Insights, Marketing & Data: Secrets of Success from Industry Leaders
Insights, Marketing & Data: Secrets of Success from Industry Leaders
ONE STRATEGY STUDIO - JONATHAN WILLIAMS (FOUNDER). PART 2. Why bootstrapping often beats chasing VC money. Addressing client concerns on AI; and why even with SaaS they often still want you to press 'the button' for them.
If you’ve just raised $20m for an insights business is that a win? Or are you just taking on debt and potentially unrealistic expectations around business model? Just a couple of the subjects we get into in part 2 of the opening interview of the new FutureView series with Jonathan Williams of One Strategy Studio. In part 1 we explored Jonathan’s entrepreneurial journey thought the insight space and we now move onto the question of funding and financial sustainability the sector, including:
- The importance of value, rather than time, based pricing
- Why it may not be a good ideas to ask consumers (in the first place, anyway)
- How far can you get using AI
- Dealing with concerns around hallucinations
- Why raising money through VC/ PE may not be the right model for insight businesses
- Does SaaS work for insight businesses>
- What Jonathan has changed his mind about recently…
All episodes available at https://www.insightplatforms.com/podcasts/
Suggestions, thoughts etc to futureviewpod@gmail.com
You know, I look at people who raise money with VCs and NPs, and you know, I I think of it as debt. You know, people celebrate, oh, you know, I've raised$20 million,$30 million. I'd be thinking, I'm now$20,$30 million in debt.
SPEAKER_01:So welcome back to FutureView and part two of the episode with Jonathan Williams. We're picking up here on the question of how an agency pricing model works when you have an AI-first approach, hence limited to hard costs, for instance, sample programming, etc. We also get into how primary research can build most effectively from an AI-driven discovery process, as well as the subject Jonathan and I first connected over, the influence and fit, or possibly otherwise, of VC and P money on insight businesses. Let's crack on with it. How does the business model work? Going back to the point that we were talking about earlier. I don't mean in terms of you should give away your all your secrets about exact pricing, but what what's what are the principles behind it when you're working with clients?
SPEAKER_00:You know, we don't have any secrets of pricing and business model. We've we don't do day rates at all. We uh we do value-based pricing. If you're going through an insights to opportunity platforms process, we call that a platforms methodology, that's has a fixed price. If you're doing ideation content development, that's our ideas methodology, that has a price. If you're doing codes, which is our semiotics, that's as a price. Same for sizing. And most projects are combination of those methodologies. Now we've we've attributed a value to those based on what value we can deliver, and we've set that so that it's as well as being quicker than traditional methods, it's more cost effective. It's a lower price. I I don't know, typically about a third of what people might pay traditionally on traditional processes. And for that price, you get the human expertise from start to finish. So that's not added on. It's it's human and AI working together. And you get as many iterations as we need to get to the answers that we need. So uh that that's how we structured it. So you you've got complete clarity on the pricing up front, and we'll keep working on the project till it till it lands and it gives you what you need.
SPEAKER_01:Now I can imagine that's very attractive to clients. So they're like, We'll just we'll we'll keep going, we'll keep going until you're happy with it. We've got you what you do. We don't want to charge you more. And critically, within that environment, it's not going to take us significantly longer. Because it's not like we're having to go back and run another thousand-person survey and spend however long that would take.
SPEAKER_00:Yeah, you can you can you can run two or three project iterations in 24 hours. I mean, it's the the the way we talk about speed generally is that it's um you no longer have to wait for the agency with this model. You know, it's just it's it doesn't have to be done quickly. It can be done slowly, it could be done in four weeks, six weeks, eight weeks, or it can be done in when I say done, I mean like like full iterations and and going through different stages of the process, it can be done in a few days. It's just up to you. But before it's just like, okay, we've got to get all our ducks in a row and then oh, let's go to the agency and see when the next debrief is going to be three weeks later or whatever.
SPEAKER_01:And then how does your approach intersect with more traditional market research? So what I might think of as primary research focus for quant, that type of world.
