Tidings of Innovation & Cheer - AI, MVPs, and Leadership Lessons to Carry Into 2025

Speaker A: It's another episode of Practical Product Management where we talk about instead of the theory of product management of which there is a ton, we talk about the practicality of product management, how it's really done in context at our companies, um, with everyday life. So today Marilyn and I have a guest, Tony, who is here to talk to us. So I'm going to turn it over to you and just let you introduce yourself and then we're just going to talk and get to know each other a little better.

Speaker B: Great, let's do that. Uh, thanks a lot for having me. Uh, so I'm Tony, I'm a B2B product consultant. Uh, I help B2B startups from early stage to scale ups. Basically how they build product that people want to buy, not just products that are good at what they do. Uh, and aside from that I'm also an entrepreneur. I launched my own SaaS a year ago which is a prospecting tool for B2B companies and uh, for that uh, product. And in the past because I was a product manager at a software company, I started using AI like two years ago intensely. Uh, and since then I dove in very deeply and so now I do uh, product newsletter called the Product Courier that shares AI use cases and we're going to talk about that practical side of it uh, every week for product teams. And with that I do coachings for, for product leaders and teaching sessions or workshops with product teams to teach them about how to use AI in the day to day work.

Speaker A: Amazing.

Speaker C: I love it. Tony, uh, so you actually have a pretty interesting career. Um, uh, and you have been at this for quite some time and in fact I have, I have so many questions for you but I'm going to start with like how did you land here?

Speaker B: Here? Well, product.

Speaker C: How did you land in product? But how did you also decide to sort of take this approach where you're helping lots of people in sort of a consultancy role instead of being sort of like really hands in, uh, in one thing.

Speaker B: Cool. So let's travel a while back very quickly. So basically I uh, always like technology, uh, growing up. But life has its ways. And when I was a teenager I had to travel to Portugal to follow my family, uh, and change the course of actions that I was, that was taking, uh, came back from, to France a few years later alone. And then instead of going into tech which was the initial plan, I went into economics, uh, because again for life, uh, for reasons of life. Then I decided I wanted to work in economics a lot but ended up doing finance which Was not my cup of tea. And I got an opportunity to work in the music business, uh, to basically be a booker or product tool manager for companies. So I ended up producing jazz tools around the world with jazz artists from people that worked with Mice Davis and Django Reinhardt, great musicians. And I did that as an employee and as an entrepreneur for five years. Then I wanted to go back into the tech part in the economics world and the banking business which you know pretty well. Uh, but here in France when you go into the uh, banking world you have to start with sales. Uh, because they consider that everything that is not sales is just like a bonus that you get as a reward for doing sales unless you have lots of degree which at the time I didn't have. Uh, so I went into a ah, French bank as customer support. While working as customer support I passed a degree in economics of innovation and corporate finance and then three years later became an account executive for big companies, uh, with revenues between 3 million and a few billion uh, euros, uh, uh, and then got an opportunity to become a product leader in another French bank which I did for three years. Uh, but things are pretty slow in big corporations as you know. So I wanted really to go uh, into, into tech. And while I was working as a product leader I passed a degree in uh, data analytics and data science because I realized that there was a lot of data in the banking world that was not used and I felt that was something to, to do with it. And which takes us also to the AI side because machine learning and stuff like that is AI, although people forget about it nowadays. And then got the opportunity to go in a fre, uh startup that's called lemist, which is a, an automation tool for prospecting that's actually 75% of the clients are in the US so pretty international company. And while being there and while I was hands on, I wanted to become an entrepreneur and launch my own SaaS, which I did. Uh, after a year at LEM List I left to do that. But aside from that and until it becomes Amazon or a big, big software uh, company, obviously you have to make a living. And since I love teaching and coaching stuff, I ended up helping entrepreneurs and realizing that with all my experience in B2B and product over the years, uh, what I had to share was useful for people, uh, to help them launch their own products. So that's why I do teachings uh, and workshops at incubators. Nowadays I do coachings, paid coachings for, for startup uh, entrepreneurs and I help entrepreneurs basically want to launch products or to grow their product market fit, uh, with their work in B2B.

Speaker C: I love it. I love it. And I noticed some like, so many questions. We're going to be here all day, people, so just buckle up. I think one of the things you said that really resonates with me is people think like, AI is new, um, because they just started hearing about open air, they started hearing about Gen AI, but really machine learning and what I would just say, you know, applied data science has been around for a long time. Um, and so is that, is that something you need to educate people on? Like, hey, this isn't new. Maybe the generative part is new. So how does that play into what you do now?

Speaker B: Yeah, that in, uh, the coaching part and with the product teams, namely also sales teams that I can teach. Uh, that's one of the things I explain is that AI is not new. It's been around for years. The, actually the vast majority of what you can do with AI is not what you see with the ChatGPT and Gen AI. It's what's underneath, analyzing lots of data. And actually genai is just a way to communicate with standard AI models that do deep learning of, uh, technology of data, uh, how you can cluster stuff. And so, yeah, part of my job when I do teaching is explain that Gen AI is not new. It's been around for years. And actually what you can do with AI goes way beyond what you can do with Gen AI. And most of the time, actually we're going to talk about that more. Is that 75% of the time when people come to me saying, hey, we want to launch an AI product, can, uh, you help us? The answer is no, you shouldn't not use generative AI AI. You should use something else. Because there's a better solution than that. It's just the shiny object. Uh, and if you want to use AI, here are the ways to use it than just having a chatgpt chatbot when.

