Why Only 16% of AI Prompts Cite Your Brand—and How to Change It with Tom Rudnai

Welcome to the B2B Growth Blueprint podcast, where we talk with entrepreneurs, founders, and leaders at B2B services, tech, and SaaS companies to learn the systems and processes they've used to create success for their business. And we'll also learn from marketing and sales experts to provide insights to help you grow your business. I'm your host, Mark Osborne, and the podcast is brought to you by Modern revenue strategies and their 10x ROI B2B growth guarantee. Schedule a free diagnostic call to learn the fastest path to double your B2B revenues in just 90 days. Hello and welcome to another episode of the B2B Growth Blueprint podcast. I'm your host, Mark Osborne, and today I'm very excited to welcome my guest, Tom Rudney. Now, Tom is the founder and CEO of DemandGenius, where he works at the front lines of how AI search is reshaping the B2B buying journey. He's helping brands shape the way ChatGPT, Claude, and Gemini talk about them at every step from prompt to renewal. Before DemandGenius, Tom started his career in sales at Zephyr, which is a dynamic paywall subscription platform, where he watched B2B buyers begin to use AI tools as thought partners, rather than search engines. He left to co-found Demand Genius with James Carter in 2023, building a platform that connects website engagement and AI citations to CRM data so revenue leaders can actually attribute what's driving pipeline. What sets Tom apart is his insistence that AI search isn't an SEO problem, it's a revenue problem. His research found that only 16% of AI prompts in complex B2B buyer journeys produce a brand citation. And his work helps companies move from chasing rankings to engineering the context AI engines use to make recommendations around them. Today, Tom works with CMOs, CROs ready to turn AI search from a tactical SEO add-on into a measurable lever for pipeline, deal velocity, and retention. So we've got some exciting, uh, topics to talk about today. Tom, welcome to the show. I'm really looking forward to it, but Why don't you give our listeners a little bit of background on, you know, sort of what led you to this and what were some of the insights you had along the way?

Yeah, well, first of all, nice to be here. Thank you for having me, Mark, and thank you for a very generous introduction. I guess I'll go back a little bit because I think the way that I've meandered my way towards where I am now is quite relevant for the problem that we actually solve, which is if you go back far enough, I mean, I've held roles in various different tech, AI, SaaS startups. marketing, operations, sales, kind of found my home as an enterprise sales rep at Zephyr where we help big media brands improve the way that they monetize their content. And so you can see how we then kind of meander our way towards now thinking about how B2B brands can use content to kind of sell through AI. And that's kind of the, I guess, the slightly unique perspective. Most people in our space come from an SEO background.

Yeah.

And they're the logical first people who adopted the problem of AEO CEO, you'll hear all sorts of different acronyms. They're all broadly different ways of saying the same thing. But we've done a lot of original research and we think it's actually a very different problem and one that particularly for B2B brands is something that people need to treat as a much bigger disruption. It's not a new channel that's opened up, although it is that it can help you from that perspective. It's actually something that very fundamentally changes the way that your buyers discover your brand, engage with your brand and ultimately, I think, will eventually buy your products, which I think just completely changes the way that we need to approach it. It's not something to be optimized. AI is something that we need to treat as a market participant. It's more like a partner than a distribution channel.

Yeah, so that's really interesting. And in fact, there's a couple things that I think are worth tapping into there. From an entrepreneurial journey perspective, it's interesting, you know, as you looked at what has become a very crowded marketplace, you know, what was the, you know, sort of the guiding, you know, point of view that you had that, you know, sort of guided how you built the solution that you built? And then, you know, really tell us about what that is. How is DemandGenius different from all of the sort of sea of solutions claiming to sort of solve this AEO, GEO thing?