SPEAKER_00:Yeah, so I mean, I'm a big believer in those methodologies, and I think it's all complementary. My view is that in the past, over the last 20, 30 years, whatever people have defaulted to doing primary research when they have questions to get answers to, or they want to be more consumer focused. They say, okay, in that case, we need to speak to people. And maybe before that was the case. The only way you could get to these answers was talking to people. But actually, now there are other ways of getting to it, like what we're doing, you know, exploratory discovery work, leveraging secondary sources, internal sources. You can get much further, much faster and in better ways. And you know, our view is use these sorts of automated processes like one strategy studio and see how far you can get. Maybe you get 100% of the way there, and you don't need to do primary research to get to the solutions. Maybe you'll get 80, 90% of the way there, in which case you then use primary research to fill in the blanks or to go to consumers with really strong hypotheses, really strong learning, good stimulus, and then you get more time from your consumers. So, in in our view, this sort of new world will get more value from the primary research because it's only being used when it's really necessary, when face-to-face time or or people-to-people time, because it won't all be face-to-face, of course, but but direct with consumer time, you're that time will be considered more valuable, right? It'll happen less. There'll be more, hopefully, more time and budgets to do it properly, with a real focus of what you need to get out of it. Because at the moment, if everything's being done on primary research, you've got all this downward pressure on prices, you've got all these problems with all this sample and and uncertainties of that the price of sample's so cheap, the rates of fraud are going up. Whereas if people can afford to pay properly for really valuable time with consumers, then we'll get more value from our consumer time. So that's kind of what I think is the important thing. And I do get worried when people talk about synthetic data in order to replace testing or any kind of research when you've got new concepts. Because I think once you've got your new new to the world stuff, your inventions, your innovations, your new strategies, and that's that's when I think you do need to be interacting with real people to understand how real people respond to something new. And I'm sure you know, with these automated AI methodologies, there's some form of scream screening, which means that you can get rid of the obvious ideas you shouldn't be taking to consumers and you can prioritize. But I have to believe that there's still a lot of value when you've got new concept, new ideas, new strategies. So actually, that's when you need to talk to people. Whereas actually the early exploratory discovery stages, which we do, um you know, just leverage as much of the secondary and internal sources and get the thinking as far as possible first.
SPEAKER_01:And in some ways, as you say, I mean, leaning back on the points you'd made earlier about your career, you're you're accelerating a lot of what's historically been done, but humans were doing it and capacity was very limited. I was like, how much of the web could you read, or social media could you analyze, that what you're doing is actually just taking a lot of the desk research that, you know, due to the technological inhibitions that we had, you know, you you could only do so much of it.
SPEAKER_00:It's more than desk research, though. I can I can see the parallel, and you're right to bring it up, but it is more than desk research. We've designed it so it's the equivalent of doing consumer research, cultural analysis, expert interviews, brand analytics, and then brought it all together and integrated it. Um but the principle you're you're exactly right. You know, to do all of that in a way to get the sorts of outcomes we're talking about now would not have been possible or feasible, or you know, the amount of time. And you know, that just just it's not just accelerating something that humans did it before, but it's actually enables forms of analysis that just weren't possible before. And I think that's that's why it's so powerful, and that's why we believe in leveraging it, seeing how far it can get you, and then you know, bringing in consumer research at the points where that can add the maximum value. Because actually going out to consumers and and you know, doing eight groups when you've just got a question and not much else in terms of hypothesis or understanding is actually not the not the greatest value you're gonna get from your time with those consumers.
SPEAKER_01:And it probably doesn't do the industry any good at all because people go, it cost me whatever, lots of money, and I actually didn't get that much money out of it, as opposed to going, actually, I had a really, really well-framed brief here. I had got a lot of background, thought, thoughtful background insight already. And now I'm gonna get lots of value out of this project.
SPEAKER_00:That's the idea is like how far, how far can you get using these AI-driven methodologies? And honestly, sometimes it's 100% of the way, sometimes it'd be 90, 80% of the way. Um, but you can get very far very quickly in a way that aligns everybody. So it can be really powerful.