Speaker A: So when people come to you with those, those requests like, hey, we want to do this, what do you think they're hoping to get that they've seen out in the world that they're like, oh, I want some of that. What do you think that they're hoping to get that you're like, well, hold on.

Speaker B: Well, yeah, and actually for me it's bad because it obviously takes a lot of work for. Away from me when we just do one session and say, okay, well, let's not work together.

Speaker A: You don't have to tell all your secrets.

Speaker B: Yeah, but, uh, but the thing is, yeah, most of the time what I expect is the shiny object and the FOMO syndrome. And I've witnessed it. I didn't talk about it, but I was a branch manager in a French bank for startups before I joined Lemlist and actually saw a lot of startups. And at the time it was NFTS and uh, blockchain stuff that was trending and I saw a multiverse also. And also many people just wanted to go there because it was the training thing and because they thought that it was the magical solution to have a growing business, which obviously was just temporary most of the time. And once the VC money stopped in 2022, many companies crashed, uh, because it was just. They were trying to show a solution for problems that don't exist.

Speaker C: Yes. Uh, I think you just hit on something that's very close to my heart is like, what is the problem we're trying to solve? And the Metaverse is the one that made me laugh, like the hardest. Like, I mean, it's cool. I'm not sure we've gotten down to like practical application beyond sort of like, you know, can you find your tribe? Is it just a. Is it just a different Facebook or MySpace or whatever? Like, whatever. Um, yeah.

Speaker A: It's funny you say that. Like, I think it doesn't feel terribly different to me than, you know, 10, 15 years ago when everybody was building a platform. We all need microservices and platforms. The difference is that it has more, it's more known in the rest of the world. It's not just a little tech, a bunch of tech junkies sitting around going, we need a platform. Right. Like, no, but you know, there was a point at which we. I remember saying often that's just because it has a back end doesn't mean you have a platform.

Speaker C: Yes.

Speaker A: I just remember having those conversations being like, you don't have a platform, you don't need a platform team. You don't have a platform, you have a database that you attach a front end to. So congrats. Right.

Speaker B: But I think just putting the name.

Speaker A: Yeah, I think this is different because it has, it's sort of. Now non tech people are engaging with the concept like the metaverse and NFTs and this. There's a different user that is touching it who doesn't understand that this is what we've been doing for centuries.

Speaker B: Yeah. But there's actually two things about that specific with AI and uh, I totally agree with what you say, is that, uh, the first thing is that AI compared to other stuff. Again, I worked in the past and I did my essay, uh, on bitcoin, it was in 2016, so it was a while back. But unfortunately and didn't buy bitcoin, uh, at the time, which was the dumbest decision in the world. But still, but at the time I remember people starting to say, hey, bitcoin is the future, blockchain is the future. And there are useful applications that can be tracked and used. But obviously, yeah, ah, it becomes a shiny object. But the difference now with AI is that because it's more like a way you communicate with computers, uh, and, and software, uh, and machine learning etc, it becomes accessible for much more people and people can understand actually what it means. I mean you put someone in front of ChatGPT, just ask a question, you get an answer, it's easier to understand that what is a platform, what is an nft, what is a blockchain? So that's the thing. So the implications are obviously much larger than ah, even traditional AI when you have to explain what clustering is, someone who doesn't understand what machine learning is. So that's why to me there's a huge paradigm uh, shift that is coming with AI. But still the limits of it is that I remind people all the time that there are 200 million users, uh, active users on ChatGPT. There's 5 billion people that have access to the Internet. So that's 96% of the people in the world that don't use ChatGPT and even less obviously use uh, AI. They may have heard it in the news, okay, but they still don't know, don't use it. And so there's a lot of stuff to do. But again, if you want to do something that's going to be sustainable, you have to find a problem first and then find a solution. And if AI is the right solution for the problem you're trying to solve.

Speaker C: Yeah, I was going to say, I think, I think the number of people that have experienced some form of AI probably skews more towards the number of people on the Internet than people that have used the ChatGPT interface. Most people that have interacted with AI don't realize it.

Speaker B: Right, yeah. Any segmentation that you have, ah, when you go on the bank, website or whatever institution that has big data, uh, behind it, there's a new use of AI and you don't know about it because you don't realize it. But being a segment and having uh, a personalized experience on whatever website is the result of Segmenting and often, oftentimes

Speaker A: AI segmentation or even if you even travel. Right. So everything travel industry is all about how you train the M models to find the user and say, hey, would you like, do you want to go here now? This is what you've done before. Like travel has been doing this sort of segmentation and using machine learning and training the data around their users and different user bases for a long time.

Speaker B: Yeah.

Speaker A: And so people have experienced that. They just don't know that's what they were experiencing. They just thought that Expedia really knew them well. And it's like they don't. Right. You're not that important.

Speaker B: You're not that important. Yeah, exactly. Same thing with nlp, you know, natural, uh, language processing. I mean, it's been around for years and nowadays it's just that AI does an extra layer and simplifies it, but it's been around for years and decades.

