How AI is Revolutionising Procurement and Business Strategy with James Crawford

ABOUT THE GUEST

James Crawford is an AI Business Strategy Consultant and Programme Lead at Touchpoint Change Consulting and specialises in AI facilitation, change strategy, and programme design. His expertise extends to process improvement, organisation design, people development, and fostering culture change, with a focus on delivering ethical, AI-driven transformations. Touchpoint Change Consulting supports organisations across utilities, telecoms, engineering, healthcare, and financial services, offering tailored solutions in business change, technology adoption, and continuous improvement. With clients ranging from under £1 million to over £1 billion in turnover, James works closely with leadership teams to drive impactful change. His in-depth knowledge of Large Language Models (LLM), Data Science, NLP, Data Analytics, and DALL-E underpins his strategic approach to innovation.

To learn more about James and how he may support you, go to:

Website: https://www.touchpointchange.co.uk/
LinkedIn:https://www.linkedin.com/in/jamestscrawford/

ABOUT THE HOST

My name is Dave Barr and I am the Founder and Owner of RLB Purchasing Consultancy Limited.

I have been working in Procurement for over 25 years and have had the joy of working in a number of global manufacturing and service industries throughout this time.

I am passionate about self development, business improvement, saving money, buying quality goods and services, developing positive and effective working relationships with suppliers and colleagues, and driving improvement through out the supply chain.

Now I wish to share this knowledge and that of highly skilled and competent people with you, the listener, in order that you may hopefully benefit from this information.

CONTACT DETAILS

@The Real Life Buyer
Email: david@thereallifebuyer.co.uk
Website: https://linktr.ee/thereallifebuyer

For Purchasing Consultancy services:
https://rlbpurchasingconsultancy.co.uk/
Email: contact@rlbpurchasingconsultancy.co.uk

Find and Follow me @reallifebuyer on Facebook, Instagram, X, Threads and TikTok.

Click here for some Guest Courses - https://www.thereallifebuyer.co.uk/guest-courses/

Click here for some Guest Publications - https://www.thereallifebuyer.co.uk/guest-publications

Transcript

Intro  00:00

Welcome to The Real Life Buyer podcast. In this podcast, you will hear interviews with business owners, entrepreneurs, thought leaders, authors and technical specialists in their field. These professionals will hasten your development, accelerate your career, and broaden your business. Know how now, introducing your host, Dave Barr, interviewing with a purchasing twist.

Dave Barr  00:21

Hello and welcome to The Real Life Buyer in this episode, I'm joined by James Crawford, an AI business strategy consultant and a program lead at Touchpoint Change Consulting, a firm specializing in change management and performance improvement with a deep knowledge of large language models, data science, natural language processing, data analytics and Dall-E and much more. James brings a unique perspective on driving business change through AI. His expertise covers AI, facilitation, strategy, design, program, delivery and cultural change. So whether you're a small business owner or perhaps a procurement professional, today, let's discover how AI can transform your business, streamline operations and create long, lasting impact. So without further ado, I welcome James onto the podcast. Hi James.

James Crawford  01:17

Hi Dave. Thanks very much. It's good to be here, and it's an interesting topic to explore in this world of procurement.

Dave Barr  01:21

Thanks very much, and it's lovely to have you on board. I hope everything's going to go well for us today. We've had our challenges with Zoom, but there we go. I'm sure it'll be fantastic. So let's dive into the first question. Now, you've held numerous senior positions over your career, but one repeating theme is your involvement in transformations and change. So you're clearly not content with the status quo. So what is it about your background that has driven you to embrace change and transform the way things are done? 

James Crawford  01:53

Well, I guess that is a theme in my career, and I guess I've never been quite satisfied with many of the services that I've received, and it's very much been my passion to improve them for other people. So whether I'm on the phone to utility and I'm now listening to someone else's choice of music for 20 minutes, whether I'm trying to get some basic service in terms of IT support, whatever it is, in terms of the reliability of motor vehicles booking a concert ticket, there are so many glitches that happen, and I've experienced that all the way through my life, and I guess in terms of customer service and performance improvement, what I've tried to do is play my part to improve that for many organizations. And there'd be many different phases of practice and ideas and methodologies and technology that's come along, of which, AI is the latest and most exciting, and that's very much what I'm focused on at the moment.