SPEAKER_01:Do you get much pushback from clients, though, going, where's this from? Like, what's the data? AI models hallucinate, etc., etc.
SPEAKER_00:Not a lot, because I think when people see the results, you know, we show we show them case studies and then we show them live projects on their pilots and then they're live projects, and they see the value in the results. Hallucinations tend to happen when you're trying to do fat face retrieval of stuff that very similar sort of facts. This isn't that kind of work. Um, and therefore, particularly also chat interfaces are set up for a certain amount of randomness and variability. And when you go environment API, you can you can set up models to act and behave in a certain way that you want them to, to go through various steps systematically, and it you actually can get a different experience. So I'm not I'm not saying that the the things that drive hallucinations you know disappear entirely, but what I'm saying is that you're you you get results that are as good as or better than what you get from a from a human process, you know, and there are biases and problems and challenges and difficulties with the human process. It's not like you're you're comparing what we're doing to a like a benchmark, which is the truth or perfection. Each system has its own strengths and weaknesses, but when people see the combination of price, speed, and and then the proof really is in the value and knowledge of what they're seeing, uh, then then then they understand you know what what the strength of it is, I think.
SPEAKER_01:Again, I'm conscious of time, and so I did want to touch on one of the subjects that was the reason actually why we first got in touch, which is the question of financing of businesses. And so, how would you go about financing a business nowadays?
SPEAKER_00:Well, that that's a very personal thing. So I wouldn't I wouldn't want to suggest to anybody there is a right or wrong way of financing their business. It's totally up to them and what what what is right for them and their business. I'm a bootstrapper, I always have been. So I like building things from scratch. I think if you've got lots of money coming in, you'll spend it rather than think about how can I solve this challenge creatively. You know, how can I how can I solve this without spending? And then you get to different solutions. So I also like the early stages rather than the late stage scaling. So all of that is well suited to to bootstrapping. So all of my startups have been bootstrapped one way or another, uh, which means you're you're you're launching with clients relatively early, you're you've got strong proof of concept. You can't just throw money or people at a challenge. So you have to to work in more sort of creative, more disciplined ways. And that suits me. Um I know there's people look at that and say, but how are you going to get to the to the unicorn, to the billion dollar or hundreds of million dollar value business? It's incredibly hard or impossible to do. It's not impossible to do that. You know, there's people who have done and do do that, but it's it's it's really, really hard to do that. You know, I look at people who raise money with VCs and then PEs, and you know, I I think of it as debt. You know, people celebrate, oh, you know, I've raised$20 million,$30 million. I'd be thinking I'm now$20,$30 million in debt. And then I'm I'm running a business that is losing five million dollars a year,$10 million a year. I I'm just not built to navigate that. I would just, it would stress me out no end. I want to get to profitability, I want to prove the business model. That's just me. Um so yeah, that's my preference. And then I I think people should go for their own preference. I think um VCs and PA firm or PE firms are a good thing because they're vital in order to drive innovation and get these big scalable businesses. So in the in in the kind of uh post that you're referencing, the kind of things I've been talking about, I I've just been talking about how applicable the application of some of those are within the insights industry and what impact it has when a model that I don't think fits starts to be pushed with lots of money behind it, uh, how that potentially damages the whole of the industry. That's kind of I'm not saying anybody should or shouldn't be doing VC and P. I'm not saying that VC and PCP is bad. I'm just thinking about what what models of business are actually right for and good for the future of the insights industry. Is that the direction we're going? And what are we missing out on by not perhaps going in the right direction?
SPEAKER_01:Have you got time, Jonathan, maybe to touch on what some of the problems are around the VC model in terms of the insights space in particular?