Speaker A: It's funny, someone recently was telling me that they went to ChatGPT and said, you know, tell me what you know about me, because they've been using it now for over a year. And, and they were, they were stunned. And I was all, uh, by the way, Amazon knows more than that. Right? Amazon knows more than that about you already. So does Google. Right. Like so, uh, carry on.

Speaker C: Right.

Speaker A: That's just the part you told it right.

Speaker B: Exactly. Yeah. It's like a batch ring.

Speaker C: I think it's also really funny what but people don't realize they leave out in the universe as digital exhaust.

Speaker A: Yeah.

Speaker C: I mean, you know, uh, every website you go to will cook you. People follow you around. Like, my car probably knows more about me than a lot of things. Right. Because it knows where I go. It knows all my habits. Actually, I've got like. So one of the fascinating things about what you talk about, Tony, is that you talk about helping founders get to product market fit in weeks. Um, let's talk about the complexity of product management just for a second. And I think this is sort of the thing that most companies chase. And if you've got a lot of money, you can waste time and not like sort of get that product market fit right away. Because you can burn cash or you can lower your margins when you're tiny. You don't have that time. Uh, you don't have the time, you have the margins. And so let's talk a little bit about how you help people find product market fit in weeks versus months or years even.

Speaker B: Sure. Uh, that's a good thing. And again, that, that sentence is Obviously a tagline, but it applies to early stage companies or first time founders most of the time. Uh, and uh, and the thing is it's having a very practical approach to product management. Uh, but not only product management as we have like, yeah, like, like we have today in many companies which is actually I've seen the product management scene grow over time and now you have product managers, product marketers, uh, product owners, product uh, product ops, ET is great but you need a structure to allow a company to have all those people and for it to work you need a great communication. And I realized that in, even in great startups there's a lot of silos that uh, occur but in a smaller company when you're early, the product manager actually should be the product marketer as well, product growth, etc. And my job is actually that because I've launched uh, my own company, I've worked in sales for a long time, is to have that very practical approach to say okay, here's how you can use your product. Let's define what's the priority to make money out of your product really fast and uh, have a plan that's going to help you uh, get that money very very fast. And also use the product and build product loops, product uh, feedback loops with the users to iterate very fast. And once you've launched the product make some money and see what works, what doesn't to improve and keep improving your product. And actually one of the big things I do is about mvp, uh, building uh, which is changing most of the mindsets that I've seen in founders. Even people have done great, uh, uh, have the great degrees etc or incubators is to change the mindset from MVP is a prototype to MVP is a full product that you actually, if you want you could sell tens of thousands of it and people should buy it. And the problem is that most of the time what I realize is people and founders think of the MVP like it's a prototype. We're gonna have people test it but don't forget it's not finished. So don't worry, we got to build the finished version. And so that creates a bias because if you go to someone and say hey, here's something for you to try but don't worry, it's not finished. Obviously they're going to be very nice to you. It's like the mom test. They're going to be nice, they're not going to say it's crap. But if you go to someone and say here's the, is a Product that solves one problem with one feature that you might have, are you willing to pay? And here's the price, that's the best way to find product market read faster, which is I give, I build the MVP as fast as possible, put it in the hands of the end user, have people buy it or people want to buy it, and if it works, then I expand on it and start working, uh, uh, more and more to expand it. Which is the big difference that I see even from big startups or to early stage, uh, that we have in traditional approach.

Speaker A: Uh, so far, I mean I think we lost our way. Right, because that was always the way it was supposed to be. But I think the way that VC's fund and the way that founders run is proofs of concept, not, not MVPs. Exactly right. And so then you get a POC out there that you can show people and you can get more money raised and you can say, isn't this exciting? This is what we're going to do instead of this is what we've done. Would you like to buy some and we'll make it even better. Yeah, but I think we've lost our way.

Speaker C: Do you think that's because there's too much money in the system and people don't have to make money right now?

Speaker B: I think there are two things and actually it's funny what you say about VCs because obviously in the bank industry for long, uh, I've said this, uh, with another on other podcast which is that we often forget that the VC funding in the startup world is the anomaly in the world. I mean if you launch uh, a shoe shop, uh, any kind of business you don't go to, people say, hey, give me money, give me some money. Then in the future we'll run. No, you can have family, family, friends, etc. Okay. But to make a business you have to show profit and then ask for lending, ask for whatever, but not give me the money first and then we'll, we'll talk uh, and see if we can make money, uh, from the companies I've accompanied, uh, when I was a banker, the ones, the startups that succeeded by raising funds were the ones that from the start, uh, had a sustainable business model and a way to make money. Whereas the ones that say, hey, I have a great idea, give me a million dollars, which I've seen and witnessed and didn't have any sustainable way to make money if the money stopped, then they went down. So yeah, I think that's the problem. VC money is one of the problems that led to this situation. But also, and that's more the human factor that we are in the product management world. So to us those things seem obvious. But the MVP approach as you described it, and as I said, is not natural for people because I see many founders and I work with them and you want to bake something perfect. So I'm building a product. Obviously I want it to be perfect and like I'm building a CRM. Of course I want it to be as good as Salesforce or OP Spot, but I'm gonna spend two years working on it and then people won't buy it because it's not solving a specific problem, although I thought it was. And that's, but that's human. And that's something I've witnessed a lot, uh, even in repeat founders. Uh, I talked with the CTO the other day from uh, a successful uh, startup in Israel and he was building stuff on the side and he said, hey, last time I spent six months developing something and he work, uh, so this time I'm coming to you for help. What's the first thing I should do? I have an idea for, I think it was bus traveling. And I said, okay. Are you in that business? No. Have you talked to uh, bus traveling companies? No. But that's a great idea. You're right, I should do that. And it seems obvious to us, but it's not because people have ideas and say this is the best idea in the world, I should pursue it and then see if people want to buy

Speaker C: and have my solution.

