← The Digital Marketing Podcast
The Digital Marketing Podcast
The Hubspot CMO Interview - Kipp Bodnar on AI, Answer Engine Optimisation and the Future of Marketing
00:20:34
Loading summary
Transcript41 lines
- [00:00]
A
Welcome back to the Digital Marketing Podcast brought to you by targetinternet.com My name is Daniel Rolls, and in this episode we're talking AI answer engine optimization and the future of marketing with Kip Bodnar, CMO of HubSpot. Okay, Kip, good to have you here with us.
- [00:22]
B
Thanks for having me, Daniel.
- [00:23]
A
It kind of feels like we're heading to another big wave of change, like we constantly are in digital marketing as we get into 2026 and January's flown by already. Well, should be well into it already.
- [00:33]
B
Yeah.
- [00:34]
A
Where do you think market is really to be keeping things on their radar? What is it they should be looking out for?
- [00:39]
B
I'll be honest with you, I think that 2026 will be the biggest year of change in marketing in history, likely. I feel like weeks or months happen in days at this moment and is completely crazy. I think if you're thinking about what marketers really have to pay attention to on a very focused level, it would be what are the new ways that your company can get discovered? And that would be things like answer engine optimization, the pending broader rollout of chat GPT ads, using AI tools to, you know, do better content production, to, you know, show up better in algorithmic discovery on social media, on YouTube, etc. Like, that is job one. Job two is probably what are the, what's the data that I need to use with my AI agents to drive personalization and a more customized buyer journey? And then what are the agents that I need to extend the effectiveness of my team? And how do I, you know, buy or build those agents? Right. And as I think those are the three things that are going to be the dominant themes of this year.
- [01:49]
A
Let's get into each of those a little bit then as well. So let's kind of maybe start with that agent workflow kind of piece. So what changes are you actually seeing in how people are changing their day to day workflows? And how is AI kind of baking into things? Because what I've seen is a lot of, there's a lot of people very excited about it, but they're at the stage at the moment where they're testing and they're testing and they're testing, but they haven't really baked it in necessarily. And obviously HubSpot offers a lot of the tools to do that. So it'd be kind of interesting to see what your take on it is.
- [02:18]
B
Yeah. So I think we're seeing a lot. I think what's interesting, right. When agents came on the scene, it was Kind of the opposite of the second generation of the Internet Web 2.0, the kind of the social web. Marketing was the most popular category of software and technology. There was just a ton of opportunity in software technology. But there's an awesome book called Range. I don't know if you've ever read it. In that book, there's that concept of kind problems versus wicked problems. You know, it compares chess, which while hard, like there are rules and you can optimize and you can learn and get feedback loops to like tennis, which is, wow, you can't predict the angles. They're not like there are far more variables and it's hard to get feedback loops on that. And if you extrapolate that to AI things like customer service, like it HubSpot, we're seeing huge adoption of our customer service agent. Our customer agent is going crazy. There's less adoption of marketing agents, I think, across the board. And if you look at even just startups getting funded in the marketing category, compared to sales or customer success, they're a lot less. Because the problems are harder and they're more complex and the best outcome is less formulaic. Which is one of the reasons I love marketing, one of the reasons I think you probably love marketing as well. And so with that, you have, I think, some very specific things that are happening in marketing. Marketing is more, I think, more about marketers using assistance, you know, like our Breeze Assistant and HubSpot, Claude, ChatGPT, etc. Than agents so far. And I think there's kind of the bleeding edge of marketers who are starting to build agents and they're building agents using HubSpot. They're building agents on N8N or Gumloop or, or with Claude code or with Lovable. All. All those things are awesome, but they're. Those are definitely like the minority of the world right now. The number one thing I would say, if you're somebody out there and you're like, hey, yeah, you know, I've gotten fairly proficient with using Claude or ChatGPT or Breeze or something to help me do some of my core marketing work. What's the next step? I have a special hack that I think is awesome. Google Gemini. I don't know if you know this. Daniel will watch any video up to an hour long and not just transcribe it, but will watch the video and can understand everything that happens. So take the thing that you have to do frequently, the most that you dislike, screen record yourself doing that thing, you know, going through the actions, narrating what you're doing upload that into Gemini and ask Google Gemini how you can use AI to help you automate that thing. And a lot of the time will tell you about an agent. Sometimes it'll tell you there's not that much you can do. That's kind of rare. But it's a great insight into like, oh, I understand how to take this thing that I know really well and take it to the new modern way of doing it.