SPEAKER_00:Yeah. So I think I think there's a classic s software as a service SaaS playbook, right? Which has played out across many, many industries successfully for these businesses. So why wouldn't you just keep you know playing playing this playbook through? Which is once you build a piece of software that people can self-serve, it's gonna it's gonna cost you a fair amount of money to set it up in terms of investment in development, investment in sales and customer success. But if you can keep that sales and customer success to a reasonably controlled size, right, then you can scale up, scale up, scale up until you you've got, you know, hundreds of, you know, tens of millions and even then hundreds of millions in annual recurring revenues, right? Then at some point you can hit profitability and then you're creating real value within the industry. You you've broken a technological challenge or a commercial opportunity challenge, you've reached that scale, it's highly profitable. Um, and then you know, not not just the VC and P firms and founders benefit, but the whole of society benefits because it fits, right? So that that's a playbook. I just think people look at the research industry, multi-billion dollar research industry, and say, well, first of all, they they assume it's all done by pen and paper for some reason. The pitch always starts, it's a terribly manual process, as if no one's done any technological things in this point. And then they think of it as a sort of a like these other models, something that can be sold by a piece of self-served software. Whereas what we're talking about is an incredibly fragmented world, right? Which is just like there's there's there's retail tracking, there's concept testing, there's exploratory research, there's innovative, there's all these different things that's highly fragmented that requires a certain amount of expertise to get value that actually solves people's challenges in each of these. There's parts of it that it's very, very hard just to get through self-service. I mean, I've had somebody say who was running one of these businesses, even if even if all the client had to do was just to press a button, and quite often a lot is going on behind the scenes that is more than pressing the button, even if all they had to do is press a button, they still want to pay someone to press the button for them. And this goes to another point, which is how well set up are client-side organizations to embrace these soft pieces of software, use them, learn them. There's lots of reasons why clients outsource as well. You know, the financial model doesn't work well, trying to integrate it into internal technological frameworks doesn't work well. Adoption is really hard. Uh, finding the time to use and the expertise to get the value out of them. So, you know, summing it all up, that that's scaling, I think it's very hard to get that right because actually what you find is every client wants it slightly differently as well. You know, it's like, you know, you win in one client, as again we were talking about before, but you don't win in the same way with another client. So you have to rebuild it in a different way to make it applicable to another client. It's not the solution that the industry is looking for, it's just the one that makes a lot of money for the VC and P firms and founders if it works. It's highly attractive and therefore it looks right, but it doesn't feel right for solving the challenges. And none of these companies get to the scale that's needed and the efficiencies to deliver the kind of profitable returns that happen in other industries. Um, that I've seen I haven't seen any evidence of a business actually getting to that point. And that's why I think it's challenging because all the time these businesses are backed by tens, hundreds of millions or whatever of dollars. It papers over the cracks of when the model isn't working. It looks like it's working. They're talking about successes, they're hiring people, they're having lavish parties and putting them on LinkedIn. It looks like a massive success story. Um, you know, they're attracting talent away from other agencies, other solutions. Um, and a lot of the sort of sales energy and stuff, they've got a lot of money spent on marketing. So that it sort of sucks the oxygen about away from the industry, moving in a way that perhaps more suited. So I I don't know, that that was the thought behind it, is just like what models are we in. We should invest. V VCMP have a role in investing, but the models that the industry need aren't the ones that look attractive and are the typical playbook for this space. And I think we've been experiencing that for 10 years plus. And it looks like we're going through another cycle of that with AI.
SPEAKER_01:Yeah, it's I mean, thank you. I think very well put, and I broadly agree with with almost all of that. I think this point around the use case, or it's a bit of business jargon, but being quite fragmented, it's it seems like you know, a lot of the SARS models work very well where you've got a very distinct singular use case. Like so it's what it is. I you know, I I need to farm a tax return in the UK, and there are only so many variations of that, and we'll help you do it very, very efficiently. Whereas I think, as we know in the insight sector, the clue's in the word, in that the client, if you're genuinely an insight business, then looking for specific insight that helps them as a business. And that's quite difficult to scale. It's not to say you can't use technology in a smart way, clearly you're doing it, but it doesn't necessarily fit the models that I think a lot of the VC firms in particular would be looking to perpetuate. You know, well, you've got to have these investments that will return their fund and so on and so on.