Speaker A: Yeah, yeah. Well, we get precious. I mean what you see with founders and I think this part of the vc, the VC cycle. Right. Yeah. Founders have an idea, they're very precious about it. They bring it to the world and say, we have this idea, don't you want to give us some money? And then they end up in a loop of trying to, to answer to their investors instead of answering to their customer. I remember sharing that, you know, products need to be self funding fairly quickly. And a room full of people looked at me like, I'm sorry, what? Like why? And I'm all, what? Of course they have to be self funding. Right. They need to pay for themselves pretty quickly or you're just going to be in a bad cycle. Yeah. And you never get to, you never get to focus on what your customer really wants and what's. What will serve the business because you're going to spend your time serving investors.

Speaker B: Yeah, that's the problem. And I, uh, think that the many companies from, from a few years back fell into the trap of. Because people are going to tell you that VC funding, obviously startups are different than usual businesses because you have network effects, et cetera. And people fell for the, for the illusion of Amazon. I say Amazon lost money for so many years but look where they are now. But what they forget is that what made money for Amazon was Amazon Web Services which was not the main business at first. And people fall for that trap and all. I think that VC funding is different but in the end, I mean yeah, network effects, uh, already uh, apply, etc. Right. But you still need to have a sustainable business if at some point you want to make money uh, without needing the VC funding. And I think that's, yeah that's uh, some path we went down that was not the best. And when 2022, when the money ran short, yeah VCs realized that maybe it was time to do something else also.

Speaker C: I think it was. So when I lived in San Francisco it was pretty widely held belief that most of the new startups because they had so much VC money did things at uh, a loss. And so while the money was free, because money was, money was pretty cheap there for a while. While the money was free you could actually get really great deals on lunch rides. Like things were abnormally cheap because the company was never going to make money in the current, in the current construct. And so people would hop from unicorn or startup to startup to get the free or cheap thing until they had to start adjusting their cost base and then everybody would abandon.

Speaker A: Yeah.

Speaker C: And you know, uh, it would like be the common joke. Oh this is, there's this new app where we can get like cheap lunch. Thanks VC founders. Um, and I just, it was pretty blatant at the time and I just, I'm consistently surprisedly back to your point. Like these things need to be self funding. Like yeah, you need to stand and walk on your own as quickly as possible. You might be a baby bird.

Speaker A: Right. You gotta go, well you shouldn't still be having a conversation about unit economics and profitability at series D or E or F. Yeah, like I'm like series C is for me. Like if you raise series C you better have a plan. Yeah. That's going to take you to profitability and good unit economics. Otherwise throw it, throw it away and start over. Do something else because you're not getting there. You've been generally that means you've been doing it for five, six more than that years and you have not been profitable. So stop. Because now your beast is bigger than what your offer is.

Speaker C: Yeah, but it's so.

Speaker B: Yeah, yeah, it's my baby. Exactly. And that, that was particularly true what you said, Marilyn, in a delivery, the food delivery business, they kept raising money and then at some point started buying each other for amazing valuations, like billions. Now, uh, I think here in Europe there's just one or two that, that are left and that, that obviously increased their prices and changed the way that their business model trying to work more with companies instead of individuals. Because at some point it's not sustainable to have that VC money forever just by lowering the price and having a commission of 0.001 cent on each, on each ride, so.

Speaker A: Well, those guys all lucked out though. A pandemic hit right before they were going to all blow up anyway.

Speaker B: Yes.

Speaker C: Right.

Speaker A: So they all like, they all got a couple more years of, of wind. Right. And it's like, come on. But you know, I think when we go back to the Amazon example, I don't, you know, uh, Marilyn and I have, have opinions on Amazon, obviously. Like the first thing that I think of is we worked in a business that did not build an mvp, that was trying to build the best thing before it went out the door and it failed hardcore. Like it's, it's our best favorite failure that we've ever, either one had. Right. And then I would also say at times, not all the time, but at times you have to. You know, I'm not. Today, I'm not a big Jeff Bezos fan because he went to Twitter and congratulated an orange monster. But I will say I think that along the way, often what he and his team were trying to do was change an industry. So they were willing to take a loss. So we're going to change how publishing works, so we will take a loss on Kindle for years. Right. And so, and to your point, then they had some real nice wins. Prime aws, like they had some, they had some big wins. But what do you think?

Speaker C: Naivety either. I think you remember even, even at, even at concept, you knew your economics.

Speaker A: Yes.

Speaker C: All the way through. And there was this notion of reinvestment. It's like Amazon's one thing.

Speaker A: Right.

Speaker C: Um, even in payment processing, we knew the cost, we were driving down cost. You, you knew the economics of what you were doing and you, you, you, you strove to offset, uh, anything that was a pure cosplay with something that drove free cash flow, whether it was gift cards or foreign currency exchange or something like there was no naivety in any of this. So people are like oh, Amazon lost money. Amazon chose to reinvest.