Dave Barr  02:39

For many people, change in itself is quite difficult, quite challenging, quite threatening, but for you, change seems to bring out the best of you. Perhaps, are there any particular moments where you've identified that embracing change actually is for the better, rather than fearing the change? And particularly when it comes to AI and you add in the dimension of a computer, shall we say, getting involved, people are even more nervous. So what is it that you find that is exciting about change, rather than fearful?

James Crawford  03:12

Well, I guess, apart from my desire for the whole world to work better, and my curiosity in terms of thinking through, you know, are the better ways, you know, in whichever field I'm working, there was a couple of moments which really made a kind of a bit of a standout impression for me. One was when I heard a quote from a management guru called Ishikawa. And I was at a quality improvement conference which is very much focused on the work of Dr Deming, who was quite influential in terms of transforming supply chains, bringing organizations together, constructing a culture of win win for management, and striking out in a slightly different way to how many other management for philosophers and gurus had set out at the time. And he was very much focused on people about culture and about improvement. And one of the quotes which stood out for me was Fauci Kawa that said, quality is about respect for humanity. So for me, it's very much about the people focus side of change. So we are all here trying to serve each other in society. Every organization has a purpose. The purpose is outside the four walls of the organization, because we're all trying to make an impact and bring benefit, whether we're working in the public sector or private sector. So that's very much been a kind of a philosophy behind what I've done, and that people centered approach. I think for me, it's very much apparent in the world of AI as well. We're not simply implementing technology for technology's sake. We're not doing anything for the pleasure of the bots themselves. It's all about helping us to do a better job for our colleagues, for our friends, for our customers, and therefore create a better industry and a better way of working. So that's really what drives me I think.

Dave Barr  04:42

Great. So thinking of our audience today, those SME business owners, procurement professionals, etc, they're very keen to leverage AI. Some may be unsure where to start. There is now a plethora of AI platforms and different things out there. Obviously we all have heard of Chat, GBT, but now there seems to be an AI for everything. So how can AI drive, let's say immediate improvements in procurement processes without overwhelming sort of those, should we say early adopters and smaller teams with complexity? What's your thoughts on that?

James Crawford  05:18

Well, for me, that was one of the big, refreshing moments when generative AI really came of age up until the end of 2022 AI had really happened in darkened machine rooms, and if you wanted to find out about it, you had to go down some stairs to a room with no windows and talk to people who found it very difficult to communicate what they were trying to achieve because it was very, very complicated, very, very scientific, and he needed a degree in computer science to make much of it. But the moment that Chat GPT came out, which was November 2022, suddenly it burst out of the machine room, spoke English and started talking to all of us. And that really caught my imagination, and I was starting to think, Okay, well, what does this mean in, you know, in project management, how do we research the market? How do we construct tenders in a better way? What does it mean for engaging our colleagues, collaborating up and down through different departments when we're trying to do procurement? How does it help us in the job of striking up contracts, managing contracts? What about HR recruitment? How do we bring people into our team? So my mind was immediately starting to race forward to say, if it can really answer these kind of questions, and it really does pull on all of that information that's on the internet. What can it do for us in all these different ways? So for me, that was the moment when AI stopped being scary and just started being useful. And since then, I very much focused on working with SMEs and charities to say, what can it do for you? Where do you start? Let's understand some practical ways that you could explore it. Speak English, be natural, you know, strike up a bit of a rapport with the with the bot. So whether it's chat GPT, which is very good at general business, talk, whether it's something called perplexity, which is really good for research if you're trying to, you know, look at the supply market for something if you're trying to generate a whole load of citations, so you can then explore and find out more about particular suppliers or trends within that particular sector, whether it's something for more of your personal life, like pi.ai which is, I find quite, you know, zany and more chatty and very good for domestic questions you might have about, you know, pictures or car maintenance, or you know which, which Film to go to. They've all got kind of different twists and slightly different personalities. If I could put it like that, then using the right one for the right job makes it more natural to have that kind of dialog with it, and therefore get more useful input out of it. And I do find that, you know, when I'm talking to people, just explaining some of the kind of practical ways to start, but also some of the limitations, many of the fears that people have start to melt away because they can see, okay, it's useful. It's not quite so scary. I can be myself. I don't need to learn any special new programming language. I can just talk to it, either with my voice or through the keyboard, and then it'll start to be useful to me. So for me, the message for everybody is, learn a bit about it, but just start the end of it, familia, you will find yourself becoming more comfortable, and it will save you time, week by week.