SPEAKER_00:Yeah, I mean that you know the the the pitch is always seems to be the same. You know, look at all these research companies working in antiquated ways, they've got all these research people. If you get a piece of software to do that, you know, you don't need to pay for all these people. But you know, you take all those researchers away and you replace them by dev people, you replace them by sales people, you replace them by customer success people, and then you find out there's not enough actual insight expertise within the business in order to be able to talk the same language of um the clients, and therefore, you know, these businesses that you know they they went in being led by tech people who didn't understand what these layers were actually doing. And so they they bring in more layers of people to bring in that expertise, and suddenly the whole thing supported by a cost base that's bigger than the one was on the traditional model in the first place, and they've got no route to profitability. So what we found is that you know, whilst whilst the investment money is flowing in, they can afford to support all these people and and the revenue can be going up. And at some point they said, now we've got to move towards breakeven and profitability. So if the revenues aren't going fast enough, then they they cut the people, and if they cut the people, the revenues start to fall as well. And they they it's not finding that equilibrium. You know, I tried to share some numbers that sort of demonstrate that. Those those weren't complete numbers, but um I've I've yet I've yet to find the numbers that prove that any of these businesses have got to really sensible levels of profit that you'd expect given the investments that have happened and the revenues they've got.
SPEAKER_01:Do you mean that to be a bit more specific on Introduction? Do you mean SARS, SARS for insight or whatever one would call it, like you know, research as a service, those types of businesses? Because I think there are plenty of others that are profitable and have been consistently so, but they may be more traditional agencies. Oh, absolutely.
SPEAKER_00:I'm talking about I'm talking about companies that have run the classic BCMP self-serve software playbook, you know, and and then manage to turn that into a large-scale business that can deliver the type of profits that make up for the investment that's gone in and therefore is is is at the profit level that you'd expect for the kind of revenues. Um and and I really hope that it it can be successful because that makes it sustainable for the industry. So the reason, you know, the reason I'm saying this is you want to find sustainable solutions for the industry, otherwise the the funding runs out, and these these these companies, you know, they they aren't able to continue in the same form, and it's it's not good for the industry. They've spent 10 years sucking out the oxygen and then then they can't deliver what what they're supposed to deliver.
SPEAKER_01:Yeah. It's really hard to say, actually, isn't it, now I think about it, because not many of the businesses actually have their figures in the public domain, which is probably telling to itself, in that they're still privately owned. Um, I guess you do I guess you do have someone like kind of Qualtrics, and probably a couple of others are in the public domain, but I don't know enough about Qualtrics' figures to really comment on the So the last annual figures they get because they went public and then they were taken back private again.
SPEAKER_00:When they went, I guess a couple of years ago, they were taken back private, but the last full annual results, they lost a billion dollars in one year. A billion dollar loss. You know, of I don't know, somewhere between one and two billion turnover. So I I don't know the exact figures, but but that's the kind of scale loss they're making. Now, maybe in the meantime, they've been taken private and um, you know, have have got to break even on moving towards profit. But you know, I can't think of another business that has reached the size of that business. And it and it looks like just their their losses escalated to match the size of their uh of their revenues, just like I was talking about. You know, if somebody comes and says, well, you're misreading the figures or you're missing this, it's actually profitable and successful, and that's not due to some funny accounting stuff, then that's great, brilliant. If someone has cracked it, fantastic. I'd love to hear that. I just haven't seen anything yet that looks like evidence that it has actually been cracked, and anything that I've seen suggests to the contrary that it hasn't been cracked and no one's really cracked it. And so, yeah, VCMP investing in the industry, brilliant, you know, innovating and new technological solutions, brilliant. Um, but it's like we know that that model needs to be some sort of hybrid, and that's not necessarily our hybrid model, but a hybrid model. But at the moment, the the way that the industry works pushes towards a self-serve model, which I don't think's working. The rewards are uh for behaviors that don't match the industry. I think that's that's kind of what needs to be changed. I don't blame the VCs or PEs, I don't blame the founders. That's what you would do. You know, I I did a secondary post on this saying that if you can get seven to eight times multiple on your revenues if you're uh a self-served software provider, or a six or seven times on profit if you're a hybrid, which would you do? Which would anybody do?
