76. Inside the $1.1B Seed Round and How The UK Government is Trying to Leverage it

Jonathan: Welcome to The Difference Engine, the show for tech founders, investors and innovators. Paul: Just to start off, um, thanks everyone for listening. Uh, we're appreciating that the growing listenership here, in case you missed it, um, it was an amazing, uh, recent episode that we did with David Friend. That one continues to be extremely popular.

This is the guy. That sold synthesizer to Led Zeppelin, um, Hendrix, et cetera, uh, and has gone on to form the amazing hot storage company Wasabi. That's a really good listen. Um, and it's also nice to see, um, friends of the pods, which is Hussain Cassai appear on other people's pods.

Glad to be helping that gentleman with his career. It's really good to see. Once again, if you like what you hear, please literally like, uh, and subscribe to us. On wherever you listen to your podcast.

Jonathan: Lots of interesting stuff there for people to delve into our archives and digest. But what's coming up today, Paul: we're gonna be turning to the teachings of the Grand Master himself, Jeff Moore, to answer a very simple question, what is a market? Jonathan: And we'll also be looking at the ever shifting sands of tech. How can you ensure in the age of AI that your category is standing out on top?

Paul: But first, we're gonna. Law, what appears to be a new tech investment strategy from the UK government? Is it and will it pay off? Jonathan: We seem to have witnessed a step change in how the UK public sector is engaging in tech and the creation of future category leaders.

Paul: Well, what, what can that possibly mean? Jonathan: Didn't you see the news about ineffable intelligence pulling in $1.1 billion in seed funding? That's a very, very big lot of seeds.

Um, it's, I mean, it's not just big. I think it's structurally significant for the, the UK tech ecosystem. Paul: Possibly. I would say it's suspiciously enormous, uh, especially for a seed round.

Um, you know, especially at this cycle in the market when some people are talking about potential, uh, bubbles. Uh, but okay, let's say the largest seed round in European history, apparently what stands out though is who's funding it. Um, 'cause you and I are doing that. But public Jonathan: capital Paul: perhaps.

Yeah. Mm. Dunno about that. So the government, uh, through the grandly titled Sovereign AI Unit and the craply titled British Business Bank isn't just supporting a company.

It's attempting and where's this ever gone wrong? It's attempting to actively shape the market Jonathan: in the past. Um, the way that. Government has, uh, dealt usually with failing businesses and, and technologies.

It's just to chuck some money, uh, at them. And, um, I think it would be fair to say that really has never worked, um, in this case. Um, there's lots of trumpeting about a co-investment model. So instead of crowding out private investors, government is crowding them in, um, in this particular round, you've got some big, big, big heavyweights like Sequoia Capital, Lightspeed Venture Partners, uh, EQT.

Uh, based up in Scandinavia and Europe's biggest native VC and the omnipresent Nvidia joining the round as well as a few other, but that, you know, that signals global confidence. But you know, is it likely it wouldn't happen at this scale without government anchoring or was really. Government just chasing where the money's already flowing. Paul: Right.

Let call me cynical. Uh, but it seems a bit like the slightly dumpy, unpopular kid in school waving at the, you know, the, the, the, uh, athlete, the, the, the top jock at the school. Well, Jonathan: getting in with the cool kids, Paul: getting in with the cool kids a little bit too late. Um, you know, you could say that it's a classic de-risking play, but all of those companies you mentioned are non British and the UK government is effectively underwriting now.

Early stage frontier tech. Um, that's typically too spec speculative, even for the top tier bcs. There are some bcs obviously who go after, you know, pre A and seed funding, but that is risky stuff. Um, and super intelligence, um, it, it's not SaaS, right?

It's very long horizon. It's capital intensive and it's onset. It sounds like something the Americans might be better at. Jonathan: This is a risky technology play, as you say.

I mean. Ineffable isn't just building bigger models, you know? Well, that would be a stupid thing to do, isn't it? It, it's, it's I guess, pushing towards reinforcement based systems that, that learn through forms of interaction.

You know, that's. Actually, if you think about it very much in the lineage of DeepMind work, um, especially with people like David Silver who's heading this, uh, the helm. Paul: I mean, I wish, I wish him all the best as as we all should. Uh, if you think about his previous systems, like, um, you know, alpha Zko Alpha Zero, they weren't just trained on SAT data sets.

