E153: B2B GTM in the AI Era

Joaquin Dominguez: Welcome to another episode of B2B Marketing Futures podcast, the podcast for B2B marketers sponsored by AdSact. we are discussing one of the biggest shifts happening in B2B marketing and go-to-market strategy right now. The impact of AI on how companies create demand, engage buyers and organize their GTM teams. AI is changing everything from audience research and segmentation to content creation, sales enablement, localization, and campaign execution.

But alongside all the excitement, there are also important questions around what should remain human led, how marketing teams should evolve, and whether AI actually changes the fundamentals of great marketing. explore all of this, I'm joined by three great guests with very different perspectives across demand generation, localization, marketing effectiveness, and GTM strategy. But before we jump in, I'd love for each of you to introduce yourselves, your role, and current focus.

⁓ Jessica, would you like to start? Welcome. Jessica Tozer: Yeah, thanks for having me. My name is Jessica Tozer.

I am the head of demand gen at 90. Joaquin Dominguez: Welcome Jessica, thanks for having me. Harry, welcome. Harry Davies: Hi, I'm Harry Davis.

I run marketing strategy and effectiveness at Sage. Joaquin Dominguez: Welcome, Perry. Thank you so much. And Andrew, welcome.

Andrew Thomas: Yes, hi, my name is Andrew Thomas. run marketing for Aclero, ⁓ global content intelligence company. Joaquin Dominguez: Thank you so much for those introductions. So let's start with the big picture.

AI is already becoming embedded into many parts of the GTM function, but the impact seems to vary depending on company size, market, marketing maturity. ⁓ are you seeing AI generally change how B2B companies go to market today? And where do you think the impact is still overstated? Andrew Thomas: I think that hopefully my panelists will agree.

We probably all experienced shift in expectation from the executives management class where they think everything can happen ⁓ faster, cheaper now. And that's a bit of a struggle because you, you the tools are new and there's always a learning curve. I feel like sometimes people don't factor that in, in their estimation of how much effort it's going to take. I do think that once you play around with the tools ⁓ you experiment, you can obviously hone them and figure out what works and what doesn't work.

And you will eventually achieve some more efficiencies ⁓ and probably some greater scale in your output, but it takes a little bit of time. And I think that's true for any new tool that you add to your process. Jessica Tozer: Yeah, I would agree with that. I would probably add that how I've seen it really change our go-to-market strategy is that people ⁓ have done a lot more research before they get to your website.

And so... Andrew Thomas: ⁓ yeah. Jessica Tozer: you really have to have shown up before. So traditional types of marketing, your SEO, your paid ads, may or may not have even seen those because they've already done their research and they've gotten their answers from AI citations ⁓ they already have their short list of vendors.

So if haven't shown up in AI citations, you don't even exist at this point. ⁓ Andrew Thomas: Mm-hmm. You don't exist. Yeah, that's right.

No, I totally agree. In fact, we... But we've also found though that the nice, that's the downside. The upside is though, that when you do show up, those leads are actually way more ready to buy and way more engaged and actually are interested in what it is that you're offering.

which is great. I mean, that's constantly the struggle in the past is you would just getting their attention, with ads and landing pages and all that kind of fun stuff. And nine out of 10, you you can't convey everything that you do in that short period of time. So you spend a lot of time in the process once the lead comes in, figuring out whether or not it's qualified.

The nice thing with the AI leads that we're getting is they're very qualified. They're just hard to get and you have to make sure that you show up. Like you said, if you don't show up, then you don't exist. Harry Davies: Yeah, it's interesting.

I'll tell a little story about a we did. So that kind of part of my job is to try and decouple the relationship between revenue growth and headcount growth. You know, like, it's very straightforward, go to market planning when you're thinking about sales capacity, you're like, right, okay, well, what's the average quota? What's our target?

Divide the target by the average quota, you've got your number of heads. Let's hire those people. Very straightforward. that means every time you scale, ⁓ you have a bigger target, you have a bigger head count, how do we break that?