SPEAKER_01:Yeah. So I I do think there's a difference, which we touched on as well. I think of some of the P funds that have been successful in the space, but I think there's a difference in that they tend to be investing in businesses that are cash flow positive, already profitable. But I think what you're really driving at, Jonathan, is this specific direction where they try to push these insight businesses into that type of SARS model, that direction. And that's often problematic.
SPEAKER_00:Well, I just don't, you know, it I don't think it matches the needs of the industry as good for the industry. Uh, and I I'm gonna be the first to put my hand up and say, I do not know the nuances of the VCP world, and I might have got some things wrong, and I might, you know, I I I I know roughly the differences between the two, but I, you know, I'm a bootstrapper, so I again I don't know that space very well. Um, but yeah, it's it's more it's more about the model, self-serve software model, and the the push to go that direction when it's not the best for the industry.
SPEAKER_01:Yeah, I I think to some extent agencies have been culpable in that they describe themselves and they try to sell themselves in as data businesses where they're not data businesses, as in many cases, they're actually insight businesses with clients who want you know relatively customized, focused insight, repeating the point. I think you can use tech in all sorts of smart ways to, and AI in all sorts of smart ways to make that more efficient and as you pointed out, to analyze data in different ways and come up with things that humans wouldn't do. But it's almost like as long as the human is the end user, then the self-service element of it becomes tricky, I think.
SPEAKER_00:Yeah, yeah, and and and there's all sorts of ways that technology can play play a role in the industry. So, you know, we're exploring one model which is like an agency-driven model with tech in the background. There are other types of hybrid models, there's lots of clients experimenting with bringing technology in-house. Um, and that's a really interesting space. I think it's in a way, if you're looking at what's going to happen over the next five years and 2030 plus, it's what what happens when clients bring kind of end-to-end workflow automation in-house. Um, what is the role of human expertise? Like, take humans completely out of the loop. Like, let's say that's for 80, 90% of the processes, and then they reserve human expertise for the 10, 20% edge cases, then what does the world look like? I don't know if that will happen, whether it'll be successful, but it's just like it's it's what lots of clients are thinking about and talking about, and that's what they're saying they're looking at and doing, and some are a bit further forward than others. But then then, you know, what is the role of an agency? What is the role of a of an outside software provider? Because there isn't any self-serve to do, because it's all end-to-end optimization. There aren't any humans in the loop at all. Um, but that's when we start looking at sort of future projections and it starts getting a bit scary, or not, because it might never happen.
SPEAKER_01:Yeah, yeah, very true. Jonathan, I'm gonna spare you the quick fire round because we're way over time.
SPEAKER_00:There's some amazing answers as well.
SPEAKER_01:So well, we can always get you could we can always dip into them if you want, but maybe I'll just pick a couple. So, what have you changed your mind about recently?
SPEAKER_00:So the the biggest thing, probably, and I'll I'll I'll stick to the work thing, is that um when we started one, it was all about automation. It was all about how much of this can you automate? Can you push it to 100% automation? Um and I and I think that that it became clear that it's all about how that automation and human expertise comes intogether. And uh ever everybody's saying it's AI and human, right? But what I'm saying is a bit more nuanced than that. It's like what happens when it's AI automation, workflow automation, and human expertise and who and how many. And so that's probably why I changed my mind about it. It's like when we started one, it was all about 100% automation, humans out of the loop. Um, just as an experiment, actually, it was just designed to be. And and then it was all about actually what does the what does the hybrid model of the future look like and what's the agency of the future look like? And that that was a a shift that has really shaped everything that we've done ever since. So that was interesting. That's not very quick fire, isn't it?