Dave Barr  08:07

Yeah. Now the key thing, I guess, with with AI, and even when you're training, somebody else, is trying to ensure that you've put things across in a manner that you make sure they get the output from the context of the input, the less you put on. Obviously, in the input, the perhaps the more confused or the more generalistic the kind of response. So how does one go about making sure that the input you give is comprehensive enough to give you the output you desire? How do you go about doing that?

James Crawford  08:38

That is absolutely key. It's rather like having a new graduate or a new employee coming to sit next to you on a Monday morning when you're doing a job and you need to turn to them and start to delegate tasks or ask them questions, and if you say to them, tell me about the supply chain for automotive components, they'll give you a very generic answer that they might have read In a book or learned on a course or had some previous experience with, but if you actually say, right, what I'm trying to do is I'm trying to upgrade this fuel injection system. I'm trying to procure these parts ready for my colleagues in operations, because they're doing some innovative research, and they want some extra facilities and extra features within the product that they can then test on their loom, and therefore work out the etc, etc. The more you give the person the kind of background to what you're after, the more relevant they can come back with information. And then at that point, of course, they'd start to talk about fuel injection systems and the experience they might have there a report, they might have read a trade magazine. The thing with AI is that not only do you kind of need to direct its attention to a particular part that you most interested in? But of course, it's already ingested the whole of the internet, everything that's been published in terms of research, product features, brochures, websites, people commenting on how well or how badly the product did for them, less than. Learned from projects that went wrong, all of that that's been published, it's learned and ingested, and what it does, it then formulates its answers back to you based on that knowledge. So the more you give it in terms of what you're trying to achieve, the insights that you like, background that you've got, you might be able to upload some documents, or you may just simply talk to it or type out, you know, a lot of information, but the more bespoke information about your own particular situation you give it, the more bespoke the output is going to be, and the more useful it's going to be to you. And the other thing to note is that you can enter into that dialog, because it won't forget the conversation halfway through, if you do a Google search with you know, a few words, it'll give you 21 million links, and you can click as many as you like and do your own research. If you ask it a follow up question, it'll start from scratch and give you another 21 million links. But there were no connection between the two searches. The genius of chatgpt and these other bots is that it strikes up a conversation with you. So each time you add more information, you ask it to clarify. You request more information. You say, Okay, well, that sounds really good. Put it into a table and schedule these actions over a three month period, and it'll do that for you, and you just have that conversation. And the more you get into a natural conversation with it, the more you'll find that you'll coax it into providing the kind of information you want in the format that you want. And it's just more useful,

Dave Barr  11:17

Yeah. So really quite interesting. You mentioned about the date, as you say, that the time of information that it's able to absorb, and I understand from if I'm correct, that it's not completely up to date, depending on what release we're on, the data is going to have some kind of lag. So you won't have the latest. Latest Is it becoming very close to real time data? So everything that's been generated immediately is going to start to appear in AI systems, or is there always going to be, say, three months or six months, or maybe even more of a delay before it gets ingested, so to speak?