Yeah, well, I'll start with the first one. I think what's— because like any kind of early stage business, we pivoted a little bit along the way. But the one thing that stayed constant from our first moment to now is the belief that content is a very underutilized lever in B2B and actually very underappreciated. So if we zoom out from any AI hype, there is a very long-term trend towards the digitization of engagement with brands, with products, digitization of how we transact. And it's something that I certainly felt going back to my days as an enterprise sales rep, right? It's harder and harder to engage directly with buyers, and there's all sorts of data around that, right? As you've got younger generations moving into buying positions, increasingly they expect to learn and transact digitally. Now, that doesn't mean that sales reps are dying and a thing of the past, but it does mean you need to be a lot more clever and a lot more deliberate with how you build. I normally desperately try to avoid words like omnichannel, but build cross-channel experiences where you allow people to self-learn and self-teach, and then sales is something that is there as a resource for them rather than the kind of traditional 7-stage deal process that everyone knows and still likes to try and push customers through. So that was like our founding belief. And then From that, we got into looking at content attribution and how we can, how can we prove that the way that people engage with content, like people don't stop engaging with content once you've decided they're an MQL, no matter how convenient it'd be for our internal attribution. And so we were looking at what patterns we can pick out in how people do engage post-MQL, and it turns out quite a lot. And then that just inevitably led us as the kind of AI and AEO train grew and that it became very clear that this is gonna be a pretty fundamental disruption in our space, that, that was the kind of lens that we needed to look at all of that through as well. But I think it's given us this very unique perspective and that's very much what's different about us. We are not interested really, there's always an element of that, but in how you can optimize for specific prompts to cite your brand, right? I've seen very little evidence actually that that is something that you can do sustainably and the tactics are very different to SEO. What we are really interested in is how you can impact the way that you are positioned and the way that you're presented in AI at both a brand level and also a category level. How can you use content, use kind of PR and different tactics, different ways that we have to influence the way that an LLM views our brand and our category to really shift perspective in a way that actually like opens up your TAM in new ways. And that's where I think the really big untapped and mostly missed opportunity lies in this. And that's why I think it needs to be on a CRO's desk and not a CEO's desk. It's not a head of SEO problem. That's the channel, right? That's the optimization. Influencing the way that the model views your category is a huge opportunity and a huge risk that I think CROs should be thinking about.

That's so interesting. Well, and as you were talking, there are, you know, a few stats that just jumped to mind. I think this research was from Gartner, but, you know, now most B2B buyers are 70% through their journey before they talk to the first provider of the service that they're thinking about, much less the second or third. And the way to influence that obviously is by creating the content that shapes the way that they think about the category and then think about the different solutions that they could use to address the challenges that they have and then the different providers of that so that you are one of the people that they reach out to. And then the other thing that is so interesting is We have seen for years that the content, you know, the right content at the right time accelerates the sales pipeline. So, as you say, after it's in MQL, giving them the right things to the right people, and oftentimes it's somewhat tailored to different folks on the buying committee, can really break through those, you know, those logjams in the sales pipeline. And really what I'm hearing you say is, You used to be able to rely on a salesperson to feed that content to those decision makers on the buying committee, but we have much less control of that buyer's journey today. And so we need to make sure that the answers that they're searching for, we're there to provide that content to them. Am I capturing that correctly?

Absolutely. It's put a blocker between you and your customer in a lot of cases, right? Because they are first going to turn to AI. And we know a lot of things about the way that people interact with AI. So we know that 89% of B2B buyers, now this is a Forrester stat, turn to AI at some point during a complex buying process, right? 89%. So it's safe to assume virtually it's having a role or it's having some level of influence for virtually every brand. That's gonna differ based on your ICP and things like that. But we also know that they relate to it very different to search. They trust it. People trust AI. It's a, and that gives it a very different role. It's not a discovery channel. It's not something that they go to and they say, What's the best CRM solution for me? And it sends you to Salesforce or it sends you to HubSpot. It's something that they talk to at every stage of their journey from problem awareness through to actually looking for solutions. Now, because they do that, and I can use my own purchasing as a bit of an example here, what I think it's gonna really do over time is compress the time to decision. Because what I don't have to do, I bought a new email platform the other week for a campaign that we wanted to run. And so I became aware of that problem. I did a whole load of research. I was looking at how I can find an email campaign that allows me to set up a different campaign for every single prospect, manage all of those things. So I'm fleshing out my problem, fleshing out my requirements, and only then did I arrive on the fact that Lemlist is the best thing for me. Now that previously was a process that would've taken me months. It would've required me to talk to the sales rep of all of these different brands, go on their websites. Instead, I went through the same cognitive process. But I did it really, really quickly and an hour later I decided, right, Lemlist is the one and I'd bought it. So that's obviously extreme. And where I think there is gonna be a lot of role for sales reps and stuff like that going forward is one of the trickiest parts about enterprise sales is consensus building, right? It's not convincing one person, it's convincing everyone and getting them to kind of coalesce in realizing this matters at the same time. Now that's where I think there is still gonna be a human role. But obviously that's a million times easier if every one of those 13, I think it is now, stakeholders in your standard buying committee, if every one of them is getting the right picture back from AI during that process, it's gonna be an awful lot easier to build consensus. If any, every one of them is getting a different picture, the CFO hears this brand might be fantastic technically, but it's bad value for money. Or the CTO hears this is great value for money, CFO loves it, but the CTO goes in and it's saying, oh, it's got this integration problems, bad support.

