that aren't in here for whatever reason that we might confuse ourselves about. So who's the 100 that we want to do the top of final take? The next thing is when I slice that 100 up in B2B segments, I'm only going to do it at the firm level. Okay? Lots of companies with really well meaning smart objectives in B2B go, well, you've got the CTO and you've got the buying committee and you've got the CEO and they're here and they're there. The minute you do that too early, it all turns into spaghetti. So when you do B2B segmentation, you segmented at the firm level. So there's segment B is, we looked at an example yesterday. Global multi-nation. Global multinationals, there's two and a half thousand of them. There were six billion dollars. We have a 10% market share, right? That's the firm, right? They all have similar behaviors because that's why they're in that segment. They buy in similar ways. When I write the portrait of that segment, I describe the firm and then I describe the buying committee within it because it should be similar, right? That's when I get into the CTO wants this and he's in charge of that and the CFO wants that and she's in charge of that, right? That's when I get into that kind of detail. That's an absolute precept. The final thing about B2B segmentation is you have to be able to name every single company in every single segment, right? It's like B2C, right? If I have 22% of the market in the happy ETA segment, I don't need their names and addresses, right? In B2B, I have to have the name of every single client so we can pass it on to the sales force and they can activate it. Otherwise, what's the point? You're point about a segmentation should be meaningful and actionable and the names of the accounts is what makes it actionable. Super feasible, right? Otherwise, what are we doing, right? Right. Right. So I think for me, the last thing I would say on this is in my experience, B2B segmentation is the key to most of the doors because what you end up seeing is, oh, eight literally 80% of the money is over here. And the sales force for whatever you tell me about them as good as they are, they cannot see that without the structuring insight of good segmentation. That's where the action is, that's where we should go. And I think for that reason alone, as I used to say in my marketing director's and medical companies, if you can do segmentation, you've earned your salary, you can pack it in for the rest of the year. So just to really put a very specific point, lots of our customers are interested in personas. Your view is you develop your segmentation, you then ultimately decide who you want to target and then you start to bring in the personas to bring that story to the company. The personas inside the target. Yeah, right. And as these ideal customer profiles and everything else is they evade the topic of targeting completely, right? You miss the point where you look. at the map and go, "Oh, fuck, over here is the giant opportunity we didn't see," right? So personas is not the start of this journey. Personas is what you do once you've identified a cluster of firms that have a certain amount of potential and are in a certain situation. There, I think personas are phenomenally useful. Is this an example where you think marketers, especially untrained marketers, jump from strategy to tactics too early? For a long time, we were selling pain medication into Australian hospitals. When we built the map, what we discovered was there was a tiny private segment of hospitals that there was literally, I don't know, there were 20 major accounts. And we went after those accounts because every one of the doctors that then was the head of practice in the big public hospitals did half a day a week there. Now, if you go to big public hospital, you can't get them just to change the form you're in all that. It takes like years. So this was a chance for these senior doctors to see what advanced pain medication the new stuff could do because you can prescribe much more easily in a private setting what you want. So we got it on the form of the hospitals. The surgeons were like, "Oh my God, this is a new generation of pain medication." And then the next time they had a meeting in the public hospitals, they talk about it. You only get that kind of play when you've built the whole picture and you can see the dynamics of the segments and everything else. So is your view that when you have your segments, then you have your targeting, then the positioning bit, I mean, is that just you're rotating the set of category entry points that are most relevant to your key targetable segments? I'm more Catholic in my taste, right? So we have this debate all the time. I love category entry points. I use them in my own business. I think they're a genuine revolution. I also think that positioning is much more basic. And I don't want it to be too dogmatic. Yeah, I think that's good. Don't position on, you know, attributes position on category entry points is just everybody just shut up. And as it turns out, Pete and I talk about this a lot, when you do category entry point research to