- [05:17]
A
I love that. I think that's a really brilliant practical tip and I hadn't thought of that at all. So we're going to, we're going to give that a go immediately, straight away afterwards as well.
- [05:24]
B
Let me know how it works.
- [05:25]
A
I think it's really interesting because I think what you say about most people haven't adopted that kind of cutting edge side of things. Like we've been playing with Claude, Cowork and all these kind of things at the moment as well. And I think it's actually realizing the fact that, you know, we're going to be in this constant loop of adoption at the moment as well. And baking that innovation and adoption into what you're doing every day is going to become, has become really important.
- [05:46]
B
Yeah, the technology is already far ahead of where it needs to be. It's the human adoption, learning, training, evolution that's really the next big thing here.
- [05:56]
A
We touched on to AEO and it was interesting. We went for this kind of evolution of what are we going to call it? Is it geo? Is it ao? And then I remember on the HubSpot you had the HubSpot AI grader and it became the AEO grader and we've been showing that to people along as well. It's kind of changing the way we think about being discoverable and that's kind of breaking down traditional attribution to what you're going to take on things at the moment.
- [06:17]
B
What's interesting, right, is there was this golden age of marketing attribution with Facebook ads and Google Ads and traditional SEO where it's like, oh, click, visit customer. Like, yeah, there's some other stuff. But like, I feel pretty confident that it's like the majority of stuff's happening there. I can do an equation around that, assign some value and move on. Right. And you're bringing up the right point, which is when it comes to being found in ChatGPT or Gemini or Perplexity or Claude, that's very different. Right? In that all of those models look at different information and the tracking is much harder because there, there are two, two aspects of showing up in those results. There's mentions which just your company name is mentioned, but there's no link or way to get back to you. And then there's a citation which is like there's a specific link for a specific reason back to you. Right. So with the citations you have some of that old school attribution that you can do, but that's not the complete picture, the complete value that you're getting. And with the mentions that gets into like the age old attribution problem that we've looked at with some of the work in social media, brand marketing, etc. Where you're looking at, oh, what's my share of voice? What's my sentiment of the way I'm being mentioned? Is my number and density of mentions going up or down? Right. And how do I correlate that with referral traffic or direct traffic or other ways to understand how I am growing? Right. And that is a big change in how we measure an attribute. What was one of the core channels of marketing? Search and discovery.
- [07:51]
A
I'm fascinated by this because I'm pretty convinced that we've already been doing it wrong all along anyway from the point of view that we were relying on attribution modeling, but it wasn't actually painting as half the picture anyway. So it's, it's interesting we get into this space. So with this, where do you think the big opportunities for marketers are with AI search and this thing about kind of first lead advantage. Do you think there's an opportunity there?
- [08:12]
B
Yeah, look, one of my core beliefs in marketing is always that like there's a huge first mover advantage because it's a function of opportunity and competition. When things are new, they're less optimized, there's more opportunity and there's less competition. And like those two things lead to a real step function change. And also when you're starting something versus optimize something, the numbers are so small, you know, for everybody and everybody's kind of learning the same way. Right. Which is awesome that you can literally make exponential progress. You know, if you go from having 50 visits from chat GPT.com, you can go to 500 visits. That's an achievable thing for any company. Right. And that is literally exponential progress. And that's with where we are at today, is that you can actually really think about how you show up. And one of the things I like to remind everybody or kind of point out, you remember the early days of traditional search engine Optimization, it was super powerful. But what was it in terms of time to value slow? Right. I mean, I imagine, Daniel, you've ran projects where took 3, 6, 12 months till you got any meaning to search traffic. Right. These AEO engines, they're much faster, you know, same day. Like you don't have the same page rank crawlability issues of traditional SEO. So you can get time to value much faster and make that change happen much faster. Which I think are a couple really big opportunities for marketers in this changing landscape.
- [09:46]
A
I think it's a hugely valid point. I remember I used to run the search agency back in the dark ages of SEO and we wouldn't do a contract less than 12 months because we knew it would take us at least six months to even start seeing some movement. So yes, you're 100% correct. I'm going to pay you a big compliment now, which I don't normally do. We're interviewing people. But the key thing is if you look at HubSpot, I think you are as renowned for your product as you are for your marketing. And you've been CMO for getting on for 11 years now. And actually the way you've looked at that marketing I think has been hugely impactful and influential across, across all kind of aspects of marketing. So what are the changes that you've made at HubSpot when it comes to this? I've seen some of the data and there's some amazing numbers kind of coming out in terms of the impact that you've had and what other marketing teams can learn from that for how you've approached aeo.