Speaker A: Yeah.

Speaker C: Every dollar they could. Because they want, they had, they had a vision on how they wanted to grow and what they wanted to sort of like take over.

Speaker A: And we were squeezing margins like.

Speaker B: Yeah. But they knew, they knew why.

Speaker A: So that we could, you know, let's buy some drones and see what they do. Let's do you know, like we were doing those other things for that reason. Right. And building really high efficiency ordering systems. These systems work like clockwork. Right. And they. Because we're reinvesting all that money into all these crazy things.

Speaker B: Exactly. And uh. And as you said, I mean. And that's something that we. As you said Marilyn, that's something. The lack, uh. There's a lack of it in uh, A lot of founders nowadays is to be stubborn on the vision, flexible on the details. And Bezos was stubborn on the vision but again he adapted it changed, improved. And as you said they knew where they were going and when they, they made the one click uh, button it was because they wanted to improve the conversion and having low margins obviously with bigger conversions you still increase the revenue etc. Etc. And everything was made to the golden we. And because there was a platform that wanted to be the number one in the world at first for publishing and then for every everyday things they had to uh, attain a critical mass. But that example is very specific to one case in particular. I mean nowadays it's better to be the, the only one in a small blue pond than to be one among. In a red ocean.

Speaker A: Yeah.

Speaker C: 100.

Speaker B: Which is the big difference from that time is now revolutionizing the world and being an Amazon or uh, now an open AI or whatever. It happens once in a, in a decade or.

Speaker A: Yeah. But they.

Speaker C: That's.

Speaker A: I think that's where sometimes we get. I think this is where the VC cycles have also fed this idea into founders minds that they are the next, next whoever they ask. And you're not. Right. Let me just be clear. I mean I worked for a guy who was like. I was. He really wanted to be Steve Jobs. And I was like that's precious.

Speaker B: Yeah.

Speaker A: You're not nice. Right.

Speaker B: Yeah.

Speaker A: Like he didn't even have a business that was going to do that. Like it was a different kind of business in the payment space that is not as shiny as this phone.

Speaker B: Yeah.

Speaker A: Right. Like sorry.

Speaker C: Right.

Speaker A: Doesn't do the same kind of thing.

Speaker B: Yeah.

Speaker C: So I think the comparators hurt us rather than sort of like leaning into your space and Your problem and your belief. You like the number of times I heard, well, this is Uber but for pets or this is Uber but stop it. Like you're just, you just took someone else's solution to a unique problem and you're trying to peanut butter it everywhere to get a cheap and quick win without really thinking about your problem.

Speaker A: Right.

Speaker C: And what people want to pay for it and what's. What problem does it solve for people? And if you really went back to that sort of fundamental, um, you might be something different and better.

Speaker A: Yeah, yeah, yeah.

Speaker B: And you might have something in your hands. You just don't realize it because you might find out, found. You might have found a real problem to solve. You're just being lazy and using a regular, uh, solution that everyone already, uh, found to a problem that might be worth solving in the first place. Also sometimes not worth solving at all, obviously.

Speaker A: Yeah.

Speaker B: Which is like the difference within a vitamin and a painkiller.

Speaker A: Yeah.

Speaker B: That we use in the B2B in the B2B market.

Speaker A: There's the name of this episode, the vitamin versus the painkiller. Just named it. Good job. I usually put things on my own. That was great. So I want to talk a little bit about fear. So. And fear and AI. We talked about this. Marilyn and I had a chat about this on our own recently. Um, and, but I, and one of the things that I think was interesting about you and your background and when we started talking about you joining us was you see these fears kind of, and you hear them. Um, so talk to us a little bit about how afraid are. How afraid are they? How afraid could they be?

Speaker B: Well, yeah, they're two, they're two major kind of people which are the ones that are obviously afraid to lose their job or afraid it might replace them. And the other kind that is blind to AI and just thinks that AI solves everything and that they can use AI for anything, even sharing a confidential data, uh, on ChatGPT. So they're the two, yeah. Two extremes. But about the fear of AI. Yeah, many product, uh, leaders, product managers, uh, are afraid that it's going to replace them. Uh, to me, my opinion is uh, beyond that thing that we read everywhere, that is, uh, AI won't replace product manager. Uh, product managers that know how to use AI will replace product managers. I think it's wrong too. To me, AI will replace bad product managers. Because nowadays in many companies what you see is that you have product managers that are basically just doing P.O. jobs, uh, writing notion documents all day long user stories, writing reports Doing stakeholder, uh, documents, which is okay, but it's not to me, it's not what makes a good pm. So those people can be replaced. Why? Because you fat, you feed AI, uh, all the data and it's going to write, uh, like you even better. It's going to find, uh, connections that you didn't, didn't even think of. So that's going to replace bad people. Uh, but the good ones will have more time because they won't be using their time to do documents and analysis and basic stuff to do what matters most, which is aligning people, talking to users, uh, going beyond, uh, what a user can say, uh, analyzing the sentiment and the feeling when they're talking to people, finding, uh, creative ways, uh, to, to launch new products, uh, thinking of product sense and all those things. So to me, that the fear of being replaced by AI should be a real fear only for bad product managers.

