You need Factory Workers & Free Thinkers. Which are you? - with James Drury

Welcome to Building Your Leader Brand. Today on the show, Bob is speaking with James Drury. The best thing to build is something every single day. Go on it every single day and cap your tokens at least one session per day. If you do that for 6 months, You will be a very, very good AI user. Hi there and welcome back to Building Your Leader Brand. My name is Bob Gentle and every week I speak with incredible people who share their secrets to building, marketing, and monetizing their expertise and the mindset you need for your business to grow and thrive. If you're new, take a second, take a breath, take a seat, and hit the subscribe button. I think that's Spotify, and if you're on Apple Podcasts, it's follow. And if you are here, you will almost certainly enjoy the Personal Brand Business Roadmap. Over 50 pages of everything you need to start, scale, or fix your expert business. Visit amplifyme.agency/roadmap. So my guest today spent a decade inside one of the Middle East's original e-commerce business, rising from a junior analyst all the way up to chief operating officer. Congratulations, James. And Chief AI Officer. Along the way, he rebuilt how the company reports, sells, and operates. And today he owns its AI strategy end to end. Beyond that, he advises CEOs and executive teams on AI adoption and business resilience through the Global Council for Business Resilience, holds an AGI strategy certification from Blue Dot Impact, and has been to the very fancy-sounding Oxford and London School of Economics. James Dury, I make you sound so fancy. Welcome to the show. Super fancy. Thank you very much. So I'm really excited to have you here today because I am a nosy so-and-so, and there are very few people make me jealous on Instagram or LinkedIn, rather. And you are one of those. I know you a bit. And I know that you are like a proper voodoo doctor with AI, particularly Claude. And goodness knows I'm a nerd, and I really want to know an awful lot more about it than I've ever had the time to really ask you about. So I am really looking forward to this. But for the listener who's meeting you for the very first time, can you maybe do a better job of introducing yourself than I have? The actual meat and potatoes, not just the fluff on the top. Yeah, so as you said, 10 years working up through my family's company, worked through a lot of different departments, got a really good standing on business in general, and then obviously up to COO of EMI Digital, which is the parent group, and then CAIO for Carbone, which is the company that makes the most money. If you're not pushing on AI, you're just going to get left in, left in the wind. So I had to focus on that for, for Kaboom. And then through all of that learning, coaching, developing, I then started up my own company, Outriders. Started off with advisory and consultancy, moved into app building., and now I do AI business setup in Claude Code. That's the main thing I do. So a car alarm has started up here, and I'm going to pretend it's not there. One of the things that intrigues me about what you're doing is I think there are lots of people— thank you— there are lots of people who, who know their way around a little bit of nerdology in Claude and Claude Code. Sure. And they tend to— it's a little bit like fish in a fish tank. Like, people like me do things like this. And so the kind of vision and the systems that I build are for people like me, fixing my little problems. Whereas you span leadership in a very large organisation and large organisation systems. But at the same time, You have a passion for the micro-business owner and really leaning into the content and the lead generation side of things, which is quite a contrast, really. And you and I met through the Advisory Board Centre. You're accustomed to spending time with very senior people, and I think that makes you uniquely placed to really look at big picture AI systems for business. But also small macro-level personal effectiveness tools. I think what would be useful at the beginning is to maybe look at big picture. So if you are a leader in a large organisation and you don't know your way around, I think a lot of people's experiences, they are chasing the bus in their organisation and there's maybe a few kids in silos who kind of know what they're doing with AI. But they're all freestyling and they don't understand the business. How do we take charge of that? Yeah, that's a good question. So I made some mistakes with Carbone at the beginning because I really fell into AI. You tend to, then you get everybody, everybody gets the subscription, right? And then for me, I once trained everybody in the company. Um, I spent a long time doing the training, but if you take into account— as a leader, what you should be taking into account is Plato's Principle. So what you need to be able to look for in your departments, it may not be the manager, but you want to find the free thinkers. You're going to have free thinkers and factory workers. Both of them have their place within all of your teams. If you try and teach somebody to grow a department within AI— because everything to do with AI is brand new— you have to be, in my opinion, you have to be a free thinker. Because what will happen is that you will give people the tool, and then it is down to them to go down rabbit holes. And I noticed it because when I first started doing it, I'm expecting like, I gave you the— I gave you the world. Like, why didn't you go and visit all of these places? Why didn't you go and do all of these things? It's not that you build resentment, but you keep— it's like running into a wall over and over again. And then I— it was just a post on LinkedIn that was free thinkers and factory workers. And I love little things like this because it just validates something in your head and you go, okay, 'Now I see where I went wrong.' So what you have to do as a leader is be able to recognise and find within your department— I call them your AI heroes— but for your AI heroes, you need free thinkers. And then understanding who your factory workers are, because for your factory workers, they're fantastic. They will produce your product, whatever that may be, across finance, marketing, and sales. But if you want them to succeed you have to build the system for them, not ask them to build the system. So that was a big learning— was a learning curve for me in just that with AI specifically, because once you get it right, the free thinkers will just go down amazing rabbit holes. And it's amazing feeling as a leader because what you will —people will just come up to you and say, "Oh, I tried this and I couldn't do that, but then I went away and did this." Or in your one-to-one meetings, you'll have, "I went away and built this, and it's something that I never even would've thought of." So, as a coach, if you are taking it on as you are going to coach the people within your organisation, you'll notice it pretty quickly. There'll be people— you can take like, uh, learning guitar as