be fed Aaron Boebus, it does tend to dominate what you position on because what do you know? It's an incredibly valuable concept. It's the customer need absolutely. But at the same time, many MBA, for example, two of the things we're positioned on is when you really do question the confidence in your own skills, because you haven't got a qualification, you're a director of marketing, you start questioning yourself because you've never been trained in this. That's absolutely a category entry point. That's why we get people on many MBA. They go in like, I'm ahead of marketing. I've never properly been trained. I'll bet you if I do training, I'll find out most of the stuff I already know. But I want it's one, but I want to check. That's a category where we position on that. But another one we position on is convenience, which isn't a category entry point to begin with. But we're just basically like, we're going to make it easy for you to 10 weeks of training and be busy at work. So it has to be a combination in my opinion. And it's whatever the day it says. So now we've talked about Ritz and on marketing. We've talked about Ritz and on B2B. We're going to finish with Ritz and on AI. Very exciting. The great technological revolution of our time. I'd like to start by actually just understanding your AI journey a little bit because I feel like it has been a journey. Like, what was your original impression of this technology and has that changed? Can I give a compliment before? Please, before many as you like them? I generally think of you as a lot, I think that's fair. That's that's that's that's the compliment meant by the way. That's that's my comment. Thanks, Joe. Because most things you've just dismissed as hype. Correctly. By the way, correctly. Correctly dismissed as like, I would say, AI, you correctly diagnosed as not hype and actually very valuable, but not at first. Well, I was pretty good. I mean, I was pretty good in this one. I was sort of like appalled by the generalism at the very start. But once I looked at it and we're talking two years ago, what I said was, yeah, you can see why it's different. And I also made the point John's making, which is Gary Vee is going to go crazy for this, right? Because Gary went crazy for NFTs, right? Really printing, you know, VR and AI on everything else. And I and it's a matter of record have consistently shat on all of those things. I called NFTs an idiot magnet. Yes. And they proved exactly to be that I said that the verse was stupid, made no sense. And it was. And it's so it came to pass. So when I went on record as saying, I think this is the genuine revolution in how we're going to do stuff. I deserve double points because I'm not, you know, I'm a lot. I'll give you a dope. You'll hear on the evidence about it. I'll be delighted to give you dope. You'll point. I think so. Double points. Double points. Double points. Yeah. It's good. So we reminisced about the time when we first met and he thought I was Peter. You were a lady. Peter. How about when we told you we were leaving LinkedIn? Why don't we tell that story? I think it's a good one. So we have our regular horn where we go and do a sort of carry grant number under the trains. And you broke it to me after perhaps a bottle or two of red wine that you were going to leave LinkedIn. And my reaction was your mad. You've got great salaries. They basically let you do what you want. You're fulfilled and happy. One of you had a child coming. One of you'd already had a child. I was like, you know, don't do it. And then one of you got out of phone and showed me the replicant capabilities of synthetic data. And I was like, wait, wait, I got to think about this. I got to do this because I'm pissed and I'm like, yeah, yeah, you got to do this. I remember. I remember seeing your face. I also just explained the inside joke behind your face was Christie. Totally. Yeah. She was sitting in the grass. What's wrong? They're quite. Christie. Quite. I see this synthetic data. What is it coming kind of tying this to the first impression of AI? Was it seeing the data and how real and empirical the data was and some correlations? Like was the data that informed you had you been playing with it yourself? No, no, no. I was aware of what was being said about it. But when you showed me a couple of the replications that I think Brian had already done and it was already at 80 odd percent. Yeah, right. Right. And in its most primitive form. Yeah, right. And your point has always been, this is as bad as AI will ever be. Yes. It was clear straight away. We've all had the same experience about 70% of the market as we work with. Don't get to do the marketing they want because they don't have the data for one of many different reasons. And that thought was straight away in my head. This is data that's imminently available immediately. Well, it's also going back to the original sections market orientation, which is a core Ritzzonean idea, right? Synthetic data, synthetic research gives you market orientation for very hard to reach market. I mean, it also gives you data and it gives you time. And