68. The Definitive Guide to Category Leadership in 2026

0:00Welcome to The Difference Engine, the show for tech founders, investors, investors and innovators.

0:08all right jonathan what's coming up today most of you listeners have been on your holidays recently and so have we now the problem when we go on holiday is we start to do some thinking and that can be a bit dangerous indeed today we're delivering the definitive guide to category leadership in 2026 we'll dig into all the things that can go wrong on your company journey and hopefully give you the tools to avoid them. We'll also review the latest relationship status between category creation and VCs. But first, grab a pad and pen. It's a new year, a new landscape, and there's a new set of lessons for anybody serious about owning their category in 2026.

0:53So, happy new year to all our listeners. We hope you enjoyed our predictions for 2026 in episode 67. So we thought it'd be good to kick off January with what category design feels like right now. What we see are the challenges to organizations in establishing and maintaining category leadership this year. Yeah, it's extraordinary. It's episode 67. Happy New Year, everyone. And like everything else in the world, it seems the tectonic plates of category design are moving. And for instance, we're entering what we believe is the post-SaaS world, where it's fundamentally important to not just think about naming a category, that's the easy bit, but having a strategy to re-engineer ecosystems around it.

1:39Let's face it, SaaS, in our view, has become lazy. Back in the day, you just found a niche, maybe it was paper-based, maybe it was spreadsheet-heavy, and you just developed a front-end, buy some cloud space and exaggerating for effect, away you go. Latterly, it became a bit of a game of fast following. You want a new CRM for your niche, try us or we're better at sales pipelines tools and everybody else and even the market leaders. There was no need for new categories, just rip off an existing one. But we think it's not just about anymore describing the problem you solve. It's how you define the entire future problem space before even competitors think about it.

2:24And that's what it's always been really, right? Yeah, totally. I mean, you know, back in the early 2020s, category design was still way too often thought of as marketing communications. Basically, that means messaging and branding. That's never, never, ever what we thought. you know we're clear it's always been about strategy. Yeah sure and when we work on category design we're always thinking about product architecture we're thinking about partnerships with other folks in the ecosystem we're thinking about how developers and users interact we're also thinking frankly these days about where the boundary is between software and people and processes.

3:11And of course, in this AI world, that means AI driven discovery layers. And in 26, what we think is if you need to succeed, you can't define a category without thinking about how both the people and the machines will find and engage with your category online. We'll be talking a little bit about that. And in fact, we believe the vast majority of the AI platforms, which will succeed, will be the ones that are looking at redefining actual processes again, different not better and leaving aside some of the b2c use cases um like grok let's not go there right now which are very self-serve what we've seen is the most productive business to business ai tools do require humans in the loop at least to customize them or to set them up to their specific requirements because every organization is different that's exactly right you know that That's why categories like adaptive process orchestration, or frankly, many of the other acronyms we came up with in episode 67, they're not just names.

4:17They actually reflect emergent buyer behaviors. And those behaviors are being driven by AI automation and dynamic workflows in restructured enterprises. Okay. And as we mentioned before, the SaaS world feels the pressure of continuous category creation now more than ever. I don't know about you, and maybe it's just because of the break, but we've saved probably thousands by deleting and reducing subscriptions for online services that we no longer need or we underuse. And that's just the way it's going to be for existing category players in SaaS. Subscription or ARR models no longer guarantee either investment by outsiders or market leadership, only continuous category innovation will be the way forward if you want to maintain your leadership.

5:07Yeah, right. The days of build it and they will come are absolutely over. You know, now category success requires you to shape customer expectations, you know, drive ecosystem standards and orchestrate AI enabled value chains that extend well beyond your app. Yeah, and if I may, I mean, this is, you see a lot of value extraction in late markets in tech, where you have consolidators come in, and they just milk the existing revenue streams, investing less and less over time. And, you know, obviously getting fewer new customers, but that's okay, because they have an asset stream, which yields a return, which they like.

5:48And this is very appropriate to enterprise software strategy. at this stage in the maturity cycle certainly of sass companies can't just insert themselves into these legacy portfolios they have to make their own case they need to articulate why a new category matters two very hard people to convince cfos and ccos people juggling digital and ai transformation they've got a whole world of pain you know they've got people to manage they've got ai risk to consider they've got compliance to figure out they've got of course cost pressures all at once in a world of volatility, uncertainty, complexity, and ambiguity.