an example. There will be people who will learn guitar and they will only practise the guitar when the guitar teacher is there. Now, what you're looking for when you're coaching somebody is when that changes to— and you can look for it especially with AI in their prompts, because when you're coaching them and you go into their code or their chat or whatever you're coaching them into, you'll only see the prompts that you worked on them with. The crossover is when you see 5 prompts that you didn't work on with them. That's when they start going down the rabbit holes and when they learn little things. And then that's the amazing thing, because you can turn into 30 people if you use AI properly, right? So even if you have a moderately good marketing manager who is now— who is a free thinker and uses AI properly, Fantastic, you know. Then focus on who are my factory workers, and then use your free thinkers to build the systems for your factory workers, or outsource it. But I like to do it within the company. I think one of the things I remember you talking about was the very first thing you need to look at is your systems, because if you don't understand what your systems are You can't amplify that. You're just going to amplify broken systems. And I think AI as an amplifier is something that I see again and again. If you're an imaginative person, you can amplify that. If you're a lazy person, it'll just make you effectively more lazy and produce even more mediocre work. But inherently, if you run a large organisation and the systems are largely improvised, as I think a lot of the time they are, and it can work well, that's not something you can really bring AI to because it can't support improvisation. You need to turn what works into systems, and then you can bring something like AI tools into that to make it faster, more effective, more efficient, and you can then trust people to do that. How would you approach this in a very large organisation that probably has an awful lot of systems? Where do you begin? You mean in the red tape around large organisations? I don't know. I mean, I'm looking at perhaps from the bottom to the top, or from the top to the bottom, or are there particular departments or particular functions that make sense first? Yeah, what you want— this was another LinkedIn post that I remember from years ago. There's a point, you know, um, you know, there's that video on YouTube, The Crowd Theory, you know, one person stands up and then they start dancing and then so on. So if I was dancing, just not a dancer. Yes, same here. What I would— it depends if you're looking for adoption across the whole company, where you start You want that— there's a point where you're a hive mind of people and it's safer to sit on the side of, ah, we tried AI, that didn't work, you know. Oh, you know, it's just, it didn't make, it didn't make anything better, or whatever it is. Is there a specific department to start with? I mean, it depends what the depends the goal of the business. But if it was— if I had to choose one, it would be one that the adoption that I knew the person that I would be delegating the adoption to would drive it, would push it through that department, that I could then wax lyrical about that. So then people are going like— so it then effectively— I talk about this as well, where it's like, opposed to pushing AI onto people. Once you get a percentage of your organisation over the hump, it becomes pull. And then they go, hey, why didn't I get an AI system put into my department? You know, why I have— why haven't I got, um, Claude Code? Why have I only got 5 times the, um— sorry, why haven't I only got the pro version or whatever it is? So I would look— again, it comes down to the specificity of your business. And what I would look for is the person— either it's going to be you, or know your people well enough to who's going to be the flag carrier at the start of the army, and they're going to pull the rest of your troops forward. What's really interesting is I would have expected the answer to— and I assumed the answer would be an operational obvious first place to go. But by putting it as a cultural win, that's actually really very creative, because once you create that cultural win, you've got a bit of buy-in. Yeah, you then have a broader palette to play with, and permission comes more easily. And I think from a leader's perspective, you don't want to be seen to be imposing things on people, but demonstrating the conditions for success. And that it, it made everybody stronger, it made everybody better. I guess what I'd like to maybe look at next is the macro. And I'm somebody who uses a lot of AI these days, and I see lots of people using lots of AI. But what I find is the way you use AI tends to be a reflection of your personality a little bit. I spoke about AI as an amplifier earlier. I have no shortage of ideas. I don't need AI to help me work out what I want to say. But I like that it can help me say it more quickly and more effectively and more efficiently and create space for where I really add the value.— but that's me. I think that's where I find the greatest problems. I think where I also really enjoy doing, using AI, is helping other people fill imagination gaps. And particularly, so I have a little AI tool, and I use this as an example, but one of the things I'm regularly trying to talk people through is when you're a business, if you're going one-to-one looking for each customer, that's an awful lot of energy. If you built a range of partners around you who already have your customers, same energy but greater output. That's leverage. They already have the relationships. So by cultivating relationships with strategic partners, your route to the individual customers became much more efficient. However, you can point me at anybody and I'll tell you who the strategic partner should be. But not everybody has this. So now I have an AI tool, you tell it who you are, what you do, it'll tell you, here's the palette. Enough about me, I'm kind of waffling here, but I say this simply to paint the picture of how do you help an individual work out where they should be exploiting AI, because it's almost certainly not where they're exploiting AI. Yeah, the easiest way to do this is go into— I don't like the web browser AI, they— I don't like chat. I think everybody should be working into code, so that's what I'm going to talk into. Claude Code is branded as code, but you should think about it as Claude with the ability to code, opposed to Claude for coders, because I think people get— well, I know because clients get put off by it and don't. So number one, just work into Claude Code, get your folders, folders and subfolders all set up. So the way to get them set up, the first thing you will do is you can do this into chat if you— before you get the code set up, but you can say, ask me 100 questions about me. I want to start using Claude Code. I want folders and subfolders set up. Um, ask me 100 questions about my goals, the way that I work, etc., etc. All of these things. And sit there, put