Um. You know, they did reinforcement learning and they learned through self play. Ineffable seems to extend or trying to extend that paradigm into more general environments. I don't think it's the only guys doing that by any means, and that's where this narrative of super intelligence starts to feel.

A little bit less like hype and more like a potential research trajectory. No. Less risky though. Jonathan: Yeah.

But of course research is where government is, um, comfortable because it can exist in a coast, little world, a real self-reinforcing world of the public sector. Um, civil servants, academics, kangos, and so on. Um, but it's quite clear that they are in a different world here They are. Um, coexisting as an investor with some red in tooth and claw investors.

Um, but you know, I, I still, still think calling it super intelligence. It's, um, seed stage is, um, what should we say? Um, ambitious Paul: and the ambition here is part of the funding logic clearly. Um.

Some could, some would say, you know, if you've got all of this money for funding stuff, why don't you just plain chuck it into defense tech? But governments don't really step in for these, uh, incremental gains. They're hoping to get in some sort of strategic advantage again. Makes you think why defense tech's not so good.

Uh, AI is clearly, um, now geopolitics and you know, the UK is sort of. Got its nose pressed against the window. And it wants to ensure that companies like this scale domestically, uh, and hope that they're not, uh, absorbed or relocating. Dare I say, like DeepMind, Jonathan: when you've got the majority of the money in this particular.

Investment being put in by companies who are, uh, headquartered in other parts of the world. Which I guess brings us back to the basic policy here. Um, the British business banks, additional $20 million, Paul: did you say? 20 billion?

Jonathan: No. Million. Million. Oh dear.

We're talking about small potatoes here. You know, these, these are really small potatoes, um, relative to the total. I mean, yes, it's symbolically important. Um, and I think what they're trying to say here, um, is, uh, we're in this for the long haul with the big boys.

Now they could be absolutely conning themselves. Paul: I mean, it does align with the talent retention narrative. Um, because as we know, and, you know, um. We had, uh, folks on here talk about the, uh, amazing good value that the UK used to be in producing top tier AI researchers.

If you think about Imperial UCL, Cambridge and Oxford. Um, but historically they've struggled to keep companies in the UK as they scale. And now of course you've got, um, the likes of, of Anthropic, um, where former PM Rishi Cena works. Offering very large, uh, six and sometimes seven fill figure salaries and, um, making that affordability of UK talent a lot less.

So hoping that this kind of financing could change the equation, could be wishful thinking, Jonathan: could be indeed very wishful thinking. But I guess if we're gonna try and be positive about this, the size of the round. Does set a precedent and if this model works, we might be seeing more state backed mega seeds in deep tech and that could get European startups off the blocks at the pace, which may stop them being roadkill for us competitors later on. Um, but again.

Uh, look at, look at the investment, the cap table on this. And, uh, I see a lot of hedging of bets going on, Paul: right? And the risk, uh, even if the bet is good, is concentration. You put that much capital into one company so early, you're making a very specific.

Uh, bet on one technological path as opposed to perhaps funding a defense sector and, and spreading the money around a little bit. Jonathan: Yeah. Well, we hope that, you know, even from global standards, this investment, which is relatively quite small, it's actually a bit of an outlier. Um, for what the, um, sovereign AI unit has actually been doing.

It's, um, it's backed eight companies so far. Paul: Another, another case, uh, from the, uh, from the same funders so to speak, is Lokey Labs. I hope I pronounce that right. So, LOC ai, uh, this is the self-styled, um, itself, styles itself as the UK Sovereign AI company.

Um, it's run by a, a, a young American chap of course. Um, but, um, the UK's. Sovereign AI company. I don't think that cuts it as a category either.

Let's think about British Coal. British Steel, British Rail. Hmm. British Airways, British Telecom.

Uh, and as a shareholder, I can tell you that's not having good time. So I'm not sure that British or even UK as in the UK sovereign AI company is the modifier that's gonna add value to your category globally. Um, now these guys recently announced in conjunction with SIBO who. Likewise claim to be the British Sovereign Cloud provider, the creation of, of Mercury.