6:25It's a lot. Right, so that's quite some challenge. So how do you see enterprise software adjusting to this new category design playbook? Well, there's a few of them, and maybe we can talk about these. So one of the first things we see is outcome-centric categories. It's a natural successor to SaaS, if you will, because it used to be hard to set up software, but now software as a service makes it very easy. delivery for your software. But enterprises no longer buy tools or services. The days of wagging your credit card around has gone. They are into, quite sensibly in conservative times, predictable outcomes.

7:05So category definitions revolve around outcomes that imply benefits, as they always have done, really. Things like zero friction onboarding. I can hire people and get them productive very quickly or predictive business continuity. I'm not going to suffer the sort of outages that we saw in last year, you know, really crunching downtime. As we said, in the world of AI, this may mean a blend of software and services, meaning that the reports of the death of McKinsey may be somewhat exaggerated. Indeed. And something I'm definitely seeing in the world of M &A is the importance of composable platforms.

7:44It's not just the enterprise buyers it's the strategic investors looking for acquisitions currently expect interoperability so the categories are now defined by integration and extensibility now at this point we're making early on about you've got to have something surrounding your app and and not just its standard capabilities you know this is all about context not just content if you think about it. But you should note that platforms composed of many AI inputs, LLMs, agents, etc. bring a whole pile of new risks. Yeah, there's a big role here for advisors and for guard railings. So that's interesting.

8:26So you're seeing in the M &A world, if people are not thinking about how they fit into other ecosystems as they prepare for sale, that there's an issue. There's a very, very big issue there. Yeah, and I guess related to your last point about the big role for advisors in guard railing, AI assisted adoption is something that we all need to consider. Tools that come with built-in intelligence that adapts to the context you just talked about. So not just simple automations, that's, you know, table stakes. Assistance that teach workflows as you go. And that changes SaaS categories from automation tools to something maybe more like an enterprise AI navigator.

9:07the self-taught version of the outcome-centric solutions we just discussed. So, you know, this is perhaps the biggest challenge facing enterprises because they literally need to rethink the career trajectory and training for all of these brand new AI-assisted professionals that are now in their workforce. Just like accountants had to move from Excel spreadsheets to Xero when everything moved to the cloud, it's actually fascinating times. It is, and it's times for a bit of growing up and growing up quick. And that's no more exemplified by software businesses in general, even at the early stage, needing to think like platform companies.

9:46That means that, you know, if you have category ambition, your category design has to answer a number of things. So you've got to ask yourself this if you're at this stage, you know, how do others build on top of us? how do we become indispensable in a stack not just installed ready to be replaced as the stack is inevitably rationalized yeah yeah and it's okay uh if you want to build strong foundations for your your category um if you want it to succeed you can't define a category without thinking about how the peoples and machines will find and engage you in the first place you know that's what building a category is all about and that used to mean um let's talk about it seo um and that's just because google ads and a tiny tiny little bit of bing uh used to have the monopoly or duopoly on search which was traditionally what marketing folks at least were obsessed with for the last few decades um and that's where most of the marketing spent where went that was what was behind product-led growth and to me that was a bit of a mistake um ai last year completely transformed search from keyword matching to semantic and intent forecasting at a higher level, literally more words, more context as to what we were looking for.

11:09What does this move away from SEO mean for categories in your view? Well, it's going to be an absolutely huge shift. You know, we're using the term SEO here, but such has been the change. with adaption for AI, new names for this category are emerging. EE, and we've used this one before too, generative engine optimization, geo, or answer engine optimization, EO, or AI optimization, AIO, or our favorite, AI visibility. No acronym there. I mean, to be clear, therefore, here is the task of making content easily digestible and citable by AI, like ChatGPT or Google's SGE, in contrast, and this is a really big contrast, to traditional SEO's focus on ranking links for human users.