those are two things, especially in BDB that marketers are very poor when it comes to. And so like now you have better data and more time to think about and use and apply the data to being strategic. I mean, to me, it's the greatest productivity revolution maybe in the last 100 years. I think that was apparent even on that day. And then the other thing was on that day, I remember also being aware of the fact that we could move from research, unlimited research into essentially automated AI planning. Yeah, right. Because now we could just build straight out of the data and go back and ask for more data and let the machine work. Yeah, right. I think we were already talking about that that day, right? You more than us, but we've certainly gotten there now. You were also very excited about the Quentin Qual thing. Yeah, Monte Carlo Simulation, the Monte Carlo all things that took us while. Monte Carlo was the thing that excite me the most. Remember that. It was the idea. Yeah, tell us. Well, the idea you could run 4,000s versions of a brand plan synthetically and see which one produced the best result. What combination of positions and segments, objectives. It's the Doctor Strange moment in the Marvel movies, right? The reason they defeat Thanos is because Doctor Strange runs, I think it was half a million Simulations. Yes, but there's only one where Tony Stark and Win. Yes. And that's the one he moves them towards. Yes. And it's not that an AI marketing plan is better than what a human can produce. It certainly is, but that's not the point. It's the winner of 40,000 AI marketing plans versus one that was built very poorly partially by a human. Then you go, holy shit, this is not even a fair contest. I was conscious of that. And I was conscious that at some point, chat GPT will be able to do pretty good market research from a prompt. Yes. You know, I mean, it would become in the year 2035, it'll become commodified, but not your ability to build it into strategic planning, which is why I always wanted you to do it. Right. What to actually do with it. I was coming back to the Federer point. I think if you can run 100,000 simulations and 54% of them tell you to go this plan, then that's the plan is most likely to make you successful in the market. You can do that now. To get what you Federer into this multiple times, right? Well, there's a very subtle point you've made here, which I think you've been bang on right about from day one, which is people are just making this false comparison between the perfect marketing plan and the AI marketing plan, or the perfect market research and the synthetic market research. In fact, you're actually comparing the AI marketing plan to no marketing plan or a very shitty marketing plan, which is what most marketers have, or you're comparing it to no market research at all. They just want to talk to sales and hope that represents the voice of the customer. That's the point, right? It's not, you know, yes, we're using AI to replace radiographers. Radiographers are very talented data-filled professionals. Right. We're using AI to replace something that has been done either not at all or horrendously bad with crayons. The similar one is when everyone linked in a buffoon turns up and goes, "Oh, we're on the chat GPT," and I did some synthetic market research, and it was way off, right? And I'm like, "What?" It's like me saying, you know, "So I picked up a flute on Monday, and I started playing the flute, and on Friday I can tell you the flute shit." The flute's horror, flute music sucks, man. It's like you can't base the criticism on synthetic data of you and your underpants having a quick go on chat GPT. Yeah, it's been deeply frustrating to us. I think their bigger point is that people, for some reason, they don't think AI is a skill. They think if you give everyone chat GPT, you get the same outcome. It's true. And like what Brian does, first of all, Brian doesn't even use chat GPT. He uses 10 different models at 100 different configurations, and he's figured out how to mix and match those those until now. like a symphony orchestra that perfectly plays synthetic research. And the idea that like any random person can use Chatsy-B-T and establish anything about whether it's good or bad, it's just a wild misconception. - But you get these two headlines right. Anyone can use it to do anything. And yet Zuckerberg's still paying 250 million bucks for the guys that actually do the engineering work. You know, clearly we're leaving the world of the prompt very soon. - Amazingly, we get this criticism on both ends. We get people who are like, I tried it in Chatsy-B-T and it doesn't work. - It doesn't work. - So it doesn't work. And then we get other ones being like, can't I just do this in Chatsy-B-T? And we're like, no, you can't. You can do very primitive forms. It's certainly better than nothing. But are you gonna build a model that is like perfectly calibrated to this use case? Or are you gonna create a thousand small permeations of a CFO and run the segmentation 10,000 times? Like, absolutely not. - People are not, this comes back to