yourself in a Google Meet with your Fireflies or whatever your AI note-taker is, and talk. Do not type it out. Because once you get through 15 questions, you're gonna get piss-bored of doing it. I sat there for— when I first did this, I sat there for 6 hours just talking. So you can imagine all of that transcript. What is going to be this— what will separate you as a, a good AI user compared to a crap one is just going to be context and memory. So when you work into Claude code, build your folders, say within those folders add memory.md and build.log, and then what are you going to fill that with, right? That's when you do your own context interview. Now you don't have to sit there for 6 hours, you know, I have— I run 2 C-suite jobs on my own company, so there was a lot to talk about. Generally it can just be an hour or an hour and a half, but the point is, is to let your conscious flow when you're talking, because the question would have gone, and if you were typing it, you get to a point and then stop. When you're talking, you will flow and you will continue to flow better. If you then take that as a practise you just do with yourself once a month, it's even a good thing to like review your business, review your goals, review, review everything, redo all of your context document. It will allow you to use AI a lot better because you've built the brain of your AI into your local folders. That's the main thing that I would advise people to do with Claude Code. So what I can imagine, you spent a lot of time explaining to AI who you are, what you do, who you're for, what their problems are, what you care about. You get to the point where you're then looking at, okay, I've got an AI here that really does understand me, but it's still just knowledge in a bottle. How do we decide what we're going to amplify? I hate to use the word build, but we need to execute on something. How are you deciding with AI's help, what's the most effective way to execute? I mean, I do strategy sessions with my AI every single day, with Claude every single day. I would say, yeah, if you come to a point where you don't know what to do, when you have all of this context and everything put in there, open up a fresh session and say, I'm not sure what I want to do. I know that I want to build something. I know that I'm not using you to my full advantage. I want you to do a full look through all of the con— and at this point there is a caveat here, which the other thing that I teach and tell people to do is you have to have all of your— have your email connected through the MCP connection so it can read through. Read through my last 3 years of emails, right? If you don't know what to execute on, yes, do Effectively, I would— the way I would do it, I might have some bias because this is the way I would like to work. I would just ask it to go look through all of my things and then come up with a strategy plan for all of the things that it can, that it can build. Now, if you didn't do that, a good place to start, probably a content system. You'll get a good, you'll get a good benefit out of a content system. Ask your Claude, I want to post on LinkedIn, I want to build a content system for LinkedIn that creates posts 3 times a week, run it as a routine, help me get started, help me plan all of this out. I could tell you, I can tell you now what you would do. I mean, I build them a lot, but it's a good place for you to start learning And then the— I think that's more the main point that I'm trying to get to. Is there one answer to the thing you should build? No. What you should look for is to constantly be building. I, I am much better at AI than I was 2 years ago, you know. So the best thing— the— yeah, the best thing to build is something every single day. Go on it every single day and, and cap your tokens at least one session per day. If you do that for 6 months, you will be a very, very good AI user. I guess where I was coming from here was the whole question of an alien drops out of space and gives you this magic blob. You think, what am I doing? Okay, you have no idea what it can do. It can do literally anything. But if you don't know what's possible with this magic blob, you might just try and mash it against your face, which is what most people do with AI. So understanding where it can be applied is actually the hardest part a lot of the time. What you described there in terms of asking AI to take a broad overview of who you are and what you do, and then say, where do you recommend we look? Is a brilliant place to begin. It's like you turn on the computer and you click the help file and you say, what can we do? Yeah, let me— okay, I see what you mean. You, you want to— I get where you're going with it now. So with the— I think where people take a misstep is, again, it's like the development department or all the APIs and MCPs Forget it. API, MCP, it just means you connect this with this thing. It's very, very simple. 2 years ago I wouldn't have said that. Now it's nothing, you know what I mean? So what would be a little secret sauce is like where you don't have an MCP connection within Claude Code or whatever app that you've got. Let's just take QuickBooks for example. I am my accountant for my company, right? So I went in to try and get my QuickBooks MCP set up. All the MCP is, if you don't know what it is, is that Claude has already set up with QuickBooks the connection so you don't have to do, you know, give it the— you don't have to give it the API. So I went in there, I was like, cool, QuickBooks has the MCP. So then I went to go and connect it and, you know, because you live in the UAE, you can't have this connection. So I was like, hogwash, my friend. So I went on to Inuit, which is QuickBooks Inuit's developer site, going through all of that. It took me about 2 hours to get— I had to set a load of stuff up, but I got the API key for my QuickBooks. I then took 12 months worth of financial data for my company. I put that into my Claude. It wasn't as simple as that, it's a couple more prompts here and there, but now I have all of my accounting done for the past 12 months. So the misstep, the, the thing, the secret sauce is you want your Claude to be a strategist and an operator. So for that, whatever tool you use, make sure you have an MCP or an API to it. And then as a strategist, you can use, use Claude in strategy sessions, have the conversations. Another little thing that I like to do is, um, ask me— it's generally the, like, ask me 10 questions on this, ask me 30 questions on this. So getting your Claude to dive into your brain a little bit from where you want to go, that's where I would go with it. The other side of it is that if you do run your own business, then you can— I do see this with clients— is that you fall into decision fatigue. So you have to make a lot of decisions. So for that, if you're firing off you know, 12 sessions at once. There could be some element to where it's— you're writing a job, um, not application, you're writing a job proposal for somebody to come on as a creative director, right? And it's just firing