Right.

Any one of those is gonna be putting up blockers to your deal. And I think that's something that people need to really switch on to.

Interesting. So there's 2 things I wanna double-click on there. The first is something you said here at the end, which is, you know, potentially the CTO could surface, you know, have different questions into their answer engine that points them in a different direction than what the CFO might say or even the champion, the end user. you know, might be saying about it. How do you think that it'll be then important for companies to keep track of how different stakeholders in that buying committee would be interacting with an answer engine related to their brand specific, and then have content that maybe counteracts or is a counterpoint to what, you know, the answer engines might be delivering? Delivering today?

Yeah, definitely. I think it's something that they need to be understanding how they're presented to all of these different buyers based on the different criteria that they apply. And then the difficult part is then how do we influence that? And we can get onto that maybe as the conversation goes on. I think what's a really useful thing for people to be aware of is that because of the lingering SEO mindset in how we apply this, the most tempting thing to do is, right, we need to know how it talks about us to these different buyers. So we build a list of prompts and we look at those responses. That's actually not very predictive because there's a huge number. And I did the kind of napkin maths. Maybe it's, you sort of take a bit of a step back and think about the way that people prompt, right? So search forces you to consolidate your intent into a very short statement, 3 to 5 words.

Right.

And they've actually, as search has got more complicated or more sophisticated and closer to AI, we've seen keywords lengthen, right? It can handle full questions rather than best CRM. So it forced that, and that meant that they were pretty simple and it meant you could actually optimize for each keyword that you wanted to optimize cuz there's only be so many variations of a question.

That's right.

Write a piece of content specific to that. AI prompts are long, they're verbose. What do people do? They give it a role, they assign criteria, they tell them about who they are, they give them all of their context. It's, I'm a CFO at a 50-person business that really cares about this, and this is our tech stack. So when you actually look at what that— like, that one keyword that was best CRM solution, when you look at all of those different things that multiply, it could be 22,500 different But different prompts. Yeah. So the challenging thing is how do we know how we're being presented across all of those different 22,500 variations?

Yeah.

And the answer is you never can. And if you're trying to optimize for each one of them, you're gonna create an awful lot of content. And we've seen that's a, it's a whole different rabbit hole I could go down into the, the impact that that has. Maybe we could do that later as a bit of a watch out. But so what we've found works really well is just try and get to the bottom, run sentiment analysis on what It comes back to properly understanding these buyers, right? But think about, okay, and most brands have this for your CTO. What are the things they really care about? And figure out, does the model think that we are good at that? Each of these different models, how do we stack up in each of those areas? And then do the same exercise for the CFO. And you can do that mapping and pretty quickly you identify where the AI risks are for you because you realize, okay, CTO, we're very easy to implement, we're very flexible, but our support is bad. And that's the thing that's going to get flagged to a CRO— a CTO, sorry. So you can just kind of traffic light it like that. And then you start to realize where your blockers might be.