your point, but they can either think about how to do their market research properly, nor think about how to build a marketing plan. People are mostly not systems thinkers. And a system is required to build incredibly good market research that answers all kinds of questions and all kinds of markets for all kinds of companies and simulates it many times to smooth out the noise and increase the signal. They're not able to stitch together all these different things to build marketing plans. I mean, people are just not systems thinkers. And so when you're not a systems thinker, you're gonna make kind of incoherent claims like, it can't work and I can't make it work. - But look, I've seen that. And I've seen the other one that we've talked about a lot is this deviation from the human surface. - Yeah, let's talk about this. 'Cause this is one of the weirdest criticisms. A normal criticism is like, that's insane. A robot CFO won't tell you what a real CFO is. The weirder criticism that we sometimes get is, oh, it's only 90% the same as market research. That means it's not good because the magic, the real truth is in the 10% difference. Can we get your response to that for better? - So that's rubbish, first of all, the 10%. But I'd say a big point here, right? There isn't, we're treating the consumer answers to a survey as if they're the Archimedean truth. - Right. - They're not. The consumer responses to a survey have long been understood to be a massively infallible reflection of actual consumer perceptions and behavior that we've known about since the advent of surveys. - Yes. - And so I continue to maintain the fact that when you see, I mean, you guys are averaging, well, 90, what's your hit rate? 95% 88% sometimes above, sometimes slightly below. - Oralization is like point eight or higher. - All right, so do a correlation. So you're a point eight, point nine correlation with the human survey. - Can I estimate? - My point has always been that may not be worse than a point nine, five correlation because the human is often off and incredibly off. And I'll give you a good example. You showed me some data where the area where a synthetic data is most fallible at the moment is in surveys where consumers talk about their future preferences and behaviors. And you showed me that and you said, so we're not very good at this. And I'm like, no, because I know from 55 other experiences that the one area where human data is always off is when you ask people, I worked a lot for betting companies. How much will you bet next year? Oh, much less. Much less this year, right? Humans literally, one of the ways to train people in survey design is keep it anchored in right now. The minute you say, will you love your husband more next year? Will you lose weight next year? We enter a world of fiction. So I think what you saw with that variance where it looked like the synthetic data is way off the human is the humans are way, way off. And synthetic data is predicting with much greater accuracy. What's going to happen? This is super interesting because of course, it's a complete evaporation of the mind for people. Yeah, it's not less accurate than human resources. More accurate. But this is always what happens to any new technology. Like, you know, the example of like when the transition from radio to TV, people just did radio on TV at the start. It was just people reading like it was rated. But TV's a totally different medium. That's right. This is a totally different technology. Well, we've had that feeling right. So we're here working, I'm not here for this podcast, obviously. You're not? No, I'm not. So I'm here working on AI systems to do marketing planning. Right? We're heavily engaged in this. There is a moment that we have where we're planning how we run these analyses and we're replicating the human calculation process. And then we slip into what I'm calling like this alien move where we don't have to do that anymore. The machine can just do that shit in a freakish, machine driven way. Yes. And we can leave the bounds of mortal analysis and enter a different paradigm. And you feel this moment where you're like, oh, shit. Yeah, this is different. But this is not the same thing that most interesting thing. Because this is, ostensibly it is, we ran correlations across 20 questions or 30 questions. And here are three to five questions where the correlations are not high. We are data. Right. And so the natural responses, our data is wrong. The customer data is right. In fact, it may be the opposite is true. Or those are questions where people are very bad at predicting or-- Well, it's interesting. There's like cantar of all places. You're the one professor who shared this research with us like number one source of inaccuracy and human market research is not fraud, although that's a big source of it. It's that people are lazy. And they're just like, I don't know, B, C. Like just get me to the end. Which time? I get my out of B's gift card. So we find, like in that data we were looking at yesterday, it was a very cognitively demanding question. Like I always find this funny. I see