B2B CMOs Deploying GenAI to Win

 

Drew: Hello, Renegade Marketers! If this is your first time listening, welcome and if you're a regular listener, welcome back. Before I present today's episode, I am beyond thrilled to announce that our second in-person CMO Super Huddle is happening November 6 and 7th, 2025. In Palo Alto last year, we brought together 101 marketing leaders for a day of sharing, caring and daring each other to greatness, and we're doing it again! Same venue, same energy, same ambition, to challenge convention with an added half-day strategy lab, exclusively for marketing leaders. We're also excited to have TrustRadius and Boomerang as founding sponsors for this event. Early Bird tickets are now available at cmohuddles.com. You can even see a video there of what we did last year. Grab yours before they're gone. I promise you we will sell out and it's going to be flocking awesomer!

You're about to listen to a recording from CMO Huddles Studio, our live show featuring the flocking awesome B2B marketing leaders of CMO Huddles. In this episode, Karen Feldman, Jeff Morgan and Adriana Gil Miner share how they're integrating generative AI into their marketing strategies, from content creation to personalization, they break down real use cases, hard-earned lessons and the guardrails that keep AI efforts on course. If you like what you hear, please subscribe to the podcast and leave a review. You'll be supporting our quest to be the number 1 B2B marketing podcast. All right, let's dive in.

Narrator: Welcome to Renegade Marketers Unite, possibly the best weekly podcast for CMOs and everyone else looking for innovative ways to transform their brand, drive demand, and just plain cut through. Proving that B2B does not mean boring to business. Here's your host and Chief Marketing Renegade, Drew Neisser.

Drew: Welcome to CMO Huddles Studio, the live streaming show dedicated to inspiring B2B greatness. I'm your host, Drew Neisser live from my home studio in New York City. One of the advantages of being a marketing veteran, yep, the gray hair is that I've seen a number of hype cycles. Some of them I got right, like the internet and social media. Those were big changes. One of them, I got wrong, remarkably wrong. I wrote a blog post when the first iPhone came out and said, and I just missed the fact that it was more than a communication tool. I just was annoyed I couldn't type anymore, and I missed my Blackberry keys. I got that one wrong, but a few I correctly predicted would fizzle. Now, each time skeptics would say it's just a tool, I'm going to wait and see. If you're a marketer and on the sidelines with generative AI, my recommendation is to consider a new profession. Generative AI will change every aspect of our business, if it hasn't already. If you need more evidence, stick around as I speak with three marketing leaders who are ready to share how Gen AI is rocking their worlds. So with that, let's bring on Karen Feldman CMO of IBM Consulting, who is joining the show for the first time. Yay. Hello, Karen and welcome. How are you and where are you?

Karen: Hi Drew. Nice to see you. And thank you for having me. I am in New York City today.

Drew: Awesome. All right. Well, we have the center of gravity here in New York City, so you got to be looking at all sorts of different ways and how you're leveraging generative AI. But I'm imagining that this is a big new practice area for IBM Consulting.

Karen: It is and for IBM across the board, generative AI is something that, you know, we had to focus on, not just in terms of how we leveraged ourselves for how we market, but, you know, how do we position our capabilities, our products, our offerings, to the market, which we began in earnest about, you know, 20 months or so ago. But of course, AI is not new, right for IBM, many of you probably saw our famous Jeopardy approach where we introduced Watson, but we have matured those capabilities quite a bit, and we have been actively looking at, how do we leverage our own technology and partner technology ourselves within our marketing organization, across other enterprise functions? Because we do fundamentally believe that you know, generative AI, is that tipping point moment that is going to transform the way we all work and the effectiveness of what we can deliver, and the types of experiences we bring to our customers.

Drew: Yeah, there's so much to unpack there. And I just, I'm so I remember the Watson moment, going back to the chess game, you know. I remember the Watson moment, certainly with Jeopardy. And you know, that was when, you know, we were just talking about AI, but it was this thing that maybe companies could do, and it was like way off here, and then suddenly Gen AI has put it at your fingertips, and it changed the game for IBM. This had been a really tough moment, because you got to pivot to another sort of use case of something that you've been thinking about for 30 years.