12:03Okay, and I think this is a message for all brothers and sisters out there in the marketing world. You might think, in contrast to traditional PR or earned media, that SEO is going to sort of like fade away, but you'd be wrong because earned media with this editorial filter is a key input to this new world of geo a call it what you will chatbot citations and it's funny that after all these years amusing to me that slinging uh two keywords together is finally found out to be not as trustworthy as giving those searching the net actually what they want um and that often involves and again irony upon irony the humble human edited story right so the big message here is now search is generative people aren't searching for phrases they're asking contextual questions and expecting eloquent and dare we say synthesized answers now that means ai agents will recommend software categories yeah imagine an AI agent advising a CIO on, say, best system for distributed workforce automation.

13:15It won't return a list for consideration. It will do the thinking and position a category and specific vendors as an integrated solution. Now, this is hugely important as it changes SEO from ranking for keywords to shaping the semantic space around your category. Right. And also a nuance here and something that we know from the work that we've done in AIO, there is a bias towards novelty, towards innovation. So playing in a category that you've lost, essentially you became the third or fourth player in it, will obviously disadvantage you when it comes to AI search, whereas novelty and new categories will be higher ranked for these AI overviews.

14:08So it sort of pays to be different and not just better. And SEO strategy in 2026 looks more like, here's a category, potential category for you, linguistic engineering. Like that, good. Yeah, and so optimizing for narrative authority or credibility, not just backlinks and cryptic goddamn keywords. You know, optimizing for AI prompt relevance. Training models that associate your category, potentially a new one, with high-value enterprise buying intents, which of course will change quickly. And that means that you need to monitor them. It's not fire and forget with a couple of like, you know, stodgy old keywords.

14:49And structured data ecosystems. AI ingests and recommends your content as authoritative if you get it right and only if you get it right. Here we go back to the old acronym garbage in, garbage out as a rule of thumb. That means having silos of data or having your website not ready to be crawled appropriately is a problem. It certainly is. And in this new world, don't forget interactive content syndication. This is really important because the reality is that AI bots will pull structured data from white papers, demos, product tables, which means that your content must be machine readable by design, by design, not just human friendly.

15:36This has been a boom to those who own authoritative review sites, the ones that list the pros and cons of everything from coffee makers to graph databases. I mean, everybody loves a good list. And nobody loves it more than an LLM crawler. But that said, let's temper that with the fact that earned media content does rank very nicely. So I think, what have we learned here? If you're a company in enterprise or B2B SaaS in 2026, you've got a problem and you need your category strategy to do four things. Okay, so listen up. One, define the future problem space, not just the current one. Yeah, and build ecosystems that extend conscious of AI assistance and AI-powered humans.

16:23Think about the integrations you need. Think about the developer communities you'll need to bring with you. And remember how AI assistants or AI-powered workforces will help your company develop and use that to consider who you're going to hire and when you're going to hire them. Yeah. Next thing you really, really must do is leverage SEO as a semantic authority strategy. And that is one that influences AI recommendations. Yeah, and just on that one, it's a thought, but maybe the SEO folks and processes used here too far ain't going to take you where you need to go. I think the other thing specifically from a category point of view is convert your category into a platform narrative so that buyers and AI systems, perhaps differently, will reference it when they're looking to solve their problems.

17:17Yeah, and it doesn't just stop there because you're going to need to layer on top of that. You need to design your product and messaging so that AI, whether that's internal search, external search, or AI procurement assistance, of which we're going to see a lot more, learns your category first. And that's the real moat. And just like moats that used to surround castles, one worth constantly maintaining.

17:51Think there are too many lookalike tech companies chasing the same markets? We do. We think it's because so few tech startups are able to build and lead their own categories. What matters most is to be different. Stop following and start building your own unassailable leadership position today. Working with the category design gurus at Categorical brings decades of differentiation expertise to your team. Book a consultation from our website today and we will send you our one pager detailing how to start designing your own highly differentiated category. Okay, so you're on a category design journey.

18:32What could possibly go wrong? We have talked a little bit about this earlier in editions of the Difference Engine and we're now on 67. so we thought it would be a good idea to revisit them because as we all know the technology business is highly dynamic though things change and you may need to focus on new things so here we go here's a list of things you really really need to think about preferably before you even go on the category journey never mind look behind you and wonder what the hell's gone wrong so let's start with the big picture. You know, category design fails most often, not because the idea is wrong, but because the company conditions and occasionally the market, we have to admit, aren't really ready for it.