And so we thought, okay, well, AI is going to be how we break that. And ⁓ started our sales development rep team. And I up to, have... ⁓ a hub based in Newcastle in the UK, ⁓ one in Atlanta in the US, but I went to the Newcastle one where one of our sales development reps had already been kind of building some stuff herself.

really basic prompts to do kind of prospect research and various things like that. I sat with her, we turned them into agents ⁓ just Microsoft Copilot, so nothing particularly impressive. And rolled it out to the rest of her pod, so eight people. Andrew Thomas: Mm-hmm.

Harry Davies: and I sat and listened into calls for the week. And what we saw is that it was probably giving them about a day back per person, per week, reducing the amount of research. And then we also built an agent that enabled them to do the sales qualified opportunity handover as well. And that was really great.

That's saving them a day a week. There was a guy called Craig, one of the SDRs. He was like, this is amazing. I don't I used to do an hour and a half writing up when I got home.

I don't need to do that anymore. Like he was getting his life back. It was like having an impact there as well as improving the productivity. But I was listening into the calls and this one call with the woman that had built these agents, woman called Laura, a fantastic sales development rep, loves being a sales development rep.

She had profiled the customer she was about to talk to. with the agent, had the conversation. It was great. It turned into an opportunity.

That was really great. What landed with me and what I took home and thought about for, I've thought about since then was the conversation at the end of the call. The prospect said to her, how long have you been at Sage? And she said, I've been at Sage for four years.

And he said, ⁓ is that when you changed from the catering manufacturer industry. And she said, I've never worked in catering manufacture. And he said, really? But you know so much about the industry in my company.

And I was like, this is not just about efficiency. This is about effectiveness. This is about having a better experience. Now, probably rest of you have similar relationship where account executives are really expensive.

Andrew Thomas: Ha ha ha. Harry Davies: sales development reps are really cheap. You know, okay, well, if we can equip sales development reps that come across as experienced as account executives, we can chop a whole load of cost out and can give the customer a better experience because their expectation, because they've done a lot of their research, isn't a kind of bank qualification. They want to go deep.

Andrew Thomas: Mm-hmm. Harry Davies: And so if we're able to enable someone on that very first call to be really good, to be really impressive at a lower price point for us, because we're bringing in more junior people, that's a fantastic improvement, both in customer experience and cost for us. And I've been so excited since then, like every little bit of the marketing and sales fun, I think, right, can I get rid of that? Can we, can we improve that?

And I think the expectation to make it more effective is important. You get it more effective and then the efficiency comes. Andrew Thomas: What? Jessica Tozer: Mm-hmm.

Andrew Thomas: was just gonna say, I'm really happy to hear that because we're literally embarking on that right now, trying to come up with AI tools to help our sales reps. And as you say, not ⁓ helping them study prospects that they're calling into or contacting, but also just knowing what we're offering better ⁓ and able to answer. any technical questions that come up. And like you said, mean, arming somebody who is maybe a little bit more entry level, just it's way more effective for the business and speeds up the sales cycle.

So I couldn't agree more. And I hope we get the same results. Harry Davies: It's fabulous. fabulous.

⁓ Jessica Tozer: Yeah, I would agree with you. just makes us a lot more effective because when we are writing content, so ⁓ different than your story, but when we're writing content and we have different personas that maybe someone on my team who's writing the content hasn't worked in that industry, it makes it a lot easier to write content that speaks to that persona. you know, if let's say construction. For an example, I don't know a ton about construction.

Maybe enough to be dangerous, but haven't been in the field. But if I'm writing content for how construction companies use our product, ⁓ AI helps make that messaging resonate better. ⁓ Harry Davies: I'd love to hear your opinion on this. So if you think about marketing and sales, the way they hand over is very much how we've always handed over.

However, the customer expectation has changed. So a customer is like not wanting to talk to a salesperson until they're ready to validate their solution. So then you've got like Jessica Tozer: Mm. Mm-hmm.

Harry Davies: marketing channels becomes super important. We still need to do the kind brand advertising at the start and we still need to demand gen to like kind of drive the leads and opportunities. But we probably also need to kind of take a few stages of the pipeline and deliver those through marketing channels. And I think that's where it becomes really interesting because That's a much more personalized experience.

Like a salesperson, when they're working through someone through a pipeline, they're answering that specific businesses and that specific person's needs. at scale. Andrew Thomas: right. That's expectation now.