Not the one I used to work with, but the self-styled UK's first AI assistant. It might just be me, but this just sounds a little bit too little, a little bit too late. Maybe a dollar short. Um, but maybe I've got this wrong and there is a defensive moat.

Here's how they describe themselves, and you can tell me if this is, um. Um, you know, I think this is just, uh, it is truthful, but their point of view is for a decade, the UK has operated as a digital tenant reliant on foreign owned hyperscale cloud providers and AI model developers for its most sensitive computational needs. I think we'll agree with that. They go on with the POV to say, these companies talking about the, the digital tenants are subject to control from their host governments, and we've seen how they've been required to remove services from individuals who have criticized these governments.

And the Mercury Series aims to end this dependency. I mean. That's a big claim, uh, with all of the, all of the Fiori here in the UK about digital id. Um, claiming that you're holier than thou?

I'm not sure. Not sure? Jonathan: Yeah. Just back to the nub of this, if, if ineffable delivers, I mean even partially, perhaps it could justify the approach, um, you know.

Looking at history, which I always like to do, you know, breakthroughs in science, medicine, engineering. These are the kinds of high risk, high reward, but socially relevant spillovers governments are willing to fund. Um, but it is this ultimately more about national pride more than you know. Seeking the cracks between categories where, you know, real value is, is created.

So, so in a way, the, the, the, the story isn't about funding. Um, it it's about how UK industrial policy or lack of it has fared. Uh, this is industrial policy in action. Tailored for an AI era.

This isn't just the UK doing a tactical funding of a promising startup. It's trying to buy a strategic position in the future of intelligence. Now, as a 1960s labor Prime Minister Harold Wilson would've said, this is forged in the white heat of the technological revolution. Um, which was a phrase he used as he was.

Trying to lead a previous attempt to modernize British industry through science and technology. Hmm. Now let's look back on that quickly. Um, that created the likes of gc, British Aerospace Corporation, ICL, um, you know, hugely technological products such as the TSR two Fighter and Concord.

But I think, you know, certainly in the case of industrial mergers as a way of regenerating competitivity, it was always too little. Too late and certainly not radical enough. Paul: Yeah. So let's, and let's be honest, we've, we've actually hoped and wished and hoped for an industrial strategy.

It's just that there seem to be big bets, not totally owned by the, the, uh, government. So they're not really bets, they're sort of hedged with all the usual suspect us. VCs, um, you, you, this is something that we, we think directionally is right. The execution is, is what I'm concerned about.

Jonathan: Yeah. And I think it, it is an attempt to, to buy a place at the table. Um, I'm just wondering if, if it's a table in another room from the table everybody else is at. Um, so I think in this case we should hope for the best, but of course.

Prepare for the worst. So just in case you hadn't noticed, the world is currently changing very rapidly and the world of strategy and technology category design is doing so as well. Now, that discipline, we've always understood as creating and owning new markets rather than competing existing one. One is, is also evolving quickly as.

Tech tectonic plates shift its power centers. Paul: Yeah, absolutely. It's changing and the upside is there's a whole lot of faster ways to get your category out there publicized and placed with the public. So we thought we'd, uh, take a look at the way that companies can execute category design in 2026 and note some of the several ways that, uh, category design has shifted.

Jonathan: I think the, the first thing, um, we have to think of, which is really a, a macro level change is from. The idea of category creation around products to problem reframing and narrative ecosystems. Now, now let me explain a bit about this. So, classic category design is, is focused on naming a category.

So, you know, CRM cloud and something like that. Um, now we've never really believed it was just about that, but now. With the changes we're seeing, the emphasis needs to be broader. So what you need to be focusing on in categories design currently is reframing the customer's worldview, not just the product.

In order to do that, you need to build a multi-layer narrative. Um, narrative is how you talk about your category and how you talk about. Your category amongst influential audiences, for instance, media analysts and other communities that support you. And that means also supporting that in turn with a full ecosystem.

Now, the ecosystem starts with the content, but it means that you need to engage your partners with it. You need to, uh, engage other creators. And crucially, you need to engage. APIs.

So modern category leaders don't just define the product. They define. Define how people think about the problem itself. So we're seeing a shift here from naming a space to owning the meaning.