it. Sometimes agencies will run this survey of marketers where they're like, what is your marketing budget? And what percentage is brand and demand? And I'm like, the marketer has absolutely no idea. Like you think the digital marketing manager knows the annual marketing budget? Like, John and I run a startup, 13% startup. And I couldn't tell you right now. You don't know. Exactly what the annual marketing budget is. Let alone what the split is. Although I hope it's roughly 50, 50. So these are questions that like a human does not have the cognitive ability or patience to answer. Which means of course you can't compare that to what a synthetic human who is infinitely patient and has infinite cognitive capacity what they're going to say. It's a false comparison. Even I mean, you prompted me to think about this. So I looked at I think places like Cantar and Gartner, when they look at spend, they actually look at the spend and then look at the actual spend. And they have to often do a calibration. And what a calibration means is what the people said isn't what actually happened. We do it with betting or we resist and calculation with betting. We ask someone, how much have you spent this year on the horses and other games? And then we apply like a 40% differential 'cause they're always overstating it. Undestanding it. So we're going to have to figure out and I think we're starting, the alien thing I love, it's kind of like part of the survey should be human. Probably maybe like current behavior stuff. Actually the AI is very good at modeling humans. There's a place where the humans are worse than the AI. - Yeah, and that's where they want a model. - You actually want a human AI synthetic survey. You don't want a fully-replication of humans. - I've thought about this point for the last couple of days we've been talking about it. You know, I think how we eventually calculated. If you look at asking the same question through a questionnaire of real humans and asking them the same question in question one, question 10, question 40, question 100. You'll get a sense of how much the human is deviating over time, currently actual truth, right? And then you run your synthetic data, which will obviously remain constant. So you're able to see the differential gap just in the human error, do you know what I mean? And how synthetic doesn't have it. These are the ways to prove to the marketer. Look, it's not perfect because no research prediction can be. But it's probably very quickly going to become more accurate than the human study. - I think the latent capability in these models is already there. Like we just have to figure out how to tease it out at a structure of a communicate. - It's like Michael Angelo, you know, he just freed David from the marble, right? Isn't it just like that? - I love you, Pietro. - It's a stretch. But the point is as well, it's 2025, right? - Yeah, right. - With 10 years away from where it's gonna be. - The earliest meaning. - The logarithmic iterations of this. - Well, so let's end here. I think there's a good sort of final bit of where it is all going. And obviously, as Niels Bohr, famous fan of the podcast, - And at Neil's - - Neil's - - Straight to always - - Straight to always - - He says famously hard to make predictions, especially about the future. And I know you are not a futurist and would never consider yourself a futurist. But if you had to sort of venture a guess, like where do you think this whole marketing AI thing is going? Do you feel optimistic about the marketers and the students in your mini-MVA? Do you feel pessimistic? What's your general vibes? - I don't think optimism or pessimism comes into it, right? I think what it means is we'll see a lot more automated marketing planning. Why does programmatic work? First, 'cause we have to have programmatic because there's no other way to do digital media. - Can we be insights? - But second, because marketers love a box that tells them what to do. And so I think there will be a rise in an automated marketing planning system, which people will still do their own accounts. But most of them will use software. And by the same token, I do think we will get to an automated marketing system to some degree. But it will be, again, much slower than anyone here is predicting. You and I have danced around a genetic and what's possible or what's not possible. The stuff you read on LinkedIn, it's so far off the base, right? The 30s are the decade where this will truly come into play. It's not the 20s, right? It's very early at this point. It's, you know, the true, we're currently programming in DOS. The advent of Windows and Google is still to arrive in my opinion. And this talking about prompts and all of this, right? We'll be long gone. And the big monsters of the 30s that have truly built these systems. And I don't think there'll be the current players. I do think we'll see a new arrival of brands and companies. - But what is the role of the marketer in an automated marketing planning world? Are they just there to click the button? - I think AI destroys price. process. It shuts it down like an