Like you said, the expectation is that it's a continued conversation that's happening earlier through the marketing materials, not through having a conversation with a salesperson that they don't want to talk to you. And Yeah, I think you're right. I think you're the nail in head and the only way you can achieve that kind of scale is by using AI. And as you were saying, Jessica, coming up with all the derivative versions ⁓ the content that you create, but I will throw out the one kind of word of caution that I think we're probably all aware of, but you still need to have a unique perspective.

You still need to have a unique idea. And anybody that's relying on AI to do the initial creation or the initial heavy lift is just going to be generating a lot of generic content, generic messaging. So I love using AI for content variation, right? It's perfect for that.

⁓ I don't like relying on it for coming up with an idea, right? Because the way AI works, obviously it's only ever going to come up with any idea that's ever been trained on. Jessica Tozer: Mm-hmm. Andrew Thomas: and you know it's pattern recognition and pattern generation and so by default everything it's going to generate is something that's already existed so you have to you know it can't give you the creativity what it can give you is all of the variation at scale which is extremely helpful.

Harry Davies: have a slightly different perspective on this. I don't think it's totally different, but it's slightly different. I think if train up ⁓ a large language model knowledge base, whether that's kind of Kopathy style ⁓ it's a database, you that up, but with marketing principles as well as custom. Andrew Thomas: Yep.

Harry Davies: stories and content like actually the kind of empirical evidence from academia. then you kind of take a lot of the human bias that sometimes creates less than average marketing. So if I think about like, I'd like to say that we're amazing, but... Not everything that we produce is amazing.

Quite a lot of it's average. some that, we have ideas that we've, as human beings, we've kind of maybe read something on LinkedIn, we've got excited about something, but it doesn't work and it's not right. But we really, really passionate about it. And then we go meetings, we argue about it, all that sort of nonsense that I'm sure we all have.

That doesn't happen. If we outsource it to a machine, I think the importance is training that machine. then we don't need to be super, super creative. We just need to be more than average.

We just need to. ⁓ And like, I'm kind of starting to think I might be able to get better results installing marketing as opposed to hiring marketers. Andrew Thomas: Yeah. Yeah, no, I take your point.

And I, and I agree that, if you, if you constructed a platform where you can, whether you're doing through prompting or fine tuning or ⁓ rag or, know, of the approaches, yes, you can absolutely rely on it more ⁓ and to, do the generation. ⁓ just, where I'm where I'm seeing a conflict is it's kind of, it's an interesting catch 22 because we know that we need to show up in those AI search results in order to get the leads now. And that's, know, AEO, GEO is becoming more important than SEO.

yet the way to show up in that is if you're generating content that the, those broader AIs are crawling on the web and indexing and learning from, and they will wait. Jessica Tozer: you Andrew Thomas: ⁓ results for having a unique perspective, right? If you're just putting out general, general content, general knowledge content, you're not going to show up, right? You're not going to get, you have to have a perspective.

And so I guess to your point, ⁓ I totally agree what you're saying. All I'm suggesting is that what's really important to get the most out of AI is that you do need to make a decision as a, as a brand. What's your unique perspective in your industry, in your niche and kind of stay true to that perspective and own it. And that can come in with, the prompting and ⁓ the that you set up the AI.

So that does continually show up. And think that's the sort of thing that I like to say that when I look at it kind of very simplistically, we use AI for scale. We use humans for judgment and we use platforms for orchestration. And so I think what you're alluding to Harry is you have to set up the platform and make sure that you're, you you, you're not just going to go to raw co-pilot or chat GPT or Claude and ask it to generate something.

You're going to make sure that you have a platform where you've, you've queued it appropriately so that what it does generate is ⁓ Sage or, you know, for a Acclaro or whoever. right? It's not just going to give you ⁓ generic results, but yeah, I agree. Joaquin Dominguez: Yeah, it looks like AI is raising the quality bar on almost everything, right?

Because in the past, only a few companies had access to very good SDRs, right? And you needed to train them ⁓ and you needed to invest a lot in that people. Today, maybe you can train a machine as you were describing, Harry, like ⁓ if you good at training a machine, then Andrew Thomas: Hmm. Joaquin Dominguez: you, you, can, you can get like a junior SDR performing very well.