Paul: Which neatly segues into our second point. If you notice what we're talking about there, when we're talking about media analyst communities, those are all people. And, um, for the time being until we have agent to agents, uh, making decisions and buying decisions, we are talking about people buying into your category. People being the, the, the friction, carbon based, heavy friction, uh, folks that we all are.

Um, and we've talked in recent. Uh, episodes about how AI is fundamentally, fundamentally changing how categories are one and created. So, um, one of the things we talked about is brands have to, if they're looking for discovery, it's no longer enough to buy a whole load of Google ads. Um, Google ads have been replaced by AI discovery, and that means pandering to LLMs, to agents, to co-pilots, et cetera.

Uh, using things like generative. Engine optimization, also called AI optimization, a EO. This, that category itself is being, uh, argued about, but this means that whatever you desc, however you describe your categories, it's gotta be machine legible with clear concepts, um, and structured data in your content. And crucially.

Consistent positioning. AI agents are increasingly the front ends. They're decision makers or recommended for the, for the, for the, you know, the long list, if not the short list. So it's important that your category is designed to be understandable not only to humans, but to algorithm.

So there's, there's a shift here from just persuading humans to having. Machines help humans understand what your category is, Jonathan: something which leads onto that, which brings us up to number three, which we've called the idea of micro category fragmentation, right? So the big categories are defined in ai, um, but its impact means tech markets are splintering. So we are seeing the development of highly specific micro categories, um, with niche use case positioning.

So for instance, rather than just ai, it's AI for, so AI for legal operations, not just the idea of. AI tools. Um, and we're also seeing the emergence because of, you know, the, almost the DIY nature of AI that we're seeing, the emergence of community led segments. So these are what we call micro-communities, and they themselves are driving adoption.

And this is a really, really important point because it shows about a, how AI is starting to drive a shift in power. Um, it, it is partially driven by saturated markets and. Associated algorithm algorithmic feeds, but it means that that's pushing individual and group hyper relevance over mass appeal. So the shift here is from single category king.

Now your company. You used to be thinking, right? Which category do we want to dominate? You know, we are Salesforce, we are gonna dominate CRM in the age of ai.

That's gonna be no longer true. You should be looking for a portfolio of niche dominance, Paul: right? So solving a bunch of problems, uh, under one category. And I think that brings us onto communities, uh, who are increasingly important.

Everybody knows the open source. Um, folks have known this for ages. Communities are massively important. Uh, and they bypass what used to be a top, very top down process.

So you, um, the high priests and the analysts, um, the, the important, um, you know, journalists, uh, and founders, uh, increasingly as they became sort of mini celebs, uh, that used to be the top down process to get your category understood. You had to have those guys. By now, you can do a lot with creators and communities. So for instance.

Reddit amazing, um, place to, to talk about your category. Um, you see some of this in, um, some of the tech papers that have comments. So for instance, the register is a good place to get your story out, and what you're doing there is you're co-creating narratives with these communities. So with the users, um, in B2C, they call it user generated content.

That's why you have all of those atrocious influencers who came to allegedly seem to be running over each other in, uh, London these days. And, um, what was, what was shifting from, from these top down influencers and top down, uh, celebs and media and analysts is more bottom up the brands, uh, learning to seed control to the communities below them and let those creators interpret and evolve your category authentically. That's really winning when you get those guys to start talking about it.

So you should deliberately move from broadcast to co-creation. Jonathan: And once those, um, human beings get involved with the pesky machines, there is a need for human and emotional differentiation in what is gonna be an AI saturated world. So as AI commoditizes features, differentiation is moving towards, um, factors such as emotion, storytelling and identity, the stuff that people can really grip onto. Um.

To help them do that. The design should be, uh, human, human centered. Um, and there's a little trick here, which is getting authenticity into your product proposition. Um, from imperfection.

You know, people are becoming very, very good at spotting what is entirely machine. Produced content. Um, and yeah, this is, this is the stuff too that has been in some of the best brands in the past. You know, and Apple are very, very good at this.

Is that brand as a belief system. Not just a solution. You know, if you buy an Apple product, if you pay the premium from Apple products, this isn't just about giving you a platform to do your word processing on. This is a belief system about design simplicity being part of a group of people.