That's interesting. And so it's around kind of building more of a conceptual map, and then using those concepts to guide, well, where are our gaps, and then where are our opportunities, rather than the more tactical keyword ranking that people would think of within the SEO space. That's really interesting. The other thing that occurred to me as you were talking— so in decision science, psychology, and analysis of the way people make decisions, there's certain sort of patterns that people have for making decisions. Some people, their sort of natural disposition is, if I see a certain number of examples— and that number of examples can vary from person to person. person within this category, then I'm convinced. That's their convincer channel. Other people are like, you know, if I, I know it when I see it. So, like, they'll make one decision because they hear from the person. They're like, they immediately know. And that's their sort of decision channel.

Yeah.

One of the components of a decision channel is time and sort of temporal. And in fact, Like, I've noticed with, you know, me personally, and this is anecdotal, but it's supported by the research that, you know, for a decision where I'm going to use it every day, so like an email platform, I don't feel comfortable making that decision in less than a week.

Mm-hmm.

And so even the compression of time that a generative answer engine partner can help me do to go through all of the logical things. I then will let it sit on the corner of my desk for a week while I think about it, but I'm not really thinking about it. Like, the decision's been made. Do you think that, um, that, that compression of time, uh, will actually start to change some of the sort of human psychology of the way people make decisions? Or do you think that they'll, you know, be like my example, they'll let the decision sit on the corner of their desk for a week as they, you know, mull it over, but not really doing anything with it?

Yeah, that's a fascinating question that I feel slightly underqualified to answer. Like, I can't give you a data-led answer, but I can say from a few data points, I think yes. I don't think it's going to change the cognitive process that a human goes through to make a decision. I don't think it's going to change the fact that trust is going to be a critical part of that. Right. And that's why I'm not one of these sales reps and I've never wasted trust with a sales rep. That will always be a part of it. You have to trust the brand. But what we know Is that AI helps people accelerate through those cognitive processes. So you've just described part of your process is you need to kind of mull on it, you need to stew on it.

Mm-hmm.

But a part of your process is you need to capture the information you need to learn. So if we break down that process, maybe this part gets compressed loads. And for you, those are slightly longer. I'm a ridiculously impulsive person who will quite happily after an hour be like, cool, that one done, bored. So it's gonna be very different for me and it's probably gonna create more actual differences between people. Because that mulling time isn't built in anymore.

Yeah.

So that will be fascinating to watch. The other thing that we know is AI, it does imbue trust. So we've got some data, like I think it's 5.7 times better leads that, leads that come to a website from AI convert 5.7 times better. I'm pretty sure I'm right on that stat, but it's around there than at least it just came from search. Why? Because it didn't. the behavior of how people talk to AI, it came with a recommendation. It's a lot closer to like a channel partner bringing that customer to you because they've come, they've brought them to you and said, this is actually the people you should speak to, they're right to you. So it's kind of your ball to fumble at that point.

Well, and it makes sense given what you were saying before because typing in, you know, best CRM for you know, a B2B SaaS company, whatever you get back from the, you know, the search engine, it's not that robust versus I made a 10-page prompt to tell it all about who I am and what my problems are. And then I went back and forth with it for 30 minutes on, well, what about this? What about that? And then it said, well, you should go talk to these folks. You can understand why that's a 5x, you know, sort of more powerful recommendation that's coming from that. And so, you gave us a framework of how to think about it, which is sort of conceptually, what are the, you know, what do you know are going to be the questions that they're going to ask? And then look at how do you show up around those questions and then create the content to either counterpoint where you have weaknesses or to, you know, buffet your success. But obviously, just creating that content You know, if it sits in a drawer, no one's ever going to see it. But how do you— you know, maybe you can give us some kind of tactical advice. So for the company out there that, you know, right now is like, boy, I want to show up when people search for these kinds of things, and I've done that mapping, I know conceptually where I need to— the content I need to create, I've created the content. How— what are some ways for people to think about how they should actually be using that content to impact those answer engines today, knowing this is a fast-moving spot and there's no silver bullet, or at least if there was one today, it probably won't work tomorrow. But what are some sort of macro trends that you see there and how should people be thinking about it?