Right. But, but today apparently it's like what you mentioned, right? And you need to, you need to build a platform where everything is connected and, and, and, and content that you used to create in the past was good enough. What was good enough in the past is not good enough today.

You need to write almost today as an academic. Andrew Thomas: Get a lot more out of it. Jessica Tozer: Mm-hmm. Joaquin Dominguez: paper on whatever you write, when you need to do a very deep research and put that into the content, otherwise you won't show up.

Andrew Thomas: Yeah, I think that the issue is in the past, you could flood the zone, right? You could spend a lot of money and get a lot of content out there. by default, you would show up with SEO. But today, that's not really true with AEO or GEO.

You really have to have something unique that people are going to be asking AI about. And then it's going to add you ⁓ to results. Jessica Tozer: I 100 % agree, you have to have a unique point of view, you need to have some stats that no one else has, you need to have something that AI can't just grab from everywhere else on the internet and create content for you. Joaquin Dominguez: And where do you think AI ⁓ improves outcomes in DTM?

And where do human judgment, relationships, creativity, and trust still matter most? Right? Because if you could ⁓ a machine maybe ⁓ improve something, like as we were discussing before, like a junior SDR, performing like a very senior Yeah, that's something that you definitely can but... Where do human judgment and relationships matter most, do you think?

Jessica Tozer: I think it's that point of view matters the most ⁓ the conversations you have with customers and people use your product and people that love your product that you can those quotes in. Stats on what matters to that no one else has that AI cannot generate. that stuff that humans still have to collect. Andrew Thomas: I would even add, I would add something else that taking a page out of, ⁓ know, I come from the localization translation industry and we've been using AI before people called it AI, you know, for decades.

And in fact, all came out of the translation space, the technology, it all came out of translation. And one thing we learned really early on with machine translation is if going to put AI generated content or AI generated anything, The human element involves trusting your customers and those recipients and ensuring that they have trust in you by clearly telling them that it was AI generated, right? When machine translation first came along and it started to replace human translators for certain types of content, if you didn't actually label it as machine translated, then when customers out, they would sometimes get very upset.

know, obviously if a human does make a pass on through the content before publishing it, then you don't need to label it that way. But if you're going to automatically create something without a human ever looking at it, and this is becoming more important with agents. And so when you have AI agents that are going to go ahead and make actions and maybe produce content and put stuff out there, I think, depending on your industry, there's going to be cases where you're going to want to label it as AI generated.

so that the customer doesn't feel duped in any particular way. And ⁓ if is a problem and because AI does ⁓ can it ⁓ is of making mistakes, but you so are humans. You also a company don't necessarily get tagged with it as much if you've clearly labeled it as AI generated. So It's not exactly directly answering your question, Joaquin, but I just wanted to bring it up because I think that trust is a very human element.

there's nothing worse than broken trust to burn a bridge with a prospect or a customer. And so that's one thing to think about when you're implementing these AI tools. And actually, Harry, it would be interesting to hear from you on your SDR example, when we've been... thinking about implementing it on our end, of the ⁓ practices that we've been told is for the agent, the person when they're on the call is to say, ⁓ I have ⁓ an tool that's assisting me and actually disclose that while they're talking.

⁓ And ⁓ assuming because of that trust issue, ⁓ why were told that that was the best practice. But I don't know if you've had any issues with that at all ⁓ in setup. Harry Davies: I mean, that's not something that we've done. an interesting question, because it's such a...

Andrew Thomas: Okay. Harry Davies: think quite soon the expectation will be that, yeah, everybody has. I personally try, I'm trying to force myself to go, right, okay, any single task that I need to do, how do I use AI to do this? And that's like my own personal learning development.

And I'm still not ready to accept AI slides because I'm very particular about slides. Andrew Thomas: It's default. Yeah, yeah. There is a window.

Yeah. Harry Davies: but it's just painful, I wish it would work better, but it's not quite there yet. I think to the broader question of what's the value of people, think there's definitely a huge number of people that prefer talking to other people and that's ⁓ to stay the case. ⁓ some cases, that's not the case.

I and one of the things that... that we've trialed is starting to use like AI SDRs where we give people the choice coming inbound. can talk to an AI or you can talk to a person. I'm just experimenting with that at the moment.