off a load of questions. Then you can just go, for all the rest of the questions, you just decide. This is another thing I really like about using AI in that when I get to a point where my brain is a little fried and I need to be focusing more on X, but I have— sorry, on one opposed, and I have 7 different sessions open, I can still move forward, right? I think being able to accept that 80%, like, you're not going to get everything you want It's a fact of life, right? So when you're moving forward on things or trying to push things across multiple departments or multiple levels of your business, or you're trying to juggle all of these plates, yeah, your, your GBT or your Claude is fantastic. Just, just please make the decision for me. Or you can say, if there's a certain person, like for example Bob, you know, make the decision from, from the perspective of Bob Gentle., right? And then it will look at it from your perspective. The metaphor that comes to mind as you're describing all this is one of these sort of top-down tabletop computer games where you've got a little guy standing in the middle of a board. It's all dark, waggling a sword, and he doesn't know where to go, and he doesn't know what to do, and he doesn't know what the landscape looks like. And if you just stand in the middle, you will learn nothing. But as soon as you start moving around, the landscape starts to light up, And this has very much been my experience. If you'd asked me 3 months ago— That's brilliant. Could I build an application? I would say, are you mad? I spent 20 years telling developers what to do, but I never built anything myself. I start moving around waggling my sword and very quickly I've built tools that make me faster, better, more competitive, stickier with my clients. And this is an environment that favours experiment. Now, there is a, there is a danger here that you can become so effective, you start moving so quickly that you think, I have to do more and more and more and more. But what I would like people to take away from this is sometimes it's about creating space for quality. And that's certainly been my experience. I was spending a lot of time doing low-level work that I wasn't really adding any value to the work, but it still had to be done. Whereas now I have tools that do the majority of that work and have added value to clients in lots of ways that don't require me to be there. And they acknowledge that Bob does not need to be participant in this. This is just systems and software. What it means is, number one, I can enjoy my work a lot more. Number two, I can go into depth and really focus on the value-add. But it also means I have time to look over the horizon, which I think most people will acknowledge you don't have time to do very often. As you start to build this flywheel effect, which if you'd never had the humility to be a student again, you'd never have got to experience. I think this is the real value of the kind of work that you do, is it just sort of nails people to the ground for a few minutes and says stop. You need to learn this stuff because it's going to be the difference between those who are treated frankly as factory workers and those who are treated as free thinkers. If we look into the future, it's just— I think there's going to be a period of difficulty we may not experience, but at some point human beings are not really going to be entirely necessary. In terms of productivity. And at some point, I would like to think we're going to have a universal basic income, which will mean the real value that we have as human beings is our ability to create joy for people. And to be able to do that at scale requires us to be able to understand these systems. Because if we're not, we're going to be very unfulfilled, just sitting there taking our money and not really knowing what life is for. A bit of a deviation, but I have actually a technical question here. So you spoke about folders. My experience of Cloud Code, and for the listener, you may switch off or you may sit and listen and laugh at Bob Gentle. When I set up Cloud Code, I didn't give a thought to folders. Now I have a desktop full of folders. If you've done this yourself listening at home, I'm hoping James is about to tell us how we can go back and fix this. If we started off being disorganised in Claude, how can we bring things back into organisation? And I'm talking specifically Claude Code on the desktop. And I echo everything James said about Claude Code. When you open it, just because it has the word code there is meaningless. You can use it the same as you would use any other instance of Claude. But it will be able to make stuff for you if you ask it to. That's really the big difference. Yes, yeah, with your folder organisation, this isn't going to be a long, long fix. Claude Code is connected to your PC, so if you want to roll the dice a little bit, you can just say go into all of my folders and fix it. If you want to do it properly, I would say I want— I would prompt it like, my folders have become very messy across my PC. What I need you to do— this is another little trick as well— like, get your Claude to scope out the work first so you can read through it. Treat it like a little development team that you've got in there. So then scope out the work first, go and review Another trick that I like to do is go do a 3-pass. Do a 3-pass across all of the data. Each time, find something new that you can organise. So keep building on the scope each time that you go in. I want logical, organised folders and subfolders. You can also act as a 4-person development team to make sure that I'm not going to lose anything, I will need you to— and now we're getting a little bit— it depends what you have set up, because if you have routines— okay, well, if you have routines and stuff set up where stuff talks to each other, then you will need to go and cross-check. You'll need to go and cross-check that. But if it's just a matter of organise, like, where would you— this is where you want Claude to look at it from a development point of view to make sure that X isn't going to fall over because I've moved Y, right? What's the file pathway? If you've got API tokens in a certain area, you know, then it'll be pulling from that area, and then all of a sudden one of your routines will fall over. So depending on how complex you have it, for me, because I have mine— mine's organised, I would— for you specifically I would go in there and also put in a thing at the bottom to say, make sure none of my routines, none of my apps are gonna fall over. I run— you can also make it important, you know, I run my whole business through Claude Code. If something falls over, it will be very bad. Do a, do a 5-pass cheque, you know, start talking. There's another one. Start talking to Claude like a human being, right? If you start talking to her like a human being, you will— it's weird as well because you start working with a friend. That's like the weird part of it, you know? Like, good work today, bro. I want to emphasise this for the listener. I think if you're