Karen: Exactly, exactly for us, our experimentation and our marketing organization kind of coincided with when we were doing our first big introduction to the market of our generative AI capabilities. And so this was 2023, April of 2023 it was the Masters, you know. And IBM has had long standing relationships with the Masters over 30 years now. US Open which is going on right now, Wimbledon, but we transform the digital fan experience, right? And we make, you know the Masters more accessible by bringing viewing and new forms of insights and new ways to personalize your experience in terms of how you watch and view your favorite golfers to market. We've been doing that, and we, you know, pride ourselves on innovating each and every year. But in 2023 we were actually launching our first generative AI functionality within the Masters app. And it was spoken voice commentary, so that, you know, you could follow your favorite golfer, but also actually hear and get the commentary you would hear right on regular television about what was actually going on. So it was, you know, kind of this awesome moment to reintroduce, you know, what IBM can do, and our capabilities and how we had modernized those, and we took a step back and said, you know, we're running this major integrated marketing campaign right to support this, ranging from, you know, television advertising to media and social and web and activations. You know, could we leverage generative AI ourselves as our first pilot, you know, to see what types of impacts we could have. And so we had just actually left Adobe Summit. We saw the beta of Adobe Firefly. We got really excited. And, you know, basically asked to be early beta users, and we used it as our first experiment. And our hypothesis had always been, you know, it's going to dramatically change content creation, right, for all of us, but it's also a lot of, you know, the derivative content creation. It's going to just drive efficiencies in terms of how we bring that to market. And so we ran an experiment with, you know, a big sort of anchor of what that campaign was all about. It was, was really designed to be, what if you could remaster the Masters, right? And it was a provocation for anybody who was watching that to say, Huh, if we could do this for the Masters, you know, how might that apply to my business? And so we took this idea of a question mark, and we produced, you know, 1000 plus images through Adobe Firefly, and we ran them in social and media on our websites, and we found we were able to produce 10x the content in a fifth of the time. But what was really interesting is the social post actually performed 26 times better than our IBM benchmark. And so we took a step back and said, All right, you know this, this is real. You know, the teams had fun using it. It is going to deliver on that promise that we've all always had around personalization at scale, right by driving both productivity and personalization, but it performs better because we were able to take the time right to focus on what was actually going to resonate in the market. So we've been, you know, scaling from there, ever since then. But you know that, I would just encourage anyone who hasn't done that yet, just get in the market, get started. It's phenomenal what it can deliver for you.

Drew: So Wow. Okay, there's a lot to unpack there. The first thing I just wanted to put a punctuation point on, which is because I've sort I've actually IBM was a client four times of Renegade over the 30 years. I'm very familiar with IBM. One of the things they're really good at is, when you sponsor something, don't just put your name on the thing, integrate into it and become a part of it. So you've always been able to showcase technology, whether it's at the US Open or the Masters. So now this was a natural you, same strategy, new tactics, right? And I think that's really sort of important in the scheme of things. When I hear 10 times the content at 1/5 the speed, I'm sort of going, Whoa, that's that's a lot of content, but the fact that the social posts performed 26 times better, I sort of my first thought was, wait a second, is that just because you had generative AI, hashtag, Gen AI, what? What could possibly create that kind of you know, that's a big increase. What? What do you what was your assessment of? Why did that do so well.

Karen: I think there's two reasons. One, I mean, you, the teams weren't spending as much time on the creation of the content, right? They were able to do that quickly. So they were able to spend more time on the creativity. Yeah. Right? What do I want that post to say? What do I want that message to be? So that's number one, the images were also beautiful, because they were able to iterate very quickly and create a lot of different images. And then two, we could personalize, right? And so that, that, to me, is the real benefit. And what we at IBM have never really been able to get to successfully, because we're producing to your pointers, so much stuff, right? You have so many products in your portfolio, so many campaigns you're running. You need, you know, so many pieces of assets for any given campaign. We're global, right? So you have to make sure you have that content globally. So the personalization has always been that last step, right? How do you tailor the message and the images to the buying personas we're trying to reach. And in the B2B world, that is only getting more and more complex, as we all know. There's, you know, so many decision makers who are part of that process, but I can't talk to the CIO right about what we delivered for the Masters the same way I would talk to a CMO, right? Because they care about different things, as we all know, and that's why it's so exciting that you're able to produce so much more content in a fraction of the time, because we finally can afford, right from both the people and a budget perspective, to personalize.

Drew: Okay, all right, now we've got to bring in Adriana Gil Miner, Chief Marketing and Strategy Officer at Iterable, an industry expert who has graced our stage before to delve into the topic of global marketing strategies and AI integration. Hello, Adri, how are you and where are you this fine day?

Adri: I am doing fantastic. I am in sunny Seattle, and I got to tell you Drew it's a very special week for me, because, actually, I, as a former New Yorker, I'm going to be going to New York next week to connect there with Iterable community, and I am celebrating 20 years of marriage. It's a special week for me. Thank you for having me today.

Drew: Awesome, amazing. That's well, that's great. And what a fun city to be able to do that in. And by the way, anytime a beautiful, sunny day in Seattle is like that, you know, just that notion alone will stick with you, because you do get about 300 days of rain, just tiny little bits.

Adri: We just wait for that. It makes all the more glorious that one sunny.

Drew: Okay, so I'm curious. You heard a lot what Karen was doing, and I'm wondering, you know, compare and contrast relative to what you're doing at Iterable.

Adri: Yeah, it's, it was really interesting, of course, like, just so great to hear Karen, well, you know, I think one thing I was reflecting is we, I have the advantage of having sort of, like, maybe, like, a two sided view on Gen AI in particular, one is the things that we're doing, you know, directly as a team here at Iterable, you know, as marketing team, but also as a martech provider, I get to work with lots of different brands and see what they're doing as well. So I kind of like see, kind of like both sides. So I'll say that, you know, similar to Karen, I think like creating the environment in sort of like a specific application, really gets to see those results. So here at Iterable, one example that we had, we created a test last year called H fat, which is high volume ad testing. And basically we did a competition in the marketing team using ChatGPT to create hundreds and hundreds of iterations of ads. And instead of like, the typical thing that you do is, you know, you kind of, like, pick the best we actually ran all of those ads in market and then work through rapidly iterating and optimizing those. And we had one of our marketers who wasn't, by the way, in digital, actually win the prize. Because we had a prize of, like, the best performing ad. The interesting thing was that it got everybody, like a real life application of Chat GPT and kind of really encouraging that, sort of like Gen AI adoption. And it produced, I believe we doubled the amount of demo requests that quarter against our normal average because of running this test that we did.