So that is actually going to be important, that there clearly is a backlash coming against these overly polished AI generated. Experiences, um, because it makes us feel a bit robotic. And what people want is a more human, tactile, expressive approach. So what we're seeing here is, is a shift.

And you know, this is anybody who knows the history of success in tech, um, should know that, that it's not always the best. Product from a technological point of view that wins. It's the best distributed product that wins. Um, so the shift here we're seeing is, is from functional superiority, which ne should never really have been your top priority in category creation to emotional resonance.

What does this do for me? How am I part of this? Paul: Yeah, and this is the irony. Uh, with all of this AI assistance we're getting, you know, user experiences are user experiences, people's experiences, uh, not product led.

And actually this is pretty positive. 'cause as you rightly say, most, uh, most tech companies are, are engineer led. They think about how to engineer products. That leads us into the sort of negative circle of, um.

Better and not different. Whereas we think everything should be different and not better. But user experiences are very different. They're individual and they are end to end.

So when you are trying to set out to solve something with some software or, or hardware, uh, solution, the way you experience differs from other folks. But it's completely end to end. I just came off a call this morning. Try explain to a supplier that their database is very nice, but I would like to take that out and do some other work with it myself with Claude on ai.

So it's important that we think about daily workflows, uh, if that's the way your product works and how, um, the outcome is what I'm, I'm interested in, not just the time that I'm in your quote unquote product. And I'd like it to be seamless, and I'd like it to be invisible, and I'd like it to be more like a utility, because consumers prefer solutions that fit into their routines and reduce friction, not just, you know, accessing a product. Figuring out where on the menu I've got to get to working the way the product works.

And so experience led categories, those where you can talk about fulfilling, uh, goals and, and creating outcomes is one of the most interesting shifts we're seeing as we move from product category to experience category Jonathan: and something else which should naturally follow on. From this is, you know, what, what are the real category fundamentals now in the era we're living in? Now, clearly data is a category, fundamental. I mean, it, it's, it's part of virtually every proposition that's worth any amount of money.

But there's two other areas which are coming in strong and we believe are gonna define categories going forward. And those are trust and ethics. So data trust. Ethics as category fundamentals.

So if you think about category leadership in tech, particularly in AI driven product, what we want to see, what we think, um, category leadership, uh, is increasingly gonna depend on is those trust frameworks. And by that we mean clear, clear, transparent, privacy, explainability responsible. Data use Absolutely. Front and center, you know, and it's something we've, we've seen, um, emerge as some of the realities of the way that some of the major social media, uh, companies have behaved, if not all of them.

I was trying to be nice there, all of them. And responsible data use has to be backed up by transparency to the users about. How systems actually work. Um, you know, pay attention LLMs at the moment.

'cause we don't believe you are being particularly transparent about that. So, for instance, high quality, trustworthy data is becoming foundational to competitive positioning. I'm gonna say that. High quality, trustworthy data is becoming foundational to competitive position positioning.

Now that means there's a shift from innovation first to trust first innovation. Yeah, this should be a natural trend as people become more aware that tech innovation for innovation's sake is fraught with issues when it actually hits the market. Right? So this.

Is intimate to the idea of a, another thing we're gonna talk about in this, um, episode about self-referencing custom behavior Paul: category design is all about design, right? So it's, it's the two things that come together. And it's true at the tech product level and the. In a purest sense, but, um, design is, is, shouldn't be and too often has been in tech.

Something that is layered on after the product market fit is found. We think it's central to category creation. Certainly. Um, you know, companies like Apple have always believed that to be the case.

Um. And what's great about AI is it's a very good co-creator is excellent at, um, names and, um, coming up with ideation, but it's less good at shaping experiences and narratives because it can't know what we would know as humans. Uh, and how we like things designed. We, you know, our flaws make things.

Um, it's necessary for us to design things. And we talked earlier about immersive, interactive and emotionally engaging interfaces. Being essential. So is strong visual and interactive design to reinforce your category identity and work with the narrative that you are saying.

So if you say you're gonna make something easier, make that very clear in your user interface. The shift here is from branding after the strategy is completed to design as part of the strategy Jonathan: again. Linked to this is, is what we've called the tension between speed and legitimacy. Now, let me explain, um, th this might resonate in particular with our European audiences because there's a growing split, uh, as we see it between fast moving, hype driven, defacto category creation.