Speaker A: Yeah, yeah, I think that's, I think that's fair. And I think that's kind of what we, where we came to as well. Like, I want to use, you know, I think, I think everybody who does tech really wants to use AI to take away the parts of their job that they're like, I don't really want to do that. Uh, I don't love to write user stories. I mean, I know that we're supposed to, but I don't. Right. So I would be happy to have something else write my user stories and make them beautiful and have the best acceptance criteria and hand them over to the engineers and I'll say, awesome, let's go. Right? But I don't want, I would, you know, I don't want something else to do. The part of the job that is the human part of the job, the part that is understanding what we're trying to solve for what we're trying to do. So I don't know. Marilyn, what do you think?

Speaker C: No, no, no, I'm a thousand percent with you. I think, uh, I think people do get wrapped around this notion of like,

Speaker A: I control the backlog, I get to

Speaker C: write the stories, I get to move things up and down. And that's actually like, how much value is. I mean, like, when you think about the, like when you think about the value that a human can add, which is the sort of like critical thinking and same like gathering information from lots of spaces and looking at other people in context to try to understand what are they feeling, how are they acting. Like those are the moments. Those are the moments where like, could I build something that would solve that person's problem in this context. And is it a problem that's so worth solving that they would pay me for it? Um, that's so much more fun than. But I think that, I think that when you're in this like, I think it's like this fixed versus not fixed mindset. When you're in a fixed mindset and you feel like something's going to be taken away from you, you just want to grab the thing that is tangible and control the crap out of it because then you're valuable. I'm super valuable. Look at what I do in this process. Like, like, uh, I cannot wait for, you know, if you're listening, chat. Btgbt I cannot wait for our, our, our AI overlords to take some of this crap work away from us. Like it's. And then like, give me perspective, right? I will, I will look at a person in context and try to understand a problem and then be a consultant in those moments to be like, hey, these ideas are great, but did you think about these other things too? Help me see connections, uh, across industries and across ecosystems and even like way outside of what we do as humans. But like, you know, I think there was, there was this, um, there was this book that talked about when we were, we were trying to figure out how to air condition buildings with determines like how do you look at completely different scenarios to see if there's a pattern that could potentially unlock a solution. That's far more interesting to me than writing a stupid document or writing user stories. Sorry to everybody.

Speaker B: Yeah, but it's true. And again, this is the two things I think that relate to that is first, because product management is not as tangible as uh, a developer or a designer. Meaning that as a coder, here's the code. As a designer is the design. As a product manager, uh, it's less tangible. So some people grab on to what they can, which is, hey, look, I wrote a notion document, I wrote a big prd, whatever. So that's one of the things. And the other thing is more like human in general, which is, is the, the fear of change. Obviously we are afraid to change. And one of the things, and actually that's something that I tell to founders most of the time is that when they want to launch something, the biggest competition is not other companies, it is the status quo because we don't like to change. So obviously, yeah, I prefer to stay where I am because I can control. I understand, I know everything. Well then having something that's going to change my My ways and having chat, uh, GPT write stuff for me. But then what am I going to do? Because that I m was paid to do this job, so I need to justify my salary by doing something else. So, so many people. That's scary. But from what you said, for example, uh, I already does most of those things for me, I mean, when I go, when I meet a new client, a new company, whatever field they are, uh, using AI, I get a full, uh, competitive analysis and a market research within like 10 minutes with the key, uh, actors in the field, the details of the market, the numbers, uh, the trends, etc, uh, in 10 minutes. Whereas in the past it would take me hours to do so. And that for me that's just something I just need to read that and then I can focus on talking with uh, the founder about the strategy because I've got that information even before a first call. I can then on the call go beyond and deeper into stuff which I wouldn't do without AI because I would take hours just to get all that information in the first place.

Speaker C: Okay, well, I think this is a perfect place to do two things. Um, the first is let's talk about resources for people that are with us, that are fearless, um, and want to lean in and don't want to be afraid and don't want to be stuck. Um, and so I want you to talk about your newsletter. But first, you just talked about a really practical application of using machines to free you from mundane work. I, I love this. Um, I'll tell you, I'm not the most technical person on the planet. I also experience fear. Um, one of the things I did is just, you know, like, what is that? Like, there's a quote by someone famous like, I don't know, Walt Disney or someone that's like the, the easiest way to get going is to just start. Um, and so I joined Reddit, I joined all the chat GPT communities. I copy and paste the prompts. I know, like, yeah, yeah, like there's so much stuff out there that you can just try in a really sort of low friction, not scary way. But Tony, I know you also have this newsletter. While you pull a lot of that stuff together, do you want to talk to us? About what? Like, where would you send people for resources?