Drew: Okay. So two, that was my first question was going to be, how, when we say the winner, the goal here was to drive demo requests so there was a specific call to action. And then, you know, stepping back for a second, how did you make sure that in 100 different iterations that they were on brand?

Adri: Well, it started really with you have to have the entire team really understanding your brand. And I actually believe that one of the big things that Gen AI is doing to all of us is really pushing us up to be everybody needs to be a brand marketer. I've spent now 25 years in marketing, and a lot of my work over my journey has been more operational, you know, going from like, Oh, I gotta upload the list to do this, create the campaign, the assets, you know. So when that really gets reduced to a minimum, that really allows us to start to think about much more of like, how do I bring to life the brand? And so you have to invest, I believe, on things like making sure that every single employee, every single person on your team considers themselves a brand marketer, knows the positioning, know the voice, the personality, the tone, your audience, your customer, your target, and knows deeply the strategy. And we move away from the idea of having like, oh, marketers are operators.

Drew: So we've got all these versions. We've tested them all. So then you know what? Any since then? How is this impacting? Because you mentioned two levels one, and you're using it, and I'm imagining, I feel like we talked about this. You've baked generative AI into your product, right?

Adri: Yes, and that is one of the most exciting things, we actually just launched a new feature called Journey Assist, which essentially allows you to just prompt a prompt the system like you would do in Chat GPT, just describe what you want in terms of creating a customer journey, right? So imagine you want to create a journey for new customers that you have and you want to onboard them, or you want to create a journey and automate, like how you're going to engage, the engagement points and the messaging points around abandoning cart, which is very common in, you know, sort of e-commerce, or so, any of that things which generally takes a ton of knowledge, a ton of, you know, again, operational effort. Now you can just do it by prompting things. And you have, we have customers, you know, like Cinemark or Next Door, who have been using this already. And what you think about like, that's, to me, you know, sort of the next step with Gen AI, a lot, you know, a lot of what many marketers have been doing is just that sort of like first step of creating content. But if you think about Gen AI, as you know, for example, this one the ability to generate a journey, the ability to generate those like strategic applications of personalization, of how you're going to engage, what are those moments in marketing that you really need to do across a customer journey, Journey Assist at Iterable, you know, the feature of Iterable really provides that it is a really concrete way to take the next leap for us marketers, in terms of that idea of like, Hey, we're going to elevate the role, because you have the machines helping you augment with all the operational and even at the strategic level. Because when you say, hey, I want to automate, you know, I want to engage this customer when they come into my platform, for example, when they sign up for us. Like many B2B companies have a freemium model, right? So let's say somebody signs up for your trial you want to automate. Like, when, how do you onboard them? And then how do you welcome them, and then how do you upgrade them to your paid subscription? That kind of moment, that kind of like strategy you may have a way, but Gen AI can help you actually suggest steps like, hey, you know what? Maybe after two weeks of usage, that's the time where you need to offer your upgrade, not later, not earlier, but two weeks is like the right the right time given all the data that you have?

Drew: Interesting. Yeah. I mean, this used to be in the world of machine learning, they would describe this as next best action, but this is sort of on steroids, if you will, in a bad metaphor, but it also includes the language that can go with it.

Adri: The conversational and I think that that's the that, yeah, the conversational aspect, and also the sort of, like human understanding, I think about what you have behind these, like LLMs, that they're really those large language models. It's not just bringing that basic prompting between the human and the machine, but it's really that two way conversation and that ongoing training that makes the generation of content or strategies or automation really, really rich. It's really a new level that we're all still trying to discover and learn how the power of it.

Drew: So far we've covered in the show, we've covered the ability to just iterate like crazy, which speeds up the process and enables folks to think about, oh, I could try it this way, or I could try it this way and do variations. And that was a common theme between you and Karen. Then we've also talked about how to bake it in. I feel like a big area for generative AI that doesn't get out enough conversation right now is in the area of sales enablement and tool development. And I'm just, you know, in that, and I'm wondering if you have made any progress in that area, or thinking about, is that on a radar, or I need to do a show on that?

Adri: I think you should do a show on that. Because, especially for us, B2B, you know, I often say, like we think about brand and moments and all this stuff. But when you have a sales led motion, like we do most of your interactions with prospect, and then that early, you know, like customer engagement is actually going to be led through a deck, you know, through a PowerPoint, and the channel, you know, the channel that you're relying on is, you know, whether it's an AE or a CSM, you know, like it is a human being that is having an interaction, especially if you're an enterprise in up market, like Iterable is. So yes, the answer is yes, we are, you know, I think particularly when it comes to enablement, there's a ton of opportunity, there's a ton of things that we are, I would say, starting with, like, Product Marketing. And our team is definitely, like, using it more for summarizing, strategizing. How do you get, like, you know, a lot of information in a very, like, short period of time, but we're in baby steps. I think, I think it's a great topic Drew to really drive into. How is this connecting, and how, how is this helping? I actually, one example that I do have it is not directly sales enablement, but one of our customers who we also use is Grammarly. I mean, a huge plug for them. Like, honestly, like, if you use Grammarly, the way that we use it is that we've trained the system with all of our brand, our tone, and even for our product launches like, it's so great, right? You can like, inform your your instance with that. And so then when you have like, your SDRs using, you know, to create their outreach sequences, or or sort of those emails, Grammarly itself is like a little tutor that you have in there to help your sales people create those communications. So I would like, really, really, you know, that's, that's one plug that I have that I had forgotten.