Yeah, yes, you're right to say there's no silver bullet. That won't stop you going on LinkedIn and finding people that say there is, but don't believe them. I'm gonna give you one high-level answer and one tactical answer that's a bit easier to implement. So high level, what's useful to understand the way that I think of it, you have 3 levers with which you can influence AI. Think of those as 3 points on the triangle. So we have positioning, who you say you are. You have your content, how you consistently reinforce that, communicate that, kind of flesh that out a little bit and explain why you are that and things like that. And then reputation, which is what other people say about you. Reviews, things people are talking about on social, Reddit, all of that kind of stuff. Kind of stuff. If you think about those as 3 points on the triangle, your goal is to get those 3 things as close together as humanly possible, because what you do then is you create this really, really clear picture of who you are, who you're for, when and how you win. And that's what the AI wants in order to be able to give a confident recommendation. And these are confident machines, right? If you don't give them a confident answer, that's where they go looking to fill gaps or they just make it up. So you're kind of minimizing the space. I have a unproven hypothesis that that's why in the early days of AEO everyone was doing loads of comparison content. People might have noticed that every brand is pumping out top 5 alternatives to this competitor. And my hypothesis is that worked really well because there was this huge gap between those 2 points. Because how do most brands evolve? You decide, okay, this is who we are. You produce content, you get reviews, and then you reposition a bit a year later. And you do the same thing and then you reposition and you do the same thing. And over time, this big gap opens up, which is what we call content debt, which is basically the difference between who you are now and what some of those pieces of content and some of those reputational touchpoints say you are. And that's a huge gap that is just creating uncertainty and clarity. So I'm winding around to a point here because what that changes is really fundamental about how you do this because it's not more, more, more. It's not keep churning out content. And this is where I think some of the advice that's out there is so detrimental. Because taking an SEO mindset encourages you to try and create pieces of content for all of those 22,500 different keywords or different—

That's right.

That's right. What you actually want to do is cultivate your overall digital footprint. Onsite is obviously the most controllable, so generally start there. It's like get your, you know, make your bed before you try and fix the world. And then offsite is harder to control, but there are levers, right? And that's where PR comes into it and social strategies and activating customers and all of those kinds of things. But so that's the biggest thing. Cultivate your overall digital footprint. Don't try and kind of play whack-a-mole churning out content. And we've seen really clear evidence recently. Lily Rae published a great study which showed the negative effects of that. And every brand that does that, you see a traffic spike from the content and then it falls and often falls to a much worse place. So be really discerning if someone's giving you that advice, 'cause it's now been proven to be really bad advice after the kind of initial gold rush.

That's super important and interesting. So I want to say that back just to, you know, sort of—

Yeah.

Make sure I get it because I don't always have access to such an insightful person. So essentially, creating content that is sort of diverse in its point of view creates confusion. for the answer engine and less certainty of who you're for and why you're for them. And so then lowers your probability of being recommended for folks. And so creating more content that isn't structured around a point of view can actually be detrimental for you. Is that— am I hearing that correctly?

Yeah, exactly right. Your goal is to create an overall footprint that is clear, consistent, Quality current. Every piece of content that you produce, you incur a little bit of debt because maintaining that across everything becomes that little bit harder. So that doesn't mean don't produce new content by any means; just means don't turn out crap because that's what makes things really hard.

Right? Well, and that also then sort of suggests that having that clear positioning, whether it's you know sort of your category POV or your positioning strategy, however you want to sort of think about that, but that real clear. Clear guide of this is who we're for, this is why we're different, uh, than everyone else, and this is the evidence that makes it— that's true. Everything should, should, uh, sort of connect to that and circle back to that. Even from the point of having customer reviews that are generic, uh, in nature could be really detrimental to you because it's not reinforcing that same point. Is that right?

Exactly right. And don't— I don't wanna give the impression like one customer review that says something generic, it's not gonna sink it. Like we don't have to be that. It's a— but in the whole, yes. And I'm actually really glad you raised that 'cause that's the kind of tactical thing. Whenever we start working with a brand or with anyone, I think they always expect us to give them a load of content that they need to create or load of content gaps, all these optimization hacks. You need to change this on your website and then everything will be great.

Right.

And I'm there, but my first thing is always like, no, go back to those fundamentals. And this is what, if you're listening to this as I know your audience, right? It's a lot of revenue leaders, a lot of founders, a lot of CEOs. This is probably the biggest thing that you can do for your marketing is get everything out of your brain and down ideally into like markdown format that is extremely machine readable.