I drive out, drive out there, dominate all the conversations around the marketplace. Put huge investment in um, two. A slower evidence-based and credibility driven approach. Again, this is linked to the idea of, uh, co-creation.

And if you're gonna win in this environment, you've gotta balance the vision, which is absolutely important to demonstrating that you can drive and dominate a new category to the proof, right? So proof, clear, measurable value. And impact. Those two things are closing down.

You used to have a lot longer to deliver the proof you don't anymore because you are immediately engaged with audience. Paul: Oh, just to add to that, I mean, the speed of news means your old news very fast. You saw that with the, with the mythos. What I call now the post mythos era.

I mean, things move so quickly that you're gonna make a claim, you better back it up and fast. Jonathan: So, I mean, if you wanna think about this in a, in a nutshell, the shift here is from move fast and define to define and validate. Continuously. Paul: Hopefully that's been helpful.

I mean, uh, we're a little bit academic here, but, um, it's because we care passionately about people and we know people are creating categories and are a little bit confused about how AI interacts category design we think is no longer about. Just creating a new market from like force of Will, it's more about intimately shaping that perception of the category across humans as well as machines and not forgetting, um, collections of humans or communities, communities simultaneously.

So humans. Machines and communities all at the same time. And if you're no longer working on your category stra strategy today, um, the key questions are no longer, you know, what is the category we're creating that that's been replaced, right? Jonno, Jonathan: what you need to be thinking about now is how is this category understood by ai, who is co-creating this with us?

What emotional and experiential space do we own? And the most existential. Why should this category exist at all? Paul: Indeed.

And we can't wait to see what new categories are created in this a IH. Jonathan: So one thing, uh, that Paul and I share is a love of Jeff Moore. Particularly his knack of turning the realities of Silicon Valley into learning that can be applied elsewhere. I mean, this is so much better than academic studies of existing published papers that haven't been relevant for at least a decade.

I talk from experience, um, you know, after an early academic career of doing just that. I had the pleasure of helping launch Jeff's inside the tornado in Europe, um, which was probably a couple of decades ago, but. Anyway, recently we're, we've been coming back to, uh, Jeff Moore's definition of a market, and, and that definition is a self-referencing group of customers. Paul: Yeah, I, so Jeff, that's interesting.

Uh, Jeffrey Moore, I first came across his work, um, when a very kind early client of mine sent me a, a draft of, he'd been to one of his events and sent me the, um, notes from the book, the, the deck, and literally wrote on it as I was studying an MBA at the time. This is worth an MBA. Uh, and whilst it, it, I wouldn't say that I plagiarized any of it. It was very helpful, I have to say.

Um, the brutal implication in Jeff's latest tome is that, um. You know, the implication of a market depends on if people speak to each other. Imagine that's an old fashioned notion, so you're not actually in the same market, um, and, uh, unless people speak to each other. And then if you're not in the same market, the chances of building a category are very small.

Jonathan: And when you think about that. It, it's more specific than most people would realize. A real market just isn't a segment on your CRM, you know, it has a few core properties. Oh, Paul: like what?

Jonathan: Ah, I'm glad you asked that. Well, first, you know, it's a defined set of customers, um, with shared needs, you know, and second they. Have to have a common use case and that they're buying for similar reasons. But, but the big one is the self-referencing behavior.

They communicate and influence each other's buying decisions. Paul: Right. And, and that's no doubt one of the hardest things about building a category after you've had the fund with the name and the, and the POV is, you know, how do you figure out how that happens, becomes viral if you will. And I guess that's sort of where word of mouth comes in.

Jonathan: Yeah. The reality is. The customers you really want to get to behave like a herd. Um, particularly when that customer base is moving from the innovators and early adopters to the early majority.

Um, you know, it's this classic crossing the chasm stuff because people want reassurance from peers. Before they commit. And it's that reassurance that starts to deliver the really steep curve. Paul: Yeah.