somebody who's accustomed to typing, getting a tool like VoiceInk or WhisperFlow or one of these— if my top pick would be WhisperFlow because it's not WhisperFlow Voice Ink because it's a lifetime deal, an independent developer, and the support is absolutely phenomenal. But do not talk using your fingers with Claude. It's the most unproductive thing. Getting accustomed to speaking to your computer is the best thing you can do for your business. If you work in an office with other people, I'm sorry. I really am sorry for you. But if you work at home, if you work alone, if you work in your own office, Learning to talk to your computer will speed you up so much. I type at something like 30 words a minute. I talk to my computer at 120 words a minute. And it knows when I'm writing an email and it will format it as an email. If I'm in WhatsApp, it knows I'm in WhatsApp and it'll chat. If I'm in Claude Code, it knows it just needs to take whatever I said and make sure if I screwed something up, it fixes it. It's so much more efficient, and it allows you— and you kind of hinted at this— you can invest so much more because the effort to get stuff in there is so much lower, and it becomes really creative. It's almost like an artist waving a brush around a canvas is a free, low-energy thing for a high-impact return. Typing is high-energy low impact most of the time. Talking, there's a good equivalency here. You get low input cost, high output with Claude. It's a difficult thing to properly express until you've experienced it, but once you start talking to your computer, you will never go back. I'm sure you agree. I gave it a go. I I personally found that I like to type because I spend longer on the prompting. I write quite— I think about the prompting if I'm trying, because it's the same thing that you said, like if you've got the speak thing, it almost felt like I was, I was just trying to get it. I needed to take a second to kind of figure out what I'm typing. Will it make you quicker? Yeah, it will make you a lot quicker. But for me personally— go ahead— after me personally, I mean, I tried Whisper, but I, I, I gave it to a couple of people in my company and now they're, you know, it's just they use it all the time. It's the only thing that they use. So it does give you— the one thing that I did like about it was the speed. The speed element, it's like 4x-ing how fast that you type. So depending on what your use case is, yeah, I, I think, and this kind of goes to where I want to go next, if I'm just saying I want that logo 4 pixels to the left, I will dictate that. If I want a carefully engineered prompt that requires planning, then we're looking at a different thing. And I think that's kind of what you're articulating there. You are somebody who, and I know from the output, you have complex, well-thought-through prompts, which kind of brings me to where I'd like to go next. And I think a lot of people are probably hoping I'm going to ask you about this. I saw a post from you. I saw the post this morning. I don't know if it was today or yesterday you posted it, but AI— not AI— LinkedIn has added a report this as AI slop button. And AI slop and content that the creation of which was supported by AI are not the same thing. And your content is most certainly not AI slop. It makes me think, which is kind of the point of it. Now, I want to leave the words to one side because it's fairly straightforward to come up with A, something to say, and then B, prompts, so that it actually is your voice. But what I think a lot of people struggle with is once words become easy, that leaves a giant gap you didn't see before, which is the visual component. My content production workflow has changed quite a lot because the words became easy, the visuals became the point of friction. So now I begin with the visuals because I know I can take care of the words quite easily. Your visuals are next level. How does that work? I do a couple of things. It's like building into— build into a content system. What I do for clients is these design documents. They're amazing. So you, you build a client a design document that they put into one of their folders within their Claude. And what do you want to do? You want to build a website? You want to build an app, whatever it is within your brand, that's the first place I would start. Build yourself a design document that your Claude will always go to. For example, if you're doing invoicing or you're doing a PDF or whatever it is, I don't even— if I'm asking, if I'm doing podcast prep, right, you know, blah blah, create me a PDF for me to read through. It's in my branding. I didn't ask it to do it, it's just because it's built into my system that way. Then when it comes down to your creators, I'm just going to take this from a perspective of a content system, right? You know the book Steal Like an Artist? Yeah, yeah. Still from many, you're, um, still from many, you're a genius. Still from one, you're a thief. So take that as a concept for your content system at the end of the day, because you don't have to reinvent the wheel. Right? So when it comes down to, again, decision fatigue, so it comes down to what style am I going to want for this, or X, Y, and Z. All of your answers are there. Go and take 20 creators. If you don't have a content system, go and copy, just snip, take, uh, take screenshots of, of content that you like and have your Claude come up with them. Now this is the thing that, again, what did I said before is like Strategist operator. So for you to be able to create the content, what do you need? You need an image generator, and then you also need Canva, because Canva has an MCP, which means if you just create an image on GPT or something and pop it onto your Claude, you get that stupid little thing on the left-hand side that's like CR, right? So what you do then is get your— if you build it into— let's say you're not building it into a system If you get an image created— okay, no system at all. Create the image, right? You can do some strategy session on Claude, get Claude to create the prompt for you within Nano or whatever you use, or GPT. Go and create the creative, then take it into Canva, put it in there, put a nice little bar on the bottom. That will get rid of your CR thing on the top, and it will just set you apart from the people that are just actually AI slopping, right? If you do it manually first to begin with, and then you start working into Claude code, then say, look, this is my process at the moment, what do I need to get connected for you to be able to run this as a routine every Sunday at, uh, 1 PM? Then once you have that set up as a routine and And then the trick would be to get your— I use, on Bob's recommendation, Publer for my scheduling. So Publer has a lovely API. Get that API, take the API into Nano, or you can do OpenAI into ChatGPT's image generator. You have all of these connectors. Build yourself a routine that will go and look at— it will go and look at 10 different profiles to see what what their posts did particularly well over the past week. It could