Right.

Your market perspective. Who are you? Why should anyone care? When I'm feeling particularly cheeky, People always come to me and ask like, how do we get cited by ChatGPT? And the easy way to push back is why should they cite you? And very few brands, you'd be surprised how few brands can answer that question, right? What is your unique point of view? And then how do you back that up with evidence? We just did a big study. We analyzed using our AI 90,000 pieces of content across the HR tech space. Because it's just a really interesting exercise and you can see how everything's evolved. And we use AI agents to score it across like 30 different attributes. The best chart that was in that was one which shows optimization and differentiation across all of that content. And what you see is optimization really strong because that's what search required. Differentiation really, really weak. 5.7% scored high for information gain. Which is, does this content actually say something that people don't know without this content? Which if you're an AI, that's what you need, right? That's what makes something citable. And 4.7% scored strong for narrative ownership. So there was a clear perspective that came through it. And what's fascinating is you see in this study an absolute explosion in 2025 of content volume. Like it's ridiculous. And you see coinciding with that differentiation falling to virtually zero. So you've got all of these brands who every one of them could be producing the same content and, and, and it's indistinguishable because they're just getting ChatGPT to spit it out. Um, well, there is no reason for a human to come to your website if that's the case. And there is absolutely no reason for an AI to cite you or bring you into an answer cuz they have no desire to do that. That's what's different to search, right? Search wants to send people your direction. Um, so, wow. Very long-winded way, but build information gain. And if you're a CEO, give that gift to your marketing team of a clear perspective and story. Wow.

Well, that is reinforcing something that we've said for a long time, which is that POV, that perspective is the touchstone that everything hinges on. And it sounds like that's even true for this next wave. So you're such an expert in some really interesting areas, and we could talk all day, but I got to be mindful of your time. But before I let you go, give us the pitch. Who exactly do you work with at Demand Genius? Like, you know, who do you work with and how do you help them?

Yeah, so we hope we work with exclusively B2B businesses. And generally speaking, the messier the sales process, the more complex the buying group, all of that kind of thing we love, just because it brings everything that we've talked about here more to the fore. There's more complexity. We help these brands navigate the transition from a search-led world to an AI-led world and everything that that So that's different for every brand. There's no magic bullet, there's no universal strategy or playbook to this, which is something we've actually had to figure out. We now have a much bigger strategy component to what we do as well as just the technology. But basically we help you understand what AI thinks of your brand and how it presents you to every one of your stakeholders in your buying committee, map that in different segments, how you can influence that and what perspective your content, what within the stuff that's really controllable by you can you tweak and improve and actually build a roadmap around things like information gain and narrative ownership on your side. site to own that? And then the last part is, is it actually having an impact on your bottom line, which is often the challenging part. And I think really important for brands who don't wanna chase hype, but wanna just understand, okay, there's this massive shift that's happening at different paces in different industries and understand like, do we have a plan for that? And how do we know when the time is to actually make something, do it, make a more drastic bet?

That sounds amazing. For folks that are listening that are like, we do have this super messy journey and we do want some of this direction, what's the best way to engage? How do they get in touch with you?

Go to our website, duan-genius.com, or follow me on LinkedIn. Get in touch with me there. I'm always happy to help. We do kind of free audits and we try to have a really consultative sales process as well where we don't just kind of pitch you, but we give you insights and help you for free and then see where it goes. So we're always happy to give you— give away a bit of the goods for free and help people understand whether this is something to be thinking about.

Fantastic. Well, Tom, I really enjoyed it. I learned a ton. Thank you so much for joining us today.

Thank you, Mark. It's been really nice to have this conversation. All right.

Talk to you soon. Thank you for joining this episode of the B2B Growth Blueprint podcast. We hope you've gained insights to grow your business. If you enjoyed this discussion, please rate, comment, like on your favorite platform and share Share to others who will find it useful. Do you know a CEO, founder, or leader at an early-stage B2B services, tech, or SaaS company who would like to be a guest and share their story? Send us a note. Let's schedule a conversation.