Drew: Well, it's so funny because I have, I still am, on the free version of Grammarly, and that is the most annoying tool in the history. So hard, it's constantly blocking things as a free thing. So thank you for they needed some positive press, because if you're not a paid user, you suffer.

Adri: Well, I mean, that's the thing Drew.

Drew: I know it's like, yeah, you still got it installed so in but guess what? We're gonna make you pay one way or another. All right, let's welcome Jeff Morgan, who's been waiting patiently. Jeff is the Head of Marketing at Elements. He's previously been on the show to talk about top of the funnel challenges. Hello, Jeff. Welcome back.

Jeff: Hey Drew. Great to see you again.

Drew: So how are you and where are you this wonderful day?

Jeff: I am doing excellent today. I am in Salt Lake City, Utah, and one of the reasons why I'm excellent is because we're headed into fall here. We're getting over the heat of the summer, and fall in Utah cannot be beat.

Drew: There you go. All right. Well, we've spanned the country today, from New York to Midwest to the West Coast. Cool. All right. You heard Karen, you heard Adri, what you know? Where is your story? Fit in between those? Those two?

Jeff: Well, it was really fascinating to hear from them, especially because they both are representing brands that are building AI products from marketers, and that is really inspiring to me to hear about how they're doing that, and how they're taking these foundational models and using them to build new, innovative things for our industry. So I'm really excited to have heard about that and learned about what they're up to over there. As far as what I've been doing, like, really what you said at the beginning about how there's these big trends over time, like the internet and the iPhone. And sometimes we can predict what's going to be huge and change our world, and sometimes we can't. I started out my marketing career in 1999 when the internet was just becoming a thing, you know, and the dot-com bubble was right in front of us, and everything felt a lot like it does today with AI. And I started in marketing and web development at that time because it was so exciting to me then, and I've really been, I think, somewhat of an early adopter of these generative AI technologies in my marketing career now, because of the same kind of excitement I have around that. The primary challenge that generative AI has helped me with in our marketing team and our executive team and our business in general, has been doing more with less. I know that's a big theme probably out there for a lot of marketers these days, budgets are getting cut. Our teams are getting cut. The amount of time we have is at a premium, and these technologies came around right at the right time for me personally, where our budgets and team were getting cut, and suddenly I could produce the same amount of work that I was with like a five or six person marketing team or a venture funded startup. So it's a small team down to like one and a half, two people over the last year, and we're doing the same or more volume, not today, than we were before. And I can attribute most of that efficiency gain to the AI tools that I've been using.

Drew: So let me add. So let's, let's go through some of them, maybe more specifically. I mean, it always it makes me sad on one side to think that there were four people and now they're one and a half because the machines are helping them work so much faster. I guess the first question would say is, yeah, but is it better?

Jeff: I personally think that our quality hasn't gone down. The metrics would say that our engagement in content is similar. Our podcast audiences are still growing, and our demand generation has continued to be consistently the bright spot in an otherwise difficult time in our industry and in our business. Little ways to measure success, I would say that the quality has been good, or at least the same as it was before.

Drew: So all right, that's reassuring, and I appreciate that we've got and I'm curious, because we've talked a lot about this on the show and in Huddles about, you know, just the content generation, and we covered that with Karen and Adri. So how else are you using it or progressing it? So we had some more ways of thinking about it, in particular, like with the podcast, which because I know from other conversations that you've had great success with your show. What tools are you using, and how is that improving the podcast specifically?

Jeff: Yeah, well, with the podcast, we've used it to brainstorm topic ideas, to outline those topic ideas, and I would say that one of the tools that we've used to make that especially effective is this idea of the retrieval, augmented generation or RAG. The easy to use RAG tools are like Claude projects or custom GPTs. What we've been able to do is upload our messaging framework, upload blog posts, upload all of our thought leadership content into those models, and then we say, "Okay, what's a podcast topic that we could do that would achieve this goal or that would help someone in our target market with this problem that they're facing?" And because we've loaded it up with all of our custom thought leadership content and our messaging framework, it actually spits out very unique and interesting ideas that previously we would have had to, like, sit around and brainstorm for hours, and now I can do it in minutes.