James Crawford  11:56

Well, that very much depends on the nature of the system you're using. Now, some of them are absolutely buying up to date, and they search the live internet. Others are 3, 6, 9, months out of date. That's partly due to the way that they're actually set up. There's a process called training, which is an astonishingly complicated process where they again, literally take the whole of the internet and they process it, sentence by sentence, word by word, concept by concept, they make quadrillions of connections between all of these different words in order to understand how these concepts relate to each other. And if someone's talking about this particular question in this particular context, then the kind of output that they want will look a bit like that. But the time it takes to actually process all of that is eye watering, and it could easily take, you know, a computer system, which costs $2 billion it can take it six months, working flat out, to produce one of these models, which is why, by the time it comes to the end and then it's published, it's six or nine months later. So that's really where chat GPT started, but now they've started to also supplement that with reaching out to the live internet with up to date information. So whether you're using Google Gemini, with all of its links to the Google estate or perplexity that I mentioned earlier, which is really good for research and has links out to live websites, or even our chat, GPT, which has got its own search function, you can actually get the best of both worlds, but these things are very much developing at the moment. But as you say, it is useful to know the system that you're actually using at the time. How was it set up? Does it have a particular cut off date? Am I looking at 2023, information, or have I back up to date, so a little bit of background research? But you can also ask it, of course, if you're using one of these systems, you can ask it what information it knows, what's its cutoff date? Can you actually supplement it with information that was uploaded to the internet yesterday, and it will tell you exactly where it's coming from and what it's able to do for you.

Dave Barr  13:46

That's fantastic, isn't it? It's amazing how fast this technology is progressing. Really exciting. Now, thinking about Touch Point, you specialize in change management for organizations from small through to large. Thinking about the small business owner, you've got limited resources, financially and with people. So how would you help or suggest to them that they prioritize their strategy? You know, how'd they go about getting their best bang for their book? So, so to speak, and adapt to new technologies like AI to ensure they get both efficiency and sustainable growth.

James Crawford  14:22

Well, that's part of the magic of this, because all of this knowledge is available in an easy format at the touch of a button. Small businesses really have the power of large businesses at their fingertips. And for me, it's it's not necessary to sit to kind of sit back and think deeply about your strategy and how you're going to use it, and they always can do too much research. The most important thing is just to start, because it's so accessible, because because you can engage just in English language, it's very easy to start to get the hang of it, and that will then inform your journey as you learn more and more, and start to apply it to different areas of your work or your domestic life. So there are some techniques, which I find quite useful in the back of my mind. But nothing beats a natural conversation. So the way you like to express yourself, the way you would typically turn to a colleague to ask some information or some analysis, or for them to go away do some planning or some research, strike up that kind of natural conversation, because the more natural it is for you, the more you'll be able to get the kind of outputs that you want. But if you are asking about a bit of a structure, I kind of use the five P's, which I find quite useful, to kind of remind myself where I'm coming from. So the first P is to do with "pick a persona". So think through, if you had an expert next sat next to you, what kind of expert would it be? Would it be the best financial accountant that you've you've ever worked with. Would it be a graphic designer? Would it be someone who's really good at innovation and bringing different ideas together? Would it be an industry expert? So you can ask about the latest suppliers and the magic quadrants and the people that you should be inviting in to have a pretender discussion, for instance. So describe kind of who you are after, the kind of skills that you'd need, the kind of knowledge you need. And the reason why that's important is because it will then trigger the model to start looking in those areas of the internet. It'll bring together the kind of knowledge that you're after and focus much more on that particular side. So that's a really good place to start. Secondly, if you could get it into the right headspace, you could "prime it" so you give it background information. You could upload maybe a strategy document or a draft tender, if you've got one, or the previous round of procurement that you went through. You could upload maybe a report which talks about some of the issues you're having with supply at the moment, or some of the problems you're having with existing suppliers that you're trying to rectify as you go out to market again, but that background information again, allows it a rich source of information to draw on, to combine with this other knowledge, and then give you something which is quite specific and bespoke for you. So then you're ready to have the third P which is to "prompt it" describe exactly, you know, what, exactly are you after? What kind of insights, and then how you present it. Is it, you know, a block of text, which is kind of its natural state, or do you want it in the format of a plan or PowerPoint slides or bullet points or a table, and then finally, perfect it. So get into that dialog with the bot to say, right, I'd like more on this. I don't like idea number two, so ignore that, but numbers three and four look really promising. Please develop those out. Give me some background information, provide citations and name the top three suppliers in this space, and it will then go away and produce that for you. And then you come back with further questions and further refinements, until eventually you've got something which is really useful to you, and you can maybe then start editing it, editing it manually, so you could really start to make your own particular points of view, your own style, you know the words you use, edit it to make it your own, and then you'll be having something at the end of it which you absolutely own. You expand by but it's probably taken you less than half the time it would have done to actually produce it.