accordion, right? Any complex process, accounting, law, where you have to go through a series of relatively generic steps with information, it just shuts it down. The bits of marketing, therefore, that are process-based, and I would very much put marketing planning in there. I put media planning in there. I think they will be closed down into an AI-driven process for any company that can do it. And that leaves you with a few outline points, which is probably still someone driving the process, probably someone still monitoring tactics, and I still believe in human creativity to also guide some of this stuff. But yeah, I do think we will have less marketers, and I do think the era of AI in the 30s will have a dramatic impact on the way we do things. The 30s can't wait. I think we'll have more marketers, not fewer marketers. Well, that's the ATM argument, you know? You have more bank tellers now. I think it's going to accelerate. It's going to change the job of a marketer quite clearly. I think it's going to eat a lot of tactics. It's going to mean there's more focus and time for strategy and planning. There'll be fewer marketers at individual firms, but there will be more firms because there will be more growth, and I think ultimately it's going to be beneficial. I mean, the entire history of these types of things is that technology is going to take jobs. Technology has been taking jobs since the industrial revolution. Spinning on you. And in all it's been doing is creative destruction, creating more jobs. And they're going to be weirder jobs and we can't imagine what they are, but I think there's going to be more marketers for more industries and many industries maybe aren't even industries today. I think to sort of bring it full circle because I do think the beauty of this long-term friendship and now partnership we have is a situation. More of a situation at this point. Situation shift, getting pretty serious. I think it's getting pretty serious. We're not seeing any other professors, by the way. I mean, well, we have. Have you explained this to your wife? I try not to. I try not to. But I think the point is AI, this incredible technology and we see a future in which it can build a marketing plan. But first you have to know what a marketing plan should look like. And if you have no proper training and you haven't taken the mini MBA or you went to University of Chicago and got taught it like it's statistics, you're not going to even know what to ask the robot to do. So there is this fusion, which I think is the original thing we were all very excited about of marketing effectiveness and the marketing efficiency of AI. And it's a very, very rare sort of once in a career time opportunity, bring those things together. And as I get older, I become more aware of those moments. Like when I launched mini MBA, it was literally the realization as I walked my dogs one day that there was no one else that had 25 years of teaching experience, but was also now free to do whatever they like because I was an adjunct professional. I was leaving. Right? There were people that wanted to be an expert lecturer that had never taught MBAs. And there are a bunch of very good MBA professors whose schools would never let them do this. Yes. And I had this sudden literally halting moment where I stopped in my steps. And when there's literally no one else in the world that could do this. Yes. And I had the same experience yesterday with you guys where it's like the most advanced synthetic data company in the world. I don't know anyone that's built more marketing plans or taught more marketing plans than me. It doesn't mean we're going to be successful, but we are in a nexus of a unique opportunity that no one else can do. And the fact that it's hard. And I think one of my takeaways from this week together is that it is very hard to build a marketing plan. That's what makes it so valuable. And that's what will make it very difficult to copy. In my mind, as you actually want to do the hard things, the very easy things like getting AI to write copy that's going to be incredibly competitive. People don't do hard things because they're hard. So that's again, a way to avoid competition. It is a way to play your game, not Roger Federer. Here we go again. Roger Federer in tennis. Get him to compete with you in a marketing plan. Thank you, Professor, for just stopping by from Australia. I know it's very convenient, conveniently located. Just jump back on the plane. We always the highlight of our career spending time with you. Just a little 24 hour off of our home. So we thank you from the bottom of our hearts. Ciao for now. Well, that's it for this week's episode of LabGrow Marketing. Make sure to join us next time for another episode on marketing effectiveness and AI intelligence. And you know, if you want to stay ahead of the two biggest revolutions in marketing, you're feel free to connect with us on LinkedIn. I'm John Lombardo, your Peter Weinberg. And if you're ready to see what LabGrow and Research and Strategy can do for your brand, please visit us, of course, at evadenza.ai. Ciao for now.