Drew: Yeah, and it just has a better memory than any, any of us. And I so I want to, I want to check off a few boxes on this. So, as you mentioned, RAG as funny I was yesterday was listening to Hard Fork, which is a great podcast on technology. And I think it was Kevin Roose who was talking about RAGs and explained that to folks. The key thing here is this is not you creating content using just what we could call because generative AI content, to me, is basically average, average intelligence when you do it that way, but if you're taking it from your own GPT, or, you know, custom project on Claude. It's your content, your thinking. And now it says, "All right, find the holes." And so the framework that I want folks to really think about this is you have a new sparring partner. It's called ChatGPT. That's it. This is something that can challenge your thinking, assuming you give it your thinking. Yeah, then it can start to challenge it. So you've been able to use it to identify holes and topics that you think that you're fairly confident will resonate. Of course, you've now had enough time to produce those podcasts. Did it work out?

Jeff: Yeah, for sure, I think that we've had a pretty interesting set of content over the last year that has been a little bit more diverse than it might have been otherwise, because of the tools that we're using to generate the ideas. The other way that we've really scaled our ability to produce podcasts is by using a tool called Descript, right? We will record an episode. And it used to, we used to outsource that to a contractor, and they would charge us somewhere between five and 10 hours worth of work to produce the end result. And now we've let that contractor go, and we just do it internally. It takes about 30 to 60 minutes to edit an episode because the tools that are available to do video and audio editing now are so much easier. You don't have to be an expert to use them, and it will do a lot of the work for you. So that's another way that we've scaled our content creation without just like it's not just writing, writing as the tool, but it also helps you to do other mediums.

Drew: So I'm a couple of things. I'm pulling from other conversations that we've had together. But first of all, and again, it's similar to the what I was saying to Karen. You have a very successful podcast because you have a hyper focused target audience to it. So now it's not like you're still looking for, it's still a hyper focused show. So the overall strategy hasn't really changed, but you're finding topics that might appeal to the target. That's part one, part two, Descript or another tool like it, we use it for. So this show will become an episode of Renegade Marketers Unite. When that happens, we'll upload it to Descript. We will use it to pull out the ums and the ahs with a couple of touch of buttons. It will automatically pull out a few highlight reels for us, it will automatically create the transcript. I mean, it's incredible how much time it saves and enables us to keep with a small team, keep producing, you know, a Renegade Marketers Unite episode every single week. So yeah, yay for that tool. Okay.

Jeff: Note on Descript. The feature that I really think is cool is when we do videos like marketing, videos that are talking head style videos, right? Problems used to be that people, if they didn't have a teleprompter in front of them, that they'd have to look over to the side right? Descript has this really cool feature where, even if they look over to the side, it will reprogram the eyes and the facial movements so that it always has direct eye contact with the camera, and all you have to do is push one button and it edits the whole video for you.

Drew: Oh, my God, in real time on this show, because I've got my window over here and my script over here, and I keep looking at you over here when I should be looking there. That's hilarious, but that is a little thing that does matter, right? And the fact that they can fix it, that's incredible, all right. Well, we've covered the gamut so far. So we're going to take a quick break while I talk about CMO Huddles. Launched in 2020, CMO Huddles is a close-knit community of over 350 highly effective B2B marketing leaders who share, care, and dare each other to greatness. Given the extraordinary time constraints on CMOs, everything about CMO Huddles is designed to help leaders save time and empower them to make faster, better decisions. So Adriene, Jeff, you've been in CMO Huddles a long time. And Karen, I know you're relatively new, but I'll start with Jeff, because you've been in the longest. How is, you know, maybe you could share a way CMO Huddles has helped you.

Jeff: I think the top value for me has been in the networking with other CMOs and other marketing leaders, getting to know them, having one-on-one meetings with them, and learning about their challenges and solutions, and being able to brainstorm together with them ways to overcome issues that I'm facing as well. That's been really fulfilling for me to help others as well as like, I've gotten a lot of value from learning from others, so that would be my top reason for doing it. But there's a lot of others.

Drew: Sort of an irony, because we're talking about networking and human connections on a show where we're talking all about machines replacing all sorts of things. That is one thing that machines will have a struggle to do. So I appreciate that it is also number one on the list of folks, of things that CMOs say when they've they're in transition. The one wish they had is networked a little bit more. So we do try to bake that in anyway. Adri, you know, I don't want to force anybody to say something nice, but jump in.

Adri: No. I mean, the networking is funny, you say. But I actually think in this digital era, the human connection is even more valuable than that. And I would say real talk, you know, like, not just in these forms, but also, like the slack conversations we have. I'd say one of the things for me that's been, like, super helpful, believe it or not. I mean, even though, yes, hiring is every job we have posted, there's like, thousands of applicants. I'm sure everybody's having the same thing, and yet it's so difficult to really find, like, what's the right people and all this stuff. So having a network that you can go in, you know, back channel people or ask for recommendations. For me, that's been another besides the networking, which is totally plus. It's been another great benefit. So I highly recommend this job is tough man, like being a CMO in B2B. So having peers, you know, here, like Karen and Jeff, that you can turn to have real talk is so so so valuable. So if you're not here, join us.

Drew: Awesome. Thank you for that. And Karen, I again, I don't want to put too much pressure on you, but if you want to, anything you want to add to that,

Karen: I think I'm just going to pile on that. I love the learning already. I love even panels like this, honestly, where we're just hearing and debating and learning from each other. And I love the fact Adri, that you said real talk. You know, too often, I think some of these networks and affiliations are a little too polished, so the fact that I'm looking forward to having real talk with everyone in this network.