Speaker B: Sure. Uh, well, first of all, subscribe to the newsletter. Uh, and uh, but uh, no kidding aside. Yeah, actually it's the motto of the newsletter, which is one actionable use case each week. About AI. It's not about the trends, it's not about the Frameworks, it's not about those theoretical stuff because you can learn that on your own. And actually I have a list that uh, we can share with the audience, which is a list that I've curated of free AI courses that anyone can take on any topic, can be introduction, can be more technical, whatever you want and you can learn for free just if you want. But practical applications is for people. Uh, again, the easiest way is just where you should go. You should go to chatgpt.com and start writing. That's it. Say, think of a problem that you have or something that you want to ask. Just ask and then see where it goes from there. Uh, that's the, the easiest way. You're talking about prompts. Yeah, we have a collection of prompts that uh, that follow private people as well. But to me, uh, again the prompts is like having a recipe. But if you don't know how to cut a tomato, you can have the best recipe in the world. You won't end up doing a lot of stuff with it. So you have to be careful with that. But prompts is a great way to start, uh, pre made prompts because it gives you an idea of the structure and uh, ideas of stuff to ask AI. But then once you've done that, it's just to understand how AI works and keep in mind that it's conversational. Uh, AI, which means that there's a sentence in outreach that says, uh, money is in the follow up. Well with AI it's the same thing. It's, it's not the first question that you ask because you get gonna, you're gonna get a probabilistic and statistical answer, which is the most generic answer people might give. So it gives you an idea, but then you have to keep talking to it to go deeper. And that's the, the best way for me, the best resource is not to go to my website, download my Framework ebook, whatever. No, it's just go on the website of ChatGPT and start asking stuff. Uh, start asking, for example, uh, I did an interview with the user. He talked about this and that. What do you think? Or for example, one of the things I did a, uh, use case that I use is uh, upload the MOM test book, um, to chatgpt and say, okay, now I'm gonna see a user in this industry and the goal of the interview is to ask them about this product, write me a list of questions that are relevant based on the MOM test. So that way it gives you a list of questions that are unbiased that you can ask instead of what if someday no it says the last time you did this, how did it go, etc. So to me that's a practical example and you can apply that to any uh, structure of your, of your workflow as a PM. The best thing if you want something structured because PMs like written stuff and frameworks is that map your day, map your day in terms of tasks and then for each task go to ChatGPT and ask a question related to it. Then see how uh, it goes from there.

Speaker C: I like that too. I'm, I mean like yes, yes, yes to everything you said. And I will tell you, I've even because I'm again not an engineer, kind of a dummy. Um, I've even told Chat GPT to be a prompt engineer and help me write a prompt, um, and then and ask refining questions to make the prompt better. So I got it and got myself into a loop where I actually ended up with a pretty good chief of staff.

Speaker A: Yeah, nice.

Speaker B: Yeah, but, and it is. And again yeah, the, the people that thrive on the, on the chat GPT and AI are the people that can find use cases that are relevant to us. Because what you say there are people selling uh, applications, uh, of software that do that, like do software prompt engineering for you. Like you put a question and actually behind the scenes it's just what you did. It asks ChatGPT back and forth uh, to, to write a prompt. So, so there are people selling it. But I think yeah, that's the best way I use it for example also for uh, freelancing work. Like uh, I do a call with a prospect. Okay, analyze the call, tell me what was right, what went wrong, what, what can I improve for anything. I wrote a quote. Is it enough? Is it too much? How would you rephrase it? I mean you can use it for I, I share actually with the PMs that I coach, uh, how to use AI for user interviews and for finding a job which is refine your, your resume, score your resume based on the job description you're applying to. And now with a new voice stuff you can have AI play role play with you like the, the interviewer, uh, being a recruiter, being the CPO of the company. You just give AI the information about the company, the job description and your own resume and say now you, you are the CPO of the company. I want to join. Let's play, let's role play and ask me the questions you would in an interview so that people can prepare.

Speaker A: I love that No, I think it's so important. And I, I mean, I'm with you. Like, I, I think. I mean, I didn't know anything about it until I started playing with it. And then now it knows me well enough that I'm like, no, no, no, not like that. Like this. It all, like, have this conversation with it and it's like, oh, yeah, okay, let me, Let me try again. And it, yeah, well, it'll adjust. And it knows how I want to sound, right. Based on things that I've fed into it that I've written. Right. And so it's like, oh, okay, she's. She wants a comic. She wants it to sound like this, right? And so. But I love that. And I love, you know, I'll be like, no, a little, little funnier, a little less funny.

Speaker B: Bit less like me this time.

Speaker A: Um, more like me. A little less like me. Right.

Speaker C: I'm not going to lie. I think the last thing I wrote to it was like, can you, can you rephrase this so it's less bitchy?

Speaker A: Could you make this seem a little like a nice. Yeah. Yeah. So, uh, the last question I have for you, because I think we're. I knew we would. I knew we would talk for a long time. We said 30 minutes, but I knew it would be closer to 50. But what are you excited about? Like, given everything that's going on and what you're kind of where your hands are, what are you excited about?