Dave Barr  17:55

That's really useful. I think having a structure for people to follow is going to be so much more beneficial than just pouring stuff in your mind down on a piece of paper. So let's just quickly go back through the five P's. So that was, Pick a persona. You mentioned Prime it, you mentioned prime it with background information? Yes, yeah. Then you do a Prompt what you're looking for, and then help me out the fourth and fifth.

James Crawford  18:18

The fourth one was "Presented". Explain how you'd like the information to be presented. And not only in terms of format, but also maybe in terms of language. If you're engaging suppliers from a different country, it could be in language. You could say, you know, I'm a non technical procurement person. I'm doing some really technical procurement here. Can you please write this next section using very technical language? Because that's going to be more interesting and more interesting and more relevant to the engineers that will be reading it. Or vice versa, you could say, do it in language which is suitable for a an undergraduate or someone who's 16 years old, an apprentice, and then the final one is "Perfect it". So get into that to and fro dialog. Have a natural conversation, and end up with something you can really be proud of.

Dave Barr  19:02

That's really good. I think it's very important for people to have that structured down, and that really helps you to think about how you lay out before you start pumping AI with lots of questions and random things. I think that's a beautiful system to use there. So thanks for sharing that now you've worked extensively with ethical AI, I believe so with AI playing an increasing role in procurement and supplier management. What would you say are some of the ethical pitfalls businesses should be aware of when integrating AI driven solutions into their business models?

James Crawford  19:37

I think there are several dimensions of this. Again, going back to our first question. If your ambition is to make sure that you provide the best possible service to your customers, your colleagues, society at large, then the way we use AI obviously has to be done in an ethical and a transparent way. And there are several different dimensions to that. Also. I'll come on to data protection in a minute as well, because that's one of the absolute vital things. Is. You're actually uploading procurement, documentation, business strategies, internal reviews, lessons learned, reports, whatever they might be. If you're uploading that to a chat bot, be aware that it may be using that for its own future learning for the next version of the systems. There's that old adage that if you're not paying for the service, you're not the customer, you're the product, and that's very much how the free versions of these work. These systems work as well. So they'll take all of the conversations that they're having with people in different sectors, and they'll use that as good material to produce a better bot for next year. And that's not necessarily what you want if you've got confidential information or personal information that's being provided. So there are several ways of making sure that your conversation is safeguarded. It's held within your enterprise. It's not part of the next training run, and it's not used in any other way. So make sure that you cover data protection first. I think that's a one of the key things to start with in terms of ethics and transparency, making sure you don't screw up by giving private information to other people. But then a bots are becoming so so good and so natural at the moment, that whether you're reading an email which was produced by one, or whether you're having a conversation with one on the phone that may have done a cold, cold call for you, I think it's really important to make sure that the reader knows where that information has come from. And that's not always as easy as this house. Certainly, if a bot cold calls you and has a natural conversation about, is your energy contract coming up for renewal, and do you live in a detached house? And right, we've got a special offer for you, and it's this. It's not always easy to say whether it's a real person, unless you start to literally ask some direct questions. So in terms of transparency and openness, I would always advocate that Bob would say, Hello, you know, this is a, you know, an automated service. I can answer all your questions, and I can put you through to a real human at any point you wish, if you want to explore it further. But I think the more we get used to these systems, the less often more that to happen, because it'll just be easier, and the last thing we want is to go through to a human and wait 10 minutes in a queue ready to speak to them. We may as well, if the conversation is going really well, talk to them, but there are very few regulations about that at the moment. So it's a bit of a wild west situation. These systems are really useful for us, but there aren't any standard rules, neither from regulation nor from business culture. That means we're all working in the same way. So we're all kind of finding our feet and finding the best way to do it. So the number of dimensions in terms of ethical AI to make sure that the information is good. It's of cell quality. It protects the user. Humans know on what basis they're engaging with these things. Those are different dimensions which just need to be considered in any particular project. So when a small business owner wants to kind of launch into chat GPT or perplexity or Gemini and start using it. That's great. Make sure you do it safely. Make sure you don't upload information that's going to be used in the wrong way. Drop me a if you've got any questions on that. But the the challenge comes, I think, when we're talking about much larger organizations, there was a famous incident a few months ago, which you might have read in the press, when a customer managed to go to the bot of a delivery company into saying how awful the service was. Now, if, if I was having a bit of banter with a customer service agent, and they said, Yeah, I get this all the time, it wouldn't make the headlines. But because it was AI, it was in all headlines. So large organizations are naturally having to be a bit a bit cautious and make sure that they got all of these different dimensions covered, so they don't drop themselves into some bad publicity at the wrong moment.