go look on— do webhooks into Reddit. What are the hot threads around your specific niche? Could be gardening, for example, right? So what are the hot topics on Reddit for gardening? What is the news for gardening, right? Gather all of this information. So that's your research element. This is one employee, right? And then once you have— your Claude has done all of this research, you can then have it bring that back to you and come up with different ideas for your particular USP. You might be for— you might sell gardening equipment or whatever it is, right? Have it then play that onto whatever you're pushing as your niche on LinkedIn. If you've built into your system that it's going to pull posts that have done well. Maybe you've created a hero post bank of maybe 100 posts, right? For that particular— can you— then you can ask your Claude to like look through all of this information and create 4 different post types, right? So once you've created those post types, that will then separate you from somebody who's just gone, write me a post on Claude. You know, you want to do— you want to Frankenstein everything. So Frankenstein, like, I really like this element of this, but I also really like this. I actually like it instead of using arrows, I want to use emojis to separate everything out. You know, just put your own— be the 20% in the loop. Yeah, right. So then once you have all of that set up, as I said, then you put in your image generator, it then creates the image, pulls it into Canva, Canva adds the thing on the bottom. You then approve everything, pushes all the way through to Publer, and then that's it. You know, that's your content system at the end of the day. But sorry, go on. I'm going to ask about image generators in a minute. I think what you described here is a very complete system, and a lot of people might listen to that and think, oh my goodness, what a lot of work. But the rewards on the other end are you have time to add the pop and sparkle. And I think the one thing that makes me sad is when people— so a lot of people focus on reach online. I want more visibility, I want more people to see me. And they use automation and AI-generated content to give them more volume, but it doesn't support recall at all. People still don't recognise you. And what this allows is, like, for somebody like me, for example, I could post a selfie every day. It doesn't cause me any emotional stress. I've got plenty to say. But it's important to have the polished corporate veneer to sit alongside that, to build status and to build credibility and clarity. But a lot of people will use the kind of workflow you described there all almost as an excuse to not have to, which I think is a real missed opportunity because if you can balance the two, almost 80/20, 80% the kind of content we've been describing here, but then a 20% that's really geared towards recall, "Oh, that's Bob." That's where things really start to work together. That's where I will happily hold my hands up and I was doing it before. But, you know, we all ebb and flow, so I have stopped. And that's why my next hire is, um, I'm interviewing over the weekend, is for a creative director marketing assistant. The main reason I'm hiring this person is, you know, like, hire for what you're not good at, right? I am very good at sitting in this chair and prompting away, you know, I'm very, very good at that. I'm I'm good at managing people. I like doing the sales stuff. Like, there's loads of stuff that— what I am not good at is doing the— I like to call it actually, it's like we become very good COOs and not very good CEOs. So you have to work, and it's like working out, working on you projecting yourself outside your business. I'm, I'm hiring this person to edit, follow me around with the camera, and take pictures of me. That's why I'm on it now. That will be the point of difference for you. I'm really excited. So we were just about to talk image generators, and I'm going to ask the question. Obviously, Claude, Claude Code. For full disclosure for the listener, I spent £25 a month on Claude. It's not a huge amount of money, but it's everything I need. I haven't really hit any meaningful limits with that. I'm sure you're on a beefier account than that, James, but if you were to choose one more tool to add to the mix, what would it be? Only one. And if the answer is not an image generator, we can have two. As a business owner? Yes. You only get one more tool. I'm talking particularly within the context of Claude Code and what we might augment our use of it with. Yeah, I would go with is probably not what I would go with. Motion AI. Motion AI is a productivity tool. It's, um, it allows you to plug in multiple calendars. So for example, if you have 3 emails like me, and it gets your organisation of all of that, it gets put into there. You can set up, you can have like private, like I'll read through it, like private workspaces. I'm getting my, um, full chest tattooed over, uh, 6 days in September. So I have all of that planned into here. Uh, Tattoos— my podcast Tattoos and Typewriters— all of the work that I have for that is in here. All of my client work per folder is in here, project. It's a very good place to manage all of your tasks, all of your projects, all of your teams. The reason why I recommend it is because with time blocking, which most productivity people who are productive do, if you work into a spreadsheet and you miss something or something goes wrong— the thing is, stuff goes wrong. That is a fact of life, right? So you can try and time block absolutely perfectly, but things will fall over, and there's a little element of guilt in there in your head. It would just— will build up. Why Motion is fantastic is because it auto move stuff. So if you have all of your stuff planned out, then when you tick something off, it will be done at that time within the time calendar. And then if you miss something, it's not a problem because it just brings it back through. So it allows you to be able to flow through your work a lot better. And it has an API into your Claude. So if you For example, let's say you are building a content system, right? As you know, as you said, it's like, it's not fine. I made it quite simple, but it will— you have to build it over a period of time. So for that, you want to map it out as a project, manage all of your time properly. Now I want you to push this, um, I want this project to be built within motion. So create all of the start and end dates, it'll do all the timings for you. This is something I don't feel particularly confident in, so I want only one small task per day that is going to be half an hour long, right? And then it will go and schedule all of that. So if you have any anxiety about building, or if you know that you need to dip your feet in a little bit to just to get interested and excited on a particular project, you're feeling nervous, then you can apply that into Motion and then it will pull it all in, schedule it all out, you know, look through all of my client history, add those in as projects under the client