Speaker B: Well, in terms of. If we think about AI, is the future of AI agents, which is based on basically AI and link and doing all that stuff we just described on its own, uh, to develop that AI with image. There's a lot of potential as well, uh, because there's a lot of amazing stuff. Uh, and it's also a big concern about deep fakes, about where, whether. Where is this going in terms of fraud. Because we see nowadays there was an article this week about that, uh, about the risk fraud and the biggest increase in fraud because of that. But, um, yeah, I'm excited to see how we can take AI to do all that boring stuff on its own and, uh, for us and give us time to do other stuff. Uh, and also, uh, I'm excited because there are so many people, companies that are launching nowadays because of AI, because it's so much easier than before to build something. Uh, and actually that's my focus now is not just on building products, but more distributing products because, uh, I've done a lot of marketing and sales and I see that nowadays best Known product bits, best product every time. You know, that's what they say. And with AI and the growth of AI, there's a lot more of that coming. And to me it's good. It's good because for my business, uh, in what I like to do, which is help founders put their product in the hands of the end user faster. And so there's a lot of stuff going on there because again, building a product, then you see that on Twitter every day people can build, uh, apps. I mean, even I, and I'm not a tech guy, I built apps in a day or two. Uh, I. Yesterday I did a scraping, uh, script to script the list of founders on YC. Uh, it took me 30 minutes to write with ChatGPT, and then it ran in the back, in the back end and, uh, did it all on its own and got me a thousand contacts, so I didn't have to do a lot. But so that part is easy. Now what becomes difficult is how, how am I going to distribute that and be unique and show my uniqueness? And that's also where we can thrive based on our expertise with product. And all that stuff is okay. Help people find what makes them unique, uh, what problem they're trying to solve, what solution they're bringing to the table. And that from what I've seen, and even with AI, people still don't know how to do it. And AI can't figure it out for you. It's going to figure out basic problems and basic solutions. Uh, but you have to think. And that's where we come in. And so that's cool. And that's what I'm excited about, is that they're going to be many founders out there and so many people that need help because they're going to all be, uh, questioning themselves, oh, I built something, it's not selling. Why is that? And so that's where we come.

Speaker A: Yeah.

Speaker C: So I do not want to kick off the next 50 minutes, but I suggest we come back and talk about this again, because my next question is when the cost comes due. Because again, let's go back to a lot of free money in the system right now. And I think there's a lot of free money in the AI ecosystem right now. And at some stage those companies are going to need to, to charge what it's really worth.

Speaker A: Yeah, yeah.

Speaker B: And it's starting.

Speaker C: What does that do to the ecosystem?

Speaker B: Actually, it's coming very much sooner than we think. Uh, first, there are many companies that started last year that went bankrupt this year. I Mean uh, I talked about human pin, uh the rabbits which were like trending stuff but they didn't solve anything. They're just shiny objects. And also companies that were doing LLMs also that raised hundreds uh of millions of that went, that filed for bankruptcy because they couldn't compete. Uh, and there are talks that uh, uh, Chad GPT is going to go from 20 to 200amonth uh soon. But again that's the problem for the companies to figure out ways to again monetize because those 200 are useful if you can make them 2000. That's always the thing. So in a sense for the average user consumer I think they're going to keep the prices low because they want still again 200 million compared to 5 billion people. So there's still a long way to go and companies like OpenAI but mostly Microsoft behind them. Uh, and again there was a uh, something today that uh, came out that Microsoft is going to put Copilot, so the AI into the user version which now, now it was only for the pro licenses uh, that were paying and now everyone's gonna get it because they want to achieve uh, we're talking about Amazon, they want to achieve that critical mass of everyone using it. Uh, chatgpt is another thing because they rely on funding so they're going to increase the prices for the pro teams and the team plans or enterprise plans from 20 to 200 or more which is in the talks. But to me it's also a good thing because companies will have to think better uh, instead of just saying okay, we have an OpenAI API behind the scenes and it's going to do the stuff. Because what I've realized and particularly in B2B SaaS is that there's a saying in AI which is garbage in Galbridge out uh, meaning that many companies put AI on their websites. For example I work for the um, outreach automation company and they say okay, you just write what you want and AI generates a sequence. But if you say I'm a uh, B2B company that helps entrepreneurs uh, do whatever if you, you remain generic, you're going to get a generic answer. So you rely, as soon as you rely on the user input you have to think of ways to minimize the risk. And so having prices increase, I think that's a good way for companies to become smarter and stop just throwing AI, uh out the window like that and saying we're powered by AI, where behind the scenes it's just a ChatGPT prompt running.

Speaker C: Yeah, with no data, unstructured garbage Unannotated. That's the next 50 minutes. So we've got a series of three coming up

Speaker B: for the next six months.

Speaker A: As I think the thing is, as the other thing about this, these topics is that things are changing quickly. We're learning a lot. There's a lot going on. So I think it does make sense that we will invite you back to talk about it some more in the maybe in the new year because I do think that there's, there's a lot to discuss and I think there's a lot of change coming. So. Yeah. Yeah.

Speaker B: Great.

Speaker A: Very cool. Thank you for joining us. It was lovely to have you and it was lovely to meet you. And so tell us the name of your. Where should people go go? Tell us the name of the newsletter and where they should go.

Speaker B: So it's the productkorea.com uh, and basically there you have a community on Slack. You have uh, the newsletter that you can subscribe on Substack but also directly from the site and get again use cases every week. And then there's my LinkedIn account, Dos Santos Tony, uh, where people post every day about AI stuff also. So that's where people can go. And if they have questions they can reach me there.

Speaker A: Awesome. Awesome. And, and we'll share all that when we, when we launch this.

Speaker B: Yeah, and I'll share the, the list of free a courses and free courses for, for, for the audience. So if they want to.

Speaker A: Yeah, if you want to share that, we'll actually put it on our resource page on our website so people can tie to this episode and that way people can get access to you and to that, to that list. That'd be great. Awesome. Thanks for joining us.

Speaker B: Thanks a lot. See you soon. By.

Speaker A: Sam.