Dave Barr  23:24

Some valuable insights there. It's interesting, the difference, you say, between a free version and a paid for version, that suggests that the terms of engagement be slightly different when you start paying for something. 

James Crawford  23:36

So yeah they are usually but they also change, because this is such a new industry, not only do the products get updated surprisingly quickly, but also the Terms Conditions still evolve as well. So whether you're using an AI to record a meeting and produce a transcript, there are some systems which automatically say, right, we will use your information. We'll try and de personalize it, but then we'll use it to train the next bot. But there are others which have much more enterprise grade security. And if you use, for instance, Microsoft teams within your Microsoft estate, you'll have the same data privacy as you would do for your other information. Similarly with Google and zoom, but some of the others, it very much depends on whether you're paying for the product or not. So it's worth doing a little bit of research. And again, something like perplexity can help you do that research, produce a table for you, mix the terms and conditions, and you can make sure that if you're in a medium or large organization, all of your colleagues are working in the right way.

Dave Barr  24:28

Really interesting. Thanks for sharing that. Now we've seen quite a few global supply chain disruptions, even now, stuff's going on that we rather not see, that's affecting everybody, and some businesses might be looking to use AI to help them build a more resilient and agile supply chain. Do you have any real world examples where your company has helped a small business or even a large business to get the AI to support them in trying to drive more resilient. Importance and agility in their supply chains.

James Crawford  25:02

I think you're right. That is very much a topic at the moment, and for the last few years, particularly since COVID, I think also probably building up before that, organizations were thinking through, how do I make sure that my supply is secured? Then there's been a big push towards globalization and virtualization of supply chains. But that does come with risks. So certainly, when there are supply glitches, then suddenly, if you've got a just in time process, within a couple of days, you'd start to be disrupted. So you get all of the benefits through the majority of your working cycle, but as soon as something does happen, you're adversely impacted. And interestingly, Japanese management, back in the 70s and 80s, actually thought that was a bit of a benefit on the basis that a glitch would cause real focus to happen, and therefore people would be really starting to think through, how do I counter measures in how do I make sure that supply is assured? How do I overcome these potential issues and manage risks in the future? But I think they had quite an interesting culture in terms of business, but those the elements of those culture that the Japanese used and many other organizations around the world. Increasingly, of course, they've been the subject of conference, speeches, case studies, books, analysis, research, all of which is known to these AI bonds. So it's an amazing resource to actually start to research the background to some of these risks. How does it apply to organizations like mine, in terms of size, scope, risk, scale, sector, country, regulation, etc. It can do a lot of that research if you just sort of prime it to say, these are the kind of areas I'm concerned about. Please supplement it with a cup another couple. It'll have that conversation with you. Then you can say, right? How are other organizations managing these risks at the moment? What kind of countermeasures are they putting in? How are they organizing their supply chains? And how are they kind of shortening their supply chains to, you know, reduce time to market, as well as incorporate the just in time planning, but also with the ability to change specification more quickly. How should I be thinking about these things? What kind of topics should I be speaking to my board, about my procurement colleagues, about the business, about and it'll come up with different sets of questions or or issues to discuss with each of those groups. You could then bring all of that output together, if you've got maybe some transcripts from a couple of conversations you've had, feed that back in and say, right? Bearing all of this in mind, produce a three page report that just summarizes where we've got to and the kind of actions that we could take in priority order. And all of these outputs will be done at such a pace, you know, within a few seconds that might have taken you several hours over a period of a week, because you've been in and out of other meetings, and suddenly you're three days ahead of yourself. You've got some really good output. You've incorporated all of your colleagues main main points, and you've got the document back out to them for review and for comment themselves. So in terms of these kind of strategic issues, it can be a really, really powerful tool, and just can improve that communication with people, make sure that it does forget or ignore one particular person's voice, and just gives you the opportunity to move more quickly. Brilliant.