19:23Yeah, and I think you're right. We've been through some of this before, but two years in, it's so worth reiterating because these issues come up time and time again in practice. Like it or not, category design is an organizational stress test. It exposes weaknesses across corporate culture, employee history, organizational structure, sales incentives and much much more and that in a way is almost a good reason to undertake the exercise but it's definitely a mixture of therapy and surgery. Yeah and talking about therapy I'm just going to jump on the consulting couch here and start off with what we think of is what is usually the first blocker and that is good old culture often quite difficult to define but you know culture when you see it i mean one way to think about it one way to think about your own culture is you know are you deeply product centric as a company do you know do the product developers do the product managers hold sway or are you a real listening company are you problem centric and if you are you've probably got an advantage because if you're product-centric, then your category design is going to struggle because you're not looking at it from the point of view of your customers.

20:43I'm avoiding mentioning Grok again at this point. All right. So yes, category design requires asking, what problem do we uniquely solve? And that could be a problem that hasn't emerged yet as a problem. It could be one that customers have but don't realize. In many tech companies, however, the ones we deal with certainly established ones, the dominant mindset is how do we ship the next feature? Or, and I've just read the Mike Lynch book, how do we hit the next quarter's numbers? And way too often, by the time you've figured that out, the problems moved on, and you're solving perfectly for a limited number of yesterday's customers.

21:20That, to me, is satisfying not designing the future. Certainly. And I also see some other issues in in risk-averse culture, and we have encountered a few of these on our journey. And, you know, the category design demands narrative leadership, conviction, and most of all, patience from the top, and we mean from the top. Got a live instance of that right now. And, of course, these internal politics will kill stone dead. your category design desire and ambition if you don't manage it. If leaders are rewarded for protecting existing products and revenue streams, guess what? They'll resist defining the category, even if they really know and deep in their hearts it's the right thing to do.

22:09We call these people Zeds, and often they hide until the very last minute, sabotaging the process very efficiently. And in a way, you could sort of say that's very obvious, they're going to do that sort of stuff. Because most of us are resistant to change. But that ignores the fact that, as we know, and you said, change is inevitable, especially in tech. Yeah, so the challenge here is to move from product-centric to problem-centric. And I think we think a classic example here is Microsoft in the pre-Satcha Nadella days. Yeah, very learned behavior, right? The culture was, let's call it, Bulmer-esque.

22:47Product-division-centric with Windows in all of its various forms, the all-powerful Office suite that sort of ruled our world for far too long, and critically to this day, the Microsoft server business, still, you know, hanging in, throwing off cash, even as your version. And each one of those seemed to divide and rule and defend its own turf. Yeah, and frankly, all this divide and rule delivered a whole series of Pyrrhic victories that made category creation really difficult. um you know the company defended inside existing categories rather than redefining them which given its market hegemony it should have done and of course they just fought against themselves while google and everybody else were making hate yeah i mean like let's remember um let's remember the mobile version of windows shall we good grief talk about uh not making a category right so very understandable right the office suite microsoft's office suite for instance captured almost every single morsel of value in end-user computing for decades with a few misses.

23:53How did they miss Adobe's rise? But you contrast that with Microsoft's later shift towards cloud productivity and intelligent cloud. They sound like categories to me. This is a major cultural reset driven by Nadella, who's widely liked and widely thought to be very visionary. and it led to the reassertion of Microsoft's leadership that we see today. Yeah, well done, Satya. So the next thing we think can be a major problem when you're trying to establish a new category is your own company history and legacy. And that matters more than most people think. So if you're a company that grew up winning feature awards in an established category, it's difficult to unlearn that behavior and move on to the next commercial innovation cycle.

24:42Yeah, especially with that success. Success is a terrible teacher in many ways, and past success can create category blindness. Teams assume the old category definition is fixed. They may have made a significant purchase, maybe a house as a result of all this. History tells us, and especially in tech, this is anything but a fixed world. Yeah, and it's not just about company history and legacy within your company. legacy customers can also be a constraint here. You know, if the company, your company, and I've heard this in so many conversations, is afraid to confuse or alienate its installed base, it won't fully commit to a new category narrative.