Because some people would prefer to talk to a machine because they're like, right, okay, I can hang up at any point on this call. ⁓ then there's also, I think, Andrew Thomas: Mm-hmm. Yeah. Harry Davies: I genuinely do think that we're going to get to a point where the and knowledge of marketing and really understanding how marketing and go-to-market and sales works becomes something where you are directing tools as to directing people.

And I think for quite a long time, unless the folks DeepMind, OpenAI and various place crack general intelligence, I think for quite a long time that is going to be quite a human And that's a great thing. And we've seen is really quite interesting that lots and lots of junior people in our organization are really thriving by adopting AI. And then lots of quite senior people are getting really ⁓ excited and using AI. But there's like kind of a middle group.

Andrew Thomas: Yep. Harry Davies: And sometimes I fall into this middle group where it's like you're so used to, you've kind of built your career on really good at the tasks and you kind of want people to go through that for the learning. But think that might disappear. And I think actually what we might find is that a massive hollowing out of middle management as opposed to junior people, which I think people are freaking out about.

I mean, there's all kinds of ethical things. Andrew Thomas: Thank Harry Davies: around this that not really going into. But I do think that kind of that ⁓ of you want to go to market is the thing that humans add. And then it's just question of, right, will the LLMs and the various different tools keep up with that vision so that you can execute.

really, really well? ⁓ Andrew Thomas: Yeah, it's. Harry Davies: Now in some cases you can execute really well now, in other cases there's some gaps. Andrew Thomas: talking about hollowing out of the middle management.

⁓ completely agree with that, but it's interesting because it always makes me think, you know, what happens a decade from now or longer because those junior people then need to become senior people over time. And I don't know about you guys, but at least in my experience, you know, the lessons you learn along the way about thinking about process and scalability and, you know, redundancies and succession planning and all these things that you kind of learn as you move up the ladder.

If we if we lose that from the workforce, which is a very real possibility as an impact of AI, you know, where are those skills going to come from? Because I don't think those kind of skills are going to go away. They're still going to be needed even with all of the technology. this is true for any role, by the way.

I mean, I think ⁓ classic thing is like thinking about developers, software developers and all of the coding that they can now do with AI and like, how do you get from a junior programmer to a senior systems architect? You you do that by making mistakes and learning as you develop and go along the way. And if there is that missing middle, to that's the big question mark. is how we are as humans, how are we going to adapt?

Because I do think, to your point, work is radically changing. yeah, it'll ⁓ I can't where it's gonna go, but it'll be very interesting to see. Joaquin Dominguez: And what skills, mindset or operating models do you think will separate strong DTM organizations from average ones over the next five years? Jessica Tozer: I think creativity, because going along with what Harry and Andrew are saying, you know, if we, if we lose these middle managers, if we, if people are not learning through making mistakes, then is just going to start to be the same.

It's, I think the term is AI is collapsing on itself. So it's just, it's ingesting the same stuff. It's regurgitating the same stuff. ⁓ Yes.

Andrew Thomas: the Ouroboros, yeah. Jessica Tozer: And so, and so. Joaquin Dominguez: and I don't know if you have an answer to this Jessica or any of you guys but how do you make an organization more creative then? Andrew Thomas: classic.

a page out of ⁓ Mad Men or, you know, but you need to, ⁓ you I it's, it's one of those funny opposites, ⁓ you know, you, me, least personally, ⁓ I, I'm creative when I the time and the space to basically be a bit removed from ⁓ the daily grind and the stress of that in some way, right? And then, maybe it's a walk in nature, maybe it's ⁓ doing a activity, you it's just something different than your daily patterns. Kind of it creates the space for creativity. That's not something that a lot of businesses take the time to support anymore.

it would be, it'll be interesting to see, maybe that's something that ⁓ as we need people because they're orchestrating AI agents, like Harry was saying to basically, be, a one person marketing team or, know, running instead of running, one campaign, you're running almost infinite campaigns simultaneously. maybe part of that will be giving those people the time and space. to be creative and giving them some ability to shut the world out for a little while just so that they can come with new ideas.

Jessica, I add to creativity, would add curiosity. I think that's the other main C word because you gotta be curious about the new tools and the new processes. Don't just use but try to dig in and learn how they work because in my experience, Anytime you learn how something actually works under the hood, then you can really start to play with it and get better results out of it. If you, if you dig in and learn how this technology works, then you know how to prompt it more effectively.