Dave Barr  28:10

That sounds really interesting. Now, before I ask my final question, I'm sure people will be very intrigued to find out more about you and the company that you have. Can you share? Where would people find such information and make contact if they needed to.

James Crawford  28:24

Sure, people are welcome to contact me on LinkedIn, there are actually a surprising number of James Crawford. There was one particular James Crawford about three or four years ago, started contacting all of the other James Crawford, so we're all connected as well. If you search for me on LinkedIn, under James Crawford, Touch Point, I should be at the very top of the list. The other place, of course, is the website, and that describes more about our services, one of which might be quite interesting, which is the 3I's workshop. And the 3I's, really, for me, are a way of helping an organization think through where are we going with this? What difference could it make? Where do we start? How do we pull our people together so that the first time is in form, and then it goes on to imagine and initiate? So initially, let's be "Informed" about what AI is out there at the moment. How are other people using it in my sector? What benefit are they getting? You know, what's coming next, but isn't quite there yet. That will then leads on to the second stage of "Imagining". So how could it affect the service that we've got, the supplier reviews we've got, how could it do some of the automated management and tracking of supplier performance? How could it spot trends more easily for us, produce some reports for different audiences and make sure that we're on top of the supply issues going forward? And if we could have the whole pile of ideas that might be really powerful for us. And then the third one is "Initiate". So what ones we actually going to start to work on? What are the priorities? Where are we going to get the easiest benefits? How will we bring people together to actually develop their skills and their routine? If you like, the rhythm of using AI in their database. Work. You could read about that and others on the website, which is touchpoint change.co.uk.

Dave Barr  30:06

Brilliant. Thanks for sharing that. So we mentioned earlier on how fast AI is developing. So with your crystal ball, love to understand from you what you think the AI driven trends and innovations. You know, what do you see as coming down the line in, let's say, three, maybe even five years? What sort of things do you think is going to make it so amazing, even beyond what we have now today?