25:30Ah, the innovator's dilemma. Yes, very good. Acquisitions create complexity. Multiple product lineages and the personnel associated with them don't call my baby ugly, my code is gorgeous. That often means multiple conflicting category stories. A company I was very familiar with and lived through that exemplifies this story is Intel. So why is Intel a classic example here, Paul? Well, I think their founder, or let's say most famous leader, it was the author of the amazing book, Only the Paranoid Survive, which is a fabulous read, a recommended read. And what's nice about that book is it has real-life management examples of, let's say, how to be paranoid and how to adapt to market changes.

26:17Yeah, the problem is they sort of didn't. And decades of dominance in the category we think of as CPU performance shaped how Intel saw value and shaped how it thought its customers received value. Unfortunately, the customer base was starting to move. I think that's a bit harsh. I mean, they dominated for a long time. They played around in RAMs before they got to CPUs, etc., etc. Looking back, history has made it harder to lead in new categories. If you look at the companies I think you're referring to, mobile computing firm ARM or NVIDIA, those guys had their own little almost accidental breakthroughs, which in retrospect looked like genius, but were big risk-taking, category-defining moves.

27:01Whether they were deliberate or accidental, Intel's strength in one category narrative slowed adaption to new ones, undoubtedly. And that left the playing field open to new category leaders, whether accidental or deliberate, as the market changed. And also something else I noticed a lot when working with Intel was another issue that can cause problems if you're trying to define a category. And it's geographical and organizational distribution. Definitely an underrated factor because it is where you are. The reality is distributed global teams often struggle to align on a single category narrative.

27:46And that should be pretty obvious. So that's something you've got to get over. Yes, there's always that push and pull between control and autonomy, right? Category design needs to be tightly aligned between the leadership, the product teams, the marketing teams, and the sales teams. time zones and regional autonomy do make that harder. I wonder if that's going away a little bit, but totally take the point. Yeah, we'll see. But it's a facet of human behavior that regional sales teams living in slight isolation may also localize messaging to fit short-term strategies and local conditions. And that, of course, although they're doing it for the right reasons, as they say, usually to meet the quarterly numbers, but it does undermine the long-term category story.

28:32Yeah, it's not a category if it's like regional, right? No. So if HQ is far removed from the core customers or the market signals equally, that's a problem. So the category definition can then feel a little bit to everybody in their little silos, like a little bit abstract or internally focused. You have to fight against that whole attitude. I think SAP is, I guess, my favorite simple illustration of such geographical and organizational challenges. And that has absolutely affected how they've failed yet to dominate a new category. Yeah, definitely. Deep European roots, dominant in Europe, strong regional autonomy, amazing technical chops, and global sales decentralization.

29:21Yeah, I mean, you know, the face of it. Fantastic. But that strategy made it difficult to tell a single unified category story, especially when shifting from what became tactical ERP, which was, again, became table stakes, to a broader strategic narrative like intelligent enterprise. Yeah, I think you're right. And oversimplifying vastly, they say no one got sacked for buying IBM, but nobody in Europe, no CFO in Europe got sat for buying SAP. And you could say Oracle did the same in the US. But these broader categories like intelligent enterprise, brand new categories do require mopping up all of that problem.

30:05Because otherwise you get regional sales teams who revert to the legacy ERP positioning. Well, I've sold it to all the CFOs I know. Let's do some more of that. Indeed. And, you know, follow that story and see how it goes. So let's talk about, I think it's the fourth one we've got now, which is, as you might expect, this has to come up at some point, your actual technology and architectures. Ironically, and we hate to say this because it shouldn't be true, but it is, strong technology can work against category design. Yeah, this is where I think the confusion between marketing and strategy when it comes to category design is missed out.

30:43So we're talking about the product strategy, the technology, and the architecture underneath it all, not just the gloss you put on this. That said, if the platform is overly complex or presented as overly complex and often built through incremental evolution, you know, some tactical acquisitions here, et cetera, et cetera, it may not map cleanly to a simple, compelling category. That's just the way it is. Yeah, right. The best category design thrives on clarity, penetrable architectures, excessive configurability, or unclear value boundaries make positioning harder. And this is the real DNA of tech and product development.