And, and so that's ties directly to this word of mouth notion and why word of mouth whilst it's been forgotten, I feel many years because of let's, um, let's just call it out, Google ads and product-led growth, uh, and people thinking that they can sell things just off of a website. Word of mouth, which involves humans, let's note, is very powerful. Um, marketing cannot of course reach and speak to everyone individually, so you do need a network effect. But it's word of mouth we're talking about.

Jonathan: Yeah, exactly. And without that, in internal communication, there's, there's no leverage. You know, the, the, the message simply doesn't propagate, which is what you want. Paul: So, um, let's test this out.

Um. Let say the ICP for, um, for our company is, uh, a hundred million to $1 billion tech companies. Jonathan: Oh, classic segmentation. Yeah.

What would sales ops say about that? Paul: Well, sales ops would say that, uh, uh, a French company in LA be France and, uh, an American company from the, the good old, uh, Midwest, that's the same market, same range, same type of company. Same ICP? Jonathan: Yeah, sure.

On paper. Just on paper it, because that approach basically drives the same firmographics. Now firmographics, if you're not familiar with that, uh, probably, probably a port manto word of some sorts. Uh, it's the descriptive attri attributes of firms such as industry size, revenue, and location.

And it's been classically used by B2B marketers to segment markets. Target our deal customers, you know. Sort of the corporate equivalent of demographic data for individuals. But, but what that means in modern marketing systems, it's the same bucket in the data.

So, you know, this is what, what you would get if you just said, right. We're looking for defense contractors in. Particular turnover group. Paul: Right?

So very little really in common, sort of like an AI slot before AI slot. In reality though, that's not, that's not the case, is it? Jonathan: They're not generating the same, same word of mouth effects because the companies may be in Paris or they may be. In San Francisco.

Paul: Yeah. Dealing with a different set of realities. I, I totally agree. And, and crucially they ain't talking to each other.

Jonathan: Yeah, sure. Exactly. I mean, it's a different geography, it's a different language, different culture, different networks, very different politics currently, if they're in defense. So sometimes these firms won't communicate because there might be some, some issue of competition between each other, but sometimes they literally.

Can't communicate Paul: if these customers aren't speaking to each other and they're not refor reinforcing each other's ideas, they're not reinforcing each other's decisions. There is in fact no heard behavior and therefore no category momentum. Jonathan: Yeah, and, and just remember this is about human beings interacting with e with each other. So, you know, we know there are many ways.

That you can interact. But the most assuring way to interact with somebody is to look them in the face and see the whites of their eyes and, and usually, uh, tell whether they're, uh, telling the truth or spinning you a line. Um, but you know, this lack of communications, this lack of heard behavior is a real problem. Particularly when you think of it in terms that, you know, Jeff invented, which is the idea of crossing the chasm Paul: and get cross the chasm and get inside the tornado and, and get carried up in the, um, momentum of the category that is in fact the whole point.

Um, and there is a gap classically for those who haven't read, uh, Jeffrey's amazing tones. Jonathan: She recommends you do. Paul: Yes, because the man, the man is responsible for the technology, uh, adoption lifecycle or talc as we like to call it. And the gap in the talc curve is between the early adopters and the early majority.

That's where the tornado all sort of happens, Jonathan: right? And, and The's whole point around this is that you cross this chasm by dominating a small, tightly connected self-referencing niche in a. Military terms, you'd describe it as a beachhead and you build out from there. Paul: Uh, and so in that sort of group, uh, o of self, self-referencing sort of herd of early adopters becoming early majorities, that's when you can, you have the shot, let's say, nothing stronger, that you have the shot becoming the standard, the category leader.

Jonathan: But if your market is actually fragmented. Your market is, you are talking to a lot of people who aren't connected. Um, you never get that compounding effect. Paul: Yeah, and this is the, uh, drawbacks too.

There's, there's a, this concept of the whole product, which is everything you need to make a category succeed. So that might be, um, you pre-sales, it may be, um, references of other people. It may be, um, may be the services that surround. The product that you are selling or the products that surround the servicing you're selling, you don't get the compounding effect.

You will miss the, the opportunity to spot that whole product opportunity. Uh, and, you know, then you won't optimize the category Jonathan: well and, and you won't get the revenues. And, you know, those self-referencing groups tend to converge on the same standards too, which is really important. Not just the core product, but all those supporting things around it.