James Crawford  30:32

Well, this is some subject to something called exponential change, which is sometimes a bit hard to visualize. But I think the COVID crisis really gave us some sort of a good example of that. And this is when change just gets quicker and quicker and quicker and quicker. And I remember back in early 2020, we were reading about some of the news reports about communities in China being locked down. And then later on, they had some cases in Italy and they got locked down. And I was thinking, Well, I can understand why China would want to lock down relations to kind of stamp out the, you know, the virus, as much as they can. But I was really surprised when Italy, a European country, did that as well. I think they'd have something like, what two deaths by that point? I thought, Well, surely that's a massive overreaction for two deaths. But of course, when you've got the virus, which is doubling so many times each month, you maybe go from two cases to four to 16 to 32 and it's quite some time until you've got 1000 cases, which, in the scheme for population, is almost nothing, but then suddenly it's 20,000 and then 100,000 and then 6 million. The line almost goes vertically. And that's a good example of how exponential change, I think, is happening also within the world of AI. So as AI is now starting to write its own programs under the guidance of people, the rate at which it's improving and becoming more capable is really accelerating. And we're talking about some things today that simply weren't possible two years ago. They were barely possible 18 months ago, and then suddenly, all of these features are made available to us, many of them for free, but all of which can be helpful in our day to day work. So the key thing is to stay abreast of it by getting into the regular use. And when you move beyond adoption to what they call augmentation, which is we work closely every day with a number of different AI systems that just help us do things better, quicker, cheaper, more easily, take out the drudgery. Hopefully we go home on time instead of late. We don't have to work in the evening. We have a two hour lunch break in the due course, that's my dream, anyway, and in due course, that will then move on to the next stage, which is called AI agents. And we're starting to see those being released at the moment, at the moment, at the end of 2024 and that's when just simply ask individual questions or give a task to an AI agent, you can actually ask it to undertake complete role. So for instance, you could say, you know, I want to set up an agent that's going to monitor the performance of this particular contract. They're going to have access to the mailboxes from people that are dealing with the supplier. They can pick up the tone, the nuances within the language, the issues that are being raised. They'll make a judgment in terms of how well the suppliers collaborating, feeding back issues. Is there a positive working relationship there? It'll either write the management reports each month, if needed, or certainly absorb them and comment on them and look for trends. And it'll basically be sort of working along in the background, almost like it's got a job description, so you don't have to tell it every single thing to do. It's just working away in the way that colleagues do. And then from time to time, maybe several times a day or several times a month, it would give you insights. It would prompt you to say this is going on. These are some issues which need to be resolved. I've spotted some early warning signs. These are the actions that you might be able to take. So that's the next stage that we're starting to move into, not just simply being a tool for us, but almost being like a virtual colleague. And that's going to be very powerful, because that will give us all a lot more capacity in terms of what we're doing, but it will maybe test our leadership skills, because we will need to be good at describing what that bot should do, setting out that job description for want of a better term, managing that performance and continually improving what it's doing to make it as relevant as it needs to be, to support a whole procurement and supply management exercise and then beyond that, it's quite interesting to see some of the robotic firms come up with very dexterous robots that can be potentially used in small businesses and small warehouses. There was a demonstration the other day, I think it was of Elon Musk's, where a robot was having a natural conversation with someone, and they said, right, well, can you just move those eggs from this pile to that pile? And it had the dexterity to reach down, pick up the eggs, one by one, without crushing them, and just move them carefully, place them over there. So the mixture of intelligence, language, the dexterity, but also the kind of the human mobility. These, these robots will be quite revolutionary for small businesses. So, for instance, you might, you know, have a team of people that leaves the warehouse in a bit of a mess in the evening, when they all go home at five o'clock, five o'clock in the evening, by the morning, you know, the two robots could have cleaned everything up, put all of the stock back on the right racks, swept the place up, loaded the vans, and when the humans come in, they'll be ready for the next stage. So effectively, you've got two shifts going on. So I think that's going to be very, very impressive. But even though the basic technology works now, it's probably going to be five or six years before the price really comes down to a level where most small businesses will find it cost effective to have boxes as part of their workforce, but that'll be really interesting, just again, taking the next step. So physically, we'll be sharing a space with these bots, as well as simply on our computer screens. So the world is starting to move on, and if we get it right, it'll work really well for us all as individuals, but it will need some head scratching and some mental gymnastics, I think, for many people to become comfortable with that and find out how best to work with them and how to make the most of it. 

Dave Barr  36:05

Well, that's, that's fantastic, really interesting. That certainly gets the mind juices flowing about all the different opportunities that could lay ahead, both in business and in your own domestic, personal life. Some, you know, huge time savings there. Yeah, I think that's that's going to make me think for a little bit. I've got some ideas in my head. I'd love to talk to you afterwards about that. 

James Crawford  36:26

Sure. 

Dave Barr  36:27

Well, that's brilliant. James, thanks very much for sharing your knowledge today. Been a wonderful podcast. And yeah, thanks for sharing all that knowledge with my listeners as well. It's really appreciated. Yeah, you're very welcome.  Thanks, and take care, mate. Cheers. By then. So there's another Real Life Buyer podcast. I do hope you enjoyed it, and it has given you some ideas and inspiration for greater action and achievement. Don't forget to subscribe so you don't miss out on future episodes and a five star review would be most appreciated. If you would like to discover more about me and what I do take a look at www.thereallifebuyer.co.uk. Bye.