Drew: Real talk. I love it. All right. Well, if you're a B2B CMO, who can share, care, and dare with the best of us, maybe you need a shortcut to B2B greatness. Take a second to sign up for a free starter program at CMOHuddles.com. So all three of you there, how do we balance? And this is a, you know, this is a softball question, but creativity, the humanness and just the ability of generative AI to sort of do massive integration. I'll give you a perfect example. We're working on our flocking, awesome t-shirt. We have a designer over here, and we got me playing on ChatGPT over here. And designer's perfect. Every little thing is in ChatGPT is lots. I'm just curious, how are you balancing that sort of the human creativity, with just the ability to push a button and get stuff.

Karen: So I'm, you know, I'm happy to start. I don't think generative AI should ever be used by pushing a button, at least for the time being. I really believe there should always be that positive friction of Gen AI supporting humans, right? And just sort of supercharging creativity. To me, it's a start, right? And when I think about marketers and what makes marketers great, it's that creativity, right? It's curiosity and it's creativity. And so I always say, Gen AI just fast tracked your starting point so that you have the time to be more creative. And I'll give you like a really great example. I love this. I was in Cannes in June, and I went to a breakfast that was actually hosted by Ted, and it was one of their lead presenters. And they laid out this premise that, you know, we all live in a sea of sameness already, and it's only going to get worse with generative AI. And they use the example of the phrase innovation is in our DNA. And, you know, sort of laughable. But, you know, she brought forward maybe eight or ten different brands that had that on their website or in a press release, and then she actually showed a prompt that she did in ChatGPT, how would you describe your company being innovative? And it came back and said, innovation is in our DNA, right? Because ChatGPT is reading all the data that's out there on the internet, and so that's why I always say it's a starting point, but it cannot be your ending point. It's got to be used in collaboration with our own creativity.

Drew: Adri what do you think? You did 100 different versions, right?

Adri: Yeah, I agree. And I mean, look it is funny. I mean, I don't know, Karen, but like, you're at IBM, so I venture to say, like, we're similar in the way that, like, I've made a career by embracing technology. But let's be honest, like 90% of, you know, marketers' jobs is like, again, operational, and 10% is creative. You know, it's like, oh, become a good idea. But then we gotta, like, build the thing. Go, like, you're creating T-shirts. So you gotta call the vendor. Go, like, you know, like, that's a lot. And so my hope is that production in that sort of, like you said, augmentation or start, help you. It goes faster, and you actually have much more time for that creative, for that thought. And I do think that there are things that, you know, another example I would put at the beginning, like, the marketer needs to be in the driver's seat. You need to be able to, like, oversee, because I agree, like, if I want to create a thank you email for all the listeners that we have today, and you want to do the same thing, we're probably going to get the exact same thing, if we just go to Gen AI, but if I bring in my personality, my voice or a quote or something that really stuck out to me, you know, and bring that into like the whatever, you know, we go to ChatGPT and try to generate it, then that's when we start to get this sort of like differentiation. And so that's why. So I think there's going to be even more bar for personalization, for individualization. And I'll give you one example too, about, like, the speed of bringing that creativity into an idea I have, one of my designers uses actually quite a bit, you know, a Gen AI, and the way that they use it is to sketch so, you know, like we're literally talking about, like, what we want, like the campaign and what we're thinking about, and, and he just comes up, like, really quickly with, like, is this the kind of things that you're looking at and, you know, like, here's what I'm hearing and stuff. And so, because he's a designer, and he's a visual person, you know, instead of writing a brief, you know, and going through all that stuff, he's capturing the idea with like, initial sketch that he can, like, generate really quickly with AI. So, you know, I think again, every single technology piece we've had has changed our jobs, and there is always the thread of like, quote replacing. But I think that the reality of having that, using it as a tool for augmentation, acceleration in an actual space for thinking and being creative, instead of, like, you know, the million times that you have to upload the list, update the tracker and all that stuff, I think there's much more upside there, or more likely to have that kind of outcome than the, you know, robots take over the world.

Drew: So Jeff counterpoint or totally in agreement on this as well?

Jeff: I think they made a lot of good points, but the way that I think about this goes back to how skilled are we at prompting and getting the most out of the tool. And so the way that I think of a prompt is that I'm giving instructions to a very smart intern. They don't know a lot about my subject matter expertise. They don't know a lot about the specialization that I want them to learn about. So if I can prompt that, if I can give them instructions that includes the expertise that I have, the specialization that I have, then the output that they can produce for me is a lot better and a lot more human sounding, and less generic, because the more generic the instructions, the more generic the response. And so I would just say specificity is the key, and that's why I mentioned earlier, like creating a Claude project around a specific webinar, for example, just as an example. And you go there and you say, here is the objective of this webinar. Here are the three insights that I really want to share with my audience, and I'd like you to come back having an understanding of my messaging framework and past webinars that I've produced because I've uploaded them into the project. I want you to come back to me with five titles, five descriptions, five calls to action. And I want each one of those calls to action to represent a different thought leader. So imagine one is written by Seth Godin and one is written by Drew and one is written by Adriana, and one's written by Jeff, and they have samples. I've uploaded samples from each one of those creators or marketing thought leaders, then what I find the output from that is like very good and requires very little editing on my part. And so what I've done like webinar investors, one webinar, but if you wanted to do it at scale, say, well, write me a 52 week email series, and I wanted to start with a problem, share a tip, give a call to action that references one of these 10 guides that I've uploaded, and have like a demo consultation with the end. So I'm telling it like these are the structures. These are the specialization notes. This is what I want to do. I am really the brain behind the strategy of that communication, and then it just goes and does like the work that the intern might do.