SPEAKER_01:It's not very quick fire. That was fine. It was still very interesting. And so, okay, from an agency perspective, and I asked clients this, by the way, the other way around, I think as I alluded to earlier, but from an agency perspective, what makes a good client?
SPEAKER_00:I think uh it's very cheesy to say partnership, but I mean something very specific in this, in that when you're launching something new and exploring a new way of working, AI automation, hybrid agencies and things like that, you you want a client who will go into this as a partnership and look at it, say, we are learning about this together. So it's it's first of all, let's let's do work that really matters and let's focus on pilots and and projects where it can make a difference, not relegated to some unimportant stuff in the corner, but then let's let's take it seriously and learn together. And some things might work, some things won't. Um, we've got to tailor things. So it's it's it's it's not about putting in a solution that's been you know been used for 20 years and you're just the the latest application of it. You're you're learning uh at the same time. So so if everybody comes at it with that kind of spirit, I've actually found that you know you get so much further working in that way, which is which is really good.
SPEAKER_01:Yeah, I I could certainly see that. And I assume it doesn't work where you have clients, particularly in the AI space, that every time something doesn't work perfectly the first time around, they go, I told you it was rubbish.
SPEAKER_00:Well, yeah, I mean it's you you you can find what you're looking for. I mean, it our our approach is all about iteration anyway. So it is a journey of optimization. It's a very quick one, but it it's it's a slightly different mindset anyway. But then it's like it's not just it if it works or it doesn't work, but what what's right for this brand, what's right for this particular challenge, what's right for this group of stakeholders. And again, it goes back to your point about you know, not a single software solution. This is not a single hybrid solution. It needs to be tailored and adaptable to be able to work with different organizations with different priorities and ways of working, which is the heart, I think, of the challenge of that finance model and the self serve in the first place. I think that's kind of what we're trying to navigate.
SPEAKER_01:Okay, so definitely the final question now. Yes. What's your favourite book or recent book, but you can't say your own?
SPEAKER_00:Yeah, no, I definitely I my own isn't my favourite anyway, so that's a good thing. I've been reading The Long Walk by Steve King. I know it's just coming out as a film. He writes under John Backman, isn't it? And and I've read a few of him under that, like the Mr. Mercedes series, and it's really good. But it reminded me he he's written, I think, the best book about writing, which is called On Writing, Stephen King on Writing. And he says, you know, about creating a story, he's he he to I think he talks about just saying, get an interesting bunch of characters together and put them in some sort of extraordinary situation and see what happens. Now, obviously, there's more to storytelling than that, but uh people talk so much about structure and three acts and you know the five stages and all you know, heroes and antagonists and stuff like that. It's quite refreshing to hear it the other way around, say just like interesting characters, extraordinary situations. What would happen if you mix those two together? And that's the kind of premise you know that the the long walk has, you know, a bunch of bunch of kids who have to keep walking, hundred kids who have to keep walking above four miles an hour. Um if they slow down or stop, they get shot until there's just one left. I mean, you you can't really come up with a premise that's more interesting characters, extraordinary situation, but then it's it's what you do with that.
SPEAKER_01:Well, I hope you enjoyed that episode. Apologies for this slightly abrupt ending. We ran into something of an internet outage. So we didn't actually end the conversation there, but hopefully you got the gist. As previewed, Jonathan is a really smart, thoughtful guy. I have to confess I struggled in the first instance to get my head round what one strategy studio were doing. But once it clicked, or I think it clicked anyway, I can see a huge amount of potential in that type of approach, where you're training AI personas to fulfill specific functions aligned with the workflow that's common and very consistent across agencies. You're also leveraging most of existing sources of information before you commission primary research, which surely has to make sense. The point around the inbuilt disincentives and agency models, i.e., if you're billing by the hour, the smarter and more efficient you are, the less you're going to get paid, is also really telling and quite thought-provoking. I'll aim to explore that further in my next interview, where we're diving more into the world of sports research, as well as video games and esports, with a great interviewee, Nicole Pike, the global head of sport for UGov. It's another good one, I think, and I hope to see you then. Thanks for listening and see you next time.