Like the services, the integrations, just everything. Paul: Yeah. And so again, group not connected. You won't get the alignment.

Jonathan: That's why I particularly like, uh, Moore's definition. But I, I think we've been thinking about this. I think it could be extended a bit to offer some practical advice as, as how to take advantage of, of what we think is, is a reality. Paul: So how, Jonathan: right, so you, you're actually trying to do this, you're actually trying to get this, this leverage from a, a self-supporting community.

So one of the first things you need to think of in this day and age is, is do they follow the same influences or thought leaders, you know. Even if they don't talk directly, they might still be shaped by the same voices. Yeah, Paul: yeah. And of course, as two massive proponents of earned media we're very in, very interested in, uh, talking to the same people and people who follow the same influencers.

Even if it's a TikTok influencer, uh, that gives you a way to point what it is you're trying to do and build a category. And so shared inputs, um, meaning you are all reading the same and, and, and watching on YouTube, the same influences can. Partially substitute, only partially substitute for this direct interaction in the market that that Jeffrey's defining. Jonathan: That's certainly true and if you think about what we've been saying on recent episodes of the pod, this, you know, listeners should be thinking, hang on a minute, this must also extend into L based search these days.

Because that's what they're picking up. Um, and the second thing, uh, you have to think of, and this is incredibly retro, is, is do those people you're trying to target attend the same events? You know, is there a conference or forum where they could real realistically, physically, or virtually bump into each other, Paul: physically meeting people? You know, you can't.

You don't switch it on your own. There's the interaction before, there's the interaction after there's all of the, um, you know, stuff that we, humans are great about, leaking, about our families and our setup, et cetera. So physically much richer experience and, and these all create. At least the potential for connection around the category.

And of course word of mouth, Jonathan: but, but we've gotta be clear. Word of mouth, you know, it, it's in the name only happens when people actually Paul: talk. Yeah. And that's just more important because as we said, uh, the market is shifting from.

Uh, see it, want it sort of, um, you know, quick buys via product led growth. Uh, and, um, those products will need more convincing as people have longer LLM searches and, and ask for something more precise. Um, so yeah. So we should be clear that, um, marketing is shifting from this paid acquisition via boring old Google ads to something more.

Richer, something richer, more personalized, and that's where category awareness and brand recommendations become even more powerful than ever. Jonathan: Yeah, totally. And, and if you, if you're building a, a, a category leadership position, you are relying on people discussing you and in their communities at meetups in their user groups. Perhaps even in the pub, Paul: you know, at the bus stops, et cetera, in the sports grounds, in the bingo hall, wherever people meet.

Um, and that brings us right back to geography. Jonathan: Geography is back into focus and, you know, we've, we've harped on about geography and proximity being, you know, the key thing that has enabled Silicon Valley to keep ahead of everybody else. So, you know, geography is back in focus and what is important is the local density. But I'd say local because we are making the case here for physical interaction, but density, whether that's online, it matters a whole lot more than people think.

Conferences, associations meet up. That's the infrastructure for word of mouth. Paul: Yeah. And we practice what we preach.

I said we, we run our, uh, our own AI dinner. Uh. Earlier, uh, the, the fantastic success, the takeaway here isn't to be dogmatic about Moore's definition of a market. Jonathan: No, no.

It's, it's just a great sanity check. Paul: It is a great sanity check. And, and what we, as with all of Jeffrey's work, it's supposed to be thought provoking. It's supposed to, to make you, uh, you know, have a conceptual model of how you build out a category.

Um. So when you say a great sanity check, what does that mean for you? Jonathan: So I think, I think, you know, when you define your ICP or target a market, ask a simple question, Paul: do these people actually talk to each other? And if so, how?

One has to imagine face-to-face conversations matter, at least as much, if not more than online, uh, in this case, as does physically being present. Jonathan: Yeah. So if you know, if the answer. So the question is no.

The audiences I'm trying to address and coalesce into the audience that are gonna give me category leadership, the answer is no. You probably don't have a real opportunity for category domination. All you've got. It's a lovely set of numbers on a spreadsheet.

Thank you for listening. If you wanna learn more about category design, head to be categorical.com. If you need help designing and dominating your category, then get in touch.

Contact details are in the show notes.