You know where it might not be well suited for a particular task, but it might be brilliant at another task. Like we've talked about the SDRs, but the other part that I'm, you you kind of mentioned it Harry about how they could use it for research in the prospects, but the other ⁓ area that I really love using AI for now is competitive intelligence and arming marketing. And, the past, product marketers would always have to generate battle cards. And as soon as you created it, it was out of date.

We don't need to do that anymore. Now we can curate a set of information and give our Salesforce an agent that they can ask any question to. And it's been, it's going to give you the right answer because we have armed that agent with the right information on the back end. That's a perfect use case for AI.

you know, some other ways that people think, ⁓ I could use AI for that, it might not be the best use case. So I think curiosity is what I would add to, to that list. Harry Davies: ⁓ I was thinking curiosity is pretty important to hire for, I think if I ⁓ thinking about how would I develop my skills and knowledge now, would I care about? I wouldn't worry so much about the kind of technical skills.

I can remember kind of having to learn all those things like ad tracking and all that. Andrew Thomas: Mm-hmm. Harry Davies: kind of nonsense stuff that you end up doing as junior marketer. Painful.

I wouldn't worry too much about that. would want to learn ⁓ about methodologies, learn about marketing theory, learn about how go-to-market things like the by design revenue architecture. I'd go deep into the go-to-market. Jessica Tozer: you Harry Davies: theories, into the academic literature.

And then I probably, ⁓ this has come very much from my own experience with AI, one of the first things I started doing, like I used to do a lot of research ⁓ I was junior, going like physically ⁓ customers and talk to them and be where they are. And I doing a discovery project for a bit of work at work. And I thought, okay, I'm going to just record all of these interviews on Teams. And then I pulled all of the, all of the transcripts together, built them into an agent.

And then I was able to have a conversation with all my interviews. I was able to code them up and kind of create the research report in seconds. All of that. I remember doing that with a pair of scissors and chopping different statements up and putting them in piles.

Andrew Thomas: Mm-hmm. Yeah. Harry Davies: Luckily, I've got good genes, I'm a bit older than I look. that kind of, I would probably want people to do some, like have some skills at first hand collecting data from customers.

Because ultimately, if we think about ⁓ drives that creativity, how do you ⁓ that creativity? Well, you need to have the insight. In order to have the insight, you need to be able to collect the information. If you can collect the information, you can generate insight, which then gives you springboards to come up with ideas.

becomes very easy to come up with creative ideas if you've got really great insights. In order to get really great insights, you need to be able to collect that information. If you can do that firsthand and you can talk directly to customers, at scale and then record it all, like you can do 50 hours of interviews and you can turn that into reports and insights really, really quickly now. You can actually probably start embedding some of that creativity into an organization because people are collecting the information, they're exposed to customers, they're exposed to the kind of go-to-market flows and then they will automatically be creative because you have to be.

because you've got new information. Andrew Thomas: Yeah, I think you bring up a great point about data collection as being a new that we really all should be thinking about ⁓ marketers. Joaquin Dominguez: you on something that don't know if you are aware of this, but that's really researched by the theory of innovation and creativity. Actually, that's how advertisers work.

They don't just do ⁓ brainstorming. They do a lot of research before sit into a brainstorming room. ⁓ And going back to your point, Andrew, they... They also value a lot your freedom and going for a walk, taking a shower, driving around.

But before that, you need to do the research and then your brain keeps working and connecting ideas while you're not thinking about the task because you are unblocking your there is a lot of ⁓ theory on this as well. Andrew Thomas: Right. Just breaking. Mm-hmm.

Joaquin Dominguez: We've covered a lot today and demand generation, localization, creativity, the future role of marketers. I think what's particularly interesting is ⁓ while AI clearly changing how teams operate, many of the fundamentals around ⁓ customer understanding, trust, think that's what... actually is becoming even more important. want to say thank you so much for all of you for participating in this podcast.

And yeah, really looking forward to seeing you all again. And for all our listeners, please don't forget to subscribe and give us a like if you like this episode. yeah, till next time. Thank you so much.