folder. Motion, I know that's probably not the direction that you thought to go, but when it comes to if you're a business owner Being organised and productive by time blocking is excellent. By using something like Motion with using AI to supercharge this, yeah, it's fantastic. So that would be my suggestion. I will have a look at that. And if you had to pick one AI image generator? Yeah, uh, Nanobanana is great. I will give you a caveat on this though. When it comes down to all of these, with, with these tools, it's like wacky, it's like Wacky Races or the Gold Rush, you know. Like at this point, you know, GPT's slightly better than Claude, and they were saying this in my team group chat and I was like, guys, just give it a month or give it 2 months because Claude's gonna be here, and then Nano Banana might create something, and then Manus might create something. So what I would say is that just right now where— I mean, right now where everything is, everything's really good, you know. But Nanobanana is the Google one, because it has so much history of pictures, is really good. So I would, if, you know, gun to your head, Nanobanana is what I would be, um, I would be using. If you want to use an aggregator, use Higgsfield. Higgsfield. But Higgsfield is, um, Higgsfield's penny is quite expensive for, in my opinion. And do you pay for Nanobanana, or are you using the free version? No, look, this is another one where it's like, AI is good, isn't going anywhere. It is possibly one of the greatest tech— is the greatest technological advancement of our time, and we're whining over spending $25 or $200. Like, what is wrong with— I don't understand what is wrong with— my brother works for, uh, I won't say what company, he works for an investment banking company, but you know, they're just pushing back on whatever subscription for this and that. You're just like, what is wrong with you? You manage £4 billion. Like, get a good AI subscription, man, you know. You can buy a £1,000 spanner and, and I'll tear the top off a nut. You give it to a craftsman, it's worthwhile. I think there's a proportionality, and if you're dabbling, dabble with free stuff. But as soon as you start to see the rewards, it's worth spending a little bit of money. Certainly with Claude Code, the rewards are very obvious. And I think with image generators, again, you get what you pay for. So I got one for you on that one. Yeah, I got one for you on that one. There is also an element as a, um, as a free thinker and a business owner that you should put yourself in uncomfortable situations. So a good way to do that is, you know, don't buy a year-long— I'm not advising people to buy a year-long subscription, but that's a, that's a silly thing to do. But what a good thing to do is go and spend on— spend some money on subscriptions. Organise it. If you have Motion, then prompt into your Claude, and then you can put all of the reminders in to cancel the subscriptions if you use them or not. But spend some money on it. Add a little bit of risk in there. You know, there's a— this is something that I did. It's like I bought a— I bought a motorbike before I learned how to ride a motorbike, right? So you just You just put yourself— fine, I sold that motorbike and then I bought a different one and now I'm selling that one. Lesson learned over a period of time. But the point is, is to put yourself in the situation that you have to, you have to do something for it. Because if you just have a free tool, there is an element to where it's like, you know, just I'll get around to it. And yeah, so the rewards are so obvious once, once you realise what these tools can do for you. You wouldn't think twice about it. It's the best £25 a month I could possibly spend, and possibly very close to the last money I would stop spending on a moment. So we need to get into James's amplifiers. 3 things listeners can do. A book, a podcast, favourite pizza, an app you love, doesn't matter. These are basically 3 things that you think everybody should do or read or watch or an app everyone should have. That we haven't all mentioned. It could be a mental model, a hack. There are no rules, but 3 things that you think nobody does this, everybody should do this, it's so obvious. 3 things that everybody should do that nobody does. Obviously I'm paraphrasing here, but 3 things that mean something to you. You can ask me the question, I'll think on this part. I'll think on that. I'll just, I'll run off the cuff. So, James Drury, your 3 amplifiers. These are things that people can do, they can read, they can watch, a practise that you swear by, a tip or a tactic, absolutely anything, but they mean something to you. So what might amplifier number 1 be? Managing your Manage your internal feed. So over your life, you will pick up a lot of things that you're interested in, maybe because some— for whatever reason. But when you, when you start to— it's just the, the doom scrolling. I mean, I don't go on the apps anymore. Once a month, YouTube will get me and I'll lose an hour of my day because these algorithms are amazing. What I mean by managing an amplifying technique would be to really manage your internal feed. What are you consuming, right? So when I was— at one point I was only— if I was on YouTube, I was only allowed to watch business or gym motivating videos, of which I found 3 that I really like: 18 Minutes of Pure Testosterone, 30 minutes of Pure Testosterone and Testosterone 3.0, right? But I've watched those videos over 100 times. I watch them over and over and over again. So my internal feed, there's a lot of just quotes in there, stuff that I really personally resonate with. And even if you don't feel like going to the gym at the time, there's an element of, I've just listened to this thing over and over and over it. Like the, the consistency element of Managing your internal feed, like get off if you're on. I mean, for me, I do it quite extreme where I don't, I don't go on any of, any of these things, but you can always make another account. To, to build on that, I think one of the things that I often pay attention to is mental hygiene, which I think is very similar. It's just being really conscious of what's coming into your consciousness and that you are responsible for your state as well. And I think this is also about emotional regulation and sort of general mindset stuff. But yes, what you're talking about there is like your information diet. You, you are what you eat to a large extent. So yeah, I really, really echo that, and I don't think about it often enough. Amplifier number 2, what might that be? Have less friends, I would say. Have less friends and do less things. Like, really decide— like, for me, I work 7 days a week, I go to the gym, and I spend time with my wife and my kid. That's it, you know. I don't go out after 4 PM. You will be very— if you are looking to be an entrepreneur or work multiple jobs. You will be very misunderstood, but that's okay. Decide on what the things that you like, and then look at it from a perspective. Do I just like this because my friends