Drew: And you know what I love about that, and so helpful to hear, hear you say, it is that when you're new to these tools, you go into them and you say, write a 3060, 90 day plan. And you know, you get crap. But the problem, you know, bad input, bad output, and we all know that bad data in that so that's part one of this. Thing is that the more input you give it, and the more of that being original IP, the more likely that your output will be quasi original IP, right? And it's, and it's, there's a direct line to it so, and it's funny. So I write these Saturday posts on LinkedIn. You guys have probably seen them and and some of them have gone very viral. Every single word that I write, except probably the 10 that Grammarly corrects, is human. Not a single thing there is involved. But the image series, my little penguin image series, is the same prompt with a new variation, each time with a different theme. So it's already learned what those series looks like. So every one of them has a consistent sort of, you know, that sort of penguin in the office feel. I call it the office CMO penguin. And so I get these great images, and though that is pressing a button, boom, with a couple of prompts, but I'm still in the frame of mind that if it's my IP, it's got to be here.

Adri: Yeah, Drew. But for example, I'll tell you one thing that I do, for example, for instead of, instead of having ChatGPT be the or the sort of creator, I have it be my editor and my character. English is not first language for me, so I always need an editor. And also, for anybody that's a writer, we need eyes, you know, and so, for example, I use a lot, and I've trained, you know, my own GPT to be like my voice and all these things, but I asked it to critique my writing. What are the things that I could improve, not even to edit it, but just to critique it. And so I think you're absolutely right. There is like, value on us being the creator and but there's more than one way to use and I think, like, that's really like the air, like, this is going so fast, like the initial like, hey. Just going to make it prompt to do something that was, like, level one, you know? And I think, you know, Jeff, you were talking about some the reality, I want to ask, like, has any of you, like, changed your organization, your actual team roles for Gen AI, like, does any of you have, like, a model trainer, for example, I haven't, and I'm scared of that. I'm like, Am I like, really, really, really, prepare for what's coming? And I don't think I am. I don't know if you all maybe you're more advanced.

Karen: We're a bit lucky, just because within our IBM consulting organization, we built a platform. We call it IBM consulting advantage. It's multi model. We can work with ABM models, external models, open source for imagery and text. And what's pretty awesome is it's designed so that anybody who's using it can build their own assistance, chat with their own documents, and drive, sort of their own personal innovation. But when we see enough demand and where we see really good assistance, we're hardening them and we're productizing them. And so we've got, you know, consultants who are doing the role that you just described Adri, but things like persona assistance, right to your point, it's not always just content creation, but we've, you know, built a persona assistant so that we can talk to any buying persona and actually test the messaging that we've developed. You know, ask them, What do you care about? What are your five most important needs? And then we can integrate one assistant to the next so that we can create something, talk to a persona, tweak right the five different messages or headlines we're creating, drive legal vetting. Now, you know, so that we don't have to put everything in front of legal, but we can have some things just pass through and then integrate into work front, which is our workflow management tool, so that we can actually start to automate the processes. And I think that's where it's starting to go, where we're really going to see a difference in number of people. You need the roles that we all play right to that point of just freeing all of us up so we can focus on more of the creativity. But I do think it sort of begs the question of, you know, what are the different types of roles we need to be prepared for as well?

Drew: Yeah, it's such a great thing. And if you think about the org structure of the marketing department 10 years ago, it was very different. And, you know, and again, I that is something that we're going to focus a lot about is in towards the end of the year, in in our regular Huddles, and talking about org structure in the world of Gen AI, because I do think it is going to change, and has to change. I'm not certain yet about whether you need a prompt engineer on your team or not, because it feels like it's a skill that everyone needs to understand. Part two of that is, I'm so in agreement. We have a Drew GPT, which uploaded my two books and all of my articles and so and my team can use that. I don't use it because I don't like what it comes out with, but that's funny, you know, but the team can use it to get a draft. I did want to make one other point. I completely agree with you that running it through those things as an editor and to challenge the thing, what's funny to me is, in my Saturday post, at least, I like the little flaws that they might find. I like the grammatical errors or whatever, because that looks human and people notice.

Adri: It's the perfect imperfection.

Drew: Yeah and so I let those things, I mean, I don't have any spelling errors that I purposely allow, but I know there's some radical error. I could have used that instead of which at that moment in time, but it didn't feel right to me. So fine, call me out on it, because engagement matters. So there's a moment here where, and I think there will be a counter trend where 100% human made will matter, but for 99.9% of us, it won't.