like doing this, right? You know, there will be an element to where you'll go out to— don't go out, you know, go out to birthday. I see it with a lot of my people I went to university with. It's like every single— you know, I don't go to the weddings, I don't go to the birthdays because I'm working on what I'm working on. So an amplifying technique is that there is a trade-off to everything that you do. So you want to be— if you want abnormal— if you want abnormal results, be abnormal. You know, I don't go out after 4 PM. Why? Because my relaxing time is with my family from 6 until 8. I watch Modern Family, I eat whatever the hell I want, and then I just chill. I just relax. It's like sacred to me. I protect it. So if you don't fit within my circle, I couldn't give two monkeys, to be honest, because I know exactly who I am and I know what I'm moving towards. So I know I started with just have less friends, but I suppose it comes down to lock in on the 3 things, 2 things that you really want to do and understand and be okay with the fact that there is a trade-off that you have made that decision. Get obsessed. There's your technique. I think you make it sound rather brutal and harsh. I know you a little bit, and that's— there's a bit of personality in that. But I think what I'm hearing is there's a season for different relationships. And for you, what you're describing is it's a question of value. That are these relationships you can give value to and get value from? And as you grow and as you evolve, particularly as a business owner, what gives you value and to whom you can add value changes. And I think there's also a question, particularly if you're a business owner. True. You have to put your own oxygen mask on first. And when you're responsible for other people, that is a really important thing. If you are not looking after and protecting yourself, you can't support other people. No. You have a mission and a goal at the end of the day. So when you get pulled off— I talk about, um, discipline with clients. So discipline is in three— I say there are three arms to discipline, where it's how long can you stay on the path, right? When you come off the path, how far do you fall? And then how long does it take you to climb back onto the path? So for example, For me, or let's not take me, like, let's say you're disciplined, you've worked, you know, 14 days straight or whatever it is, right? And then when you fall off the path, you go off, you don't do anything for 2 months, right? So that's quite far. But you can also go, like, let's say you don't drink or you don't do whatever added extras, you can go really quite far off the path and come back a year later on. So when it comes to the discipline element of it, managing, being able to manage how quickly and how, how far you drop and how quickly you can climb back up, will be able— you will be able to move forward and stuff. So, you know, when it comes to your friends or going out or whatever it is, you know, there's still an element to where, you know, fine if you want to go out for the beer and do whatever it is, then that's fine. But how long can you— does that kill you for 2 days? Do you then miss the gym for a week? You know, that's what I just said, like with everything for me, I'm quite an obsessive, intense person. I don't accept the trade-off, you know, until I've completed my goal at the end of the day. So I've never heard discipline described like that before, and it's really making me think, and I'm going to spend a bit of time with that amplifier. Number 3. Oh, that was the third one. Yeah, you know what, it would— you— it sounds stupidly obvious, but just learn AI, right? Like, for goodness sake, just go and spend 30 minutes on it a day. Or— this was the baffling thing about coaching 40 people, like 40 people in my company on it, when now I don't do it that way. But it just really baffles me, like, you can turn yourself into 30 people in 6 months. Like, let's, let's be honest, because, you know, with— and even with discipline, you know, it's like— and with moving forward on anything, it is not a like, you just— you don't just work it out in one week, like the way motivation works. Like, motivation will be like kindling to the fire, but you have to be able to learn to work when you don't feel like it, when you're sick, when you're through all of these things. So when you can be able to work faster and better, and you can see value and results quicker that will come from AI, right? If you want, if you want to start posting on LinkedIn, you could learn how to do the graphic design, you could learn how to do the copywriting, you could learn how to do the scheduling and all of this stuff, or you could learn how to do it on AI, build a system for it. And you want to be the 20%, not the 80%. You want to be using your mind creatively. Look through the post being like, that doesn't really make sense from my point of view. This is how I, this is how I would add value differently. Type away. I don't like this creative. I like the new creatives that you've got going on, by the way. They're very cool. Yeah. Um, so yeah, that's what I would say. I was like, amplifier number 3. I know it's quite generic. Um, it honestly isn't. I honestly think if somebody was going to throw you the keys to the time machine and say you can go anywhere you want, you would be quite keen on picking up those keys. And I think that's essentially what we're talking about here. James Drury, I have had so much fun. If people are listening thinking, I need a bit of James Drury in my life, this Claude Code sounds amazing. What's the next step? Yeah, connect with me on LinkedIn and send me a message. I'll make it very, very simple. And you're going to be launching a podcast soon. What's that called? Oh yeah, yeah, Tattoos and Typewriters. I reckon we are, we're probably about a month out from— let's be realistic, nothing happens, you know, two things: you're never going to get everything that you want, and nothing happens as fast as that you would like it to. So about a month. Well, I can't wait. If you're ever looking for a guest, I am there. But for now, links to all this stuff will be in the show notes. If you have got this far, listener, thank you so much for spending time with me. We have spoken a lot today about how valuable time is, and I do not take that for granted. But if I could ask you for one more thing, if you pick up your phone, hit the follow or the subscribe button, and for golden doughnuts, you can hit one of those 5 stars, and If you hit all 5 of those stars, you are literally the best person. If you can't remember how many 5 is, you count them. You've got the same number of fingers on your hands. Oh, I would actually like to say, if there's anybody listening and they'd like to recommend a guest to me, from time to time I've said that, and from time to time people do, and I would love for you to challenge me. Who should I have on the show? Thank you so much for getting this far with us, and James, thank you so much for Thank you so much for your time. You have been awesome.