Adri: Yeah, I, you know, I'll give you one for sure. Like, again, I'm bringing design, if you notice, in retail already, there's, there's a line or sort of a wave of more, like, what I would call, like, human made visual design, you know, Banana Republic, you know, like, and I think part of the reason is, like most of modern design, when it comes like visual design, is it's very like grid based, you know, very almost like machine made, even though it wasn't. And that is easy to replicate with machine. And I'm seeing, again, I like to look at trends way before they come to B2B. And I think like in consumer marketing and consumer brand, you can see a lot of those early stages and social media. And there is a wave in design for like, the human touch, the drawn out, you know, sketch. So I do think you're right in it. You know, we saw that rare CDs and LPs with my kids are upset, you know, like CDs came out in digitalization of media, and yet, at, you know, I think it was just like last year or two years ago, where, you know, vinyl overshadowed the sales of CDs. I do think, like, with every sort of like movement, there's a counter movement that, in. So in this, like, the counter movement of, hey, is this human, and is this human created? Is going to grow high in value, you know, versus, like the machine created thing. So I absolutely agree with that, but not going to go away.

Drew: All right, we had a question from the audience about a tech stack recommendation or two. It's such a broad question, I'm wondering if we can get it narrowed down, but perhaps a tech stack related to AI. Karen, any thoughts? I mean, you have your own big, giant thing that you get to use your own sandbox.

Karen: We do. And you know, I do think having the flexibility to work with multiple models and integrate is the way to go, because I still think the vision is, you know, orchestration and automation, as opposed to, you know, single use prompt. So we use, you know, our Watson X platform and models and capabilities. We use Adobe Firefly for text and image generation. We, you know, we also work with Microsoft, we work with Amazon Q and, you know, more and more, I think you're going to start to see all of these integrate, because the real value is when you can integrate the Gen AI into your tech stack.

Drew: Yep, okay. Adri,

Adri: I would, well, I mean, we're martech AI automation, so obviously I'm going to recommend interval. Here's how I would think about the stack. Let's just a little bit more sort of generic, maybe more helpful for people is like, I do think you have to all AI needs, first of all, a good, solid data stack. And when I mean about a data stack, you need to have, you know, whether that's Snowflake or Redshift or whatever database, but you need a place in or CDPs. But you need a, essentially, you need to have a place where you have that, that customer data, where you can also house your brand guidelines, like all of that. You have to think about what is your flexible and very, very, very accessible data stack that you have. Then you have to have, I think important, to have a content management system, right? So again, for allow to that scalability, you know, like we partner very close, a million of our customers use Mobile Inc, for example, you have to have a content platform that allows you to intelligently house, manage and match, you know, the different assets that you have, creative assets that you have, like your videos and stuff, to be able to do that orchestration, then it's what I call the activation layer, which is exactly, you know, like, what, what Karen was saying. So solutions like it or walk that allow you to activate all those assets and all that data into that personalization, that orchestration across channels. And I think super important is like you need to, you know, again, whether you do it in part and stuff, but having an analytics or some kind of other, like stack, like the analytics stack, you know, of course, I came from Tableau, so I'm going to plug in Tableau. But I know IBM also has, like, great solutions. It's really important for people, you know, to close the loop and to have that sort of, like Reporting Analytics system that works together with your entire stack. So that's how I think about the AI stack today, data, content, activation and analytics.

Drew: Okay, well, so we're so far over time, but I'm gonna, so Jeff, I'm gonna start with you. Give us two do's and one don't when it comes to CMOs looking to get more out of generative AI for their marketing, two do's, one don't.

Jeff: Two do's, one don't. Okay, so I would say one do is to try lots of different solutions. So what I've found is, like what we're talking about tech stack. I would just say that sometimes Perplexity is the best option. Sometimes Bot is the best option. Sometimes ChatGPT is the best option. Make sure that you know the strengths of the different tools and use them at the appropriate times. I would also say, look outside of just content creation and see other types of AI related tools that can help you save time. So I use Reclaim as a time and task prioritization solution. I'd really recommend people take a look at that. I use a tool called Scribe that automates the creation of step by step instructions for how to do something. So you just do it, you click around, you do the task, and it creates screenshotted text based, step by step instructions for your team. Oh, God, great training resource.

Drew: That's perfect. All right, let's see Adri, two do's when it comes to making more out of generative AI from marketing standpoint.

Adri: Experiment and train. I think I'm going to steal, like, I think I like Jeff's recommendation. Like, really invest on making it custom.

Drew: Got it, okay? Karen, two do's,

Karen: Motivate your team to get started and try and put it in the hands of your marketers. Just get going.

Drew: Get going. Don't sit on the bench. We'll wait and see. None of that here. Okay, well, thank you through all three of you. Karen, Adri, Jeff, you're all amazing sports. Thank you audience for staying with us. 

 

To hear more conversations like this one, and submit your questions while we’re live. Join us on the next CMO Huddles Studio, we stream to my LinkedIn profile—that’s Drew Neisser—every other week.

Show Credits

Renegade Marketers Unite is written and directed by Drew Neisser. Hey, that's me! This show is produced by Melissa Caffrey, Laura Parkyn, and Ishar Cuevas. The music is by the amazing Burns Twins and the intro Voice Over is Linda Cornelius. To find the transcripts of all episodes, suggest future guests, or learn more about B2B branding, CMO Huddles, or my CMO coaching service, check out renegade.com. I'm your host, Drew Neisser. And until next time, keep those Renegade thinking caps on and strong!