- [10:31]
B
Yeah. So if you think about what we've done on the AEO side first, I think we made a decision. I think all great marketing comes from like strategic conviction. And we made the decision like we believe this is a core way every company is going to get discovered and grow in the future. And this has to be a core competency. Right. And this is not a nice to have after hours project. Like we have to be good at this and we have to spend the time and effort and money to be good at this. So that's step one. I think that I would highly recommend, if you're going to do AEO like you do it, don't partially do it, like commit to doing it. The next part of this is looking at the science behind AEO and what's different about traditional search to AEO1 is that a lot of AEO is about like ultra long tail, really complex queries. Right. Where instead of four to six words in Google, you're looking at 40 to 60 words in ChatGPT or Gemini or cloud or what have you. And it's like, oh well here's my business, maybe here's some data about my business. This is the problem I'm trying to solve. What tool should I use or how should I think about this? And that is hugely different than traditional search. Right? And so the number one thing I think we have tried to do is have more information available and have that information available in friendly ways. The LLMs, which starts at like how the robots actually can access your site, moves to how you chunk organized data FAQs really instead of long NAR. Like the best example is on our CRM page, like there's a section of what is a CRM. Like we don't assume that you know what a CRM is, right. And we have our definition to help shape how the LLMs define and think about CRM. Right. And that you wouldn't necessarily do that in the world of SEO because like that wouldn't happen. You'd have these much longer pages because you're, you were counting on humans coming, reading and parsing that all together versus machine reading and parsing it all together. And so that's definitely a big change we've made. The other thing about AEO that I think doesn't get talked about enough is that SEO was very much like, can I get some links, can I get some links? And can I get some good quality links? And yes, it's the infrastructure and architecture, right. But a lot of it was like a link game where like AO is much broader. You know, you want to know for certain prompts, what are the sources people are pulling from. And like on ChatGPT it's Reddit, but it's also like some software ranking websites and other things that in some way are like affiliates, like affiliate marketing can impact how you rank. Any like partnerships you do with creators or YouTubers, like that can impact their content could show up and impact how you rank. You have to have a more multi channel strategy to also succeed in this new world.
- [13:23]
A
So interesting because we talk about this new world and I'm just thinking about how do we balance off then traditional demand gen with these more kind of emerging AI driven channels that we're looking at as well. So how do we balance those two things off?
- [13:36]
B
I don't know if balance might is the right way to think about it because if I look at what's happening in the world and you can correct me if you're working with anybody where this is not true. It's like there's a changing way information is discovered both through organic search and also on social media because you have the rise of AI content, AI slop and everything on social media. So there's the plays for search engine marketing. Social media marketing are not as effective as they used to be traditionally. Right. Because of that, a lot of companies dare say most companies are now are moving time, money and investment to ads and spending more on ads. And because of that, you're seeing ad costs go up a lot.
- [14:16]
A
Yeah, exactly.
- [14:17]
B
You know, I think in the last year the cost across the meta Network is up 18% I looked and found the other day. And like that's a lot of money. You know, it's like it's almost losing one out of every five ads just to inflation of cost. Right. So that's why I question balancing, because it's hard to balance when you're like your marketing CAC is being shifted in such a dramatic way. And so the only way to offset those costs increase are to adopt new, cheaper ways of reaching your customers. AO is one of them there. There's also influencer marketing, there's AI creative aden optimization to do better targeting so that you can shrink that ad cost back down. There's like lots of plays, but you have to know what those plays are and at least be playing in one or two of those areas, I think.
- [15:07]
A
So that leads us on to a good point then. So based on that, what kind of skills and capabilities should marketers be kind of delivering what their teams need to have over the maybe the next year? If you kind of use your crystal ball, where do you think this is going? The skills they're going to need and they're going to need to develop that capability.
- [15:23]
B
I think that is a really tough question. I think curiosity and learning is number one. Are you curious and learning and can you show, not tell? Like, are you prototyping? Are you building, you know, are you building quick AI decks? You building quick prototypes in lovable or replit or whatever you're using? Right. Like, is that becoming the motion and that? It's like not theoretical. It's not word documents. It's becoming like, oh, we could do this and this is what it would look like. And it's like, great, now let's take that and take that to market and bring that to market in a real and meaningful way. I think that is like almost a way of working in addition to skills. That has to really change a lot.
- [16:02]
A
Yeah, it's interesting. We had the lead consultancy from Deloitte coming and saying actually that, you know, that baking innovation into everything, it can't be a nice to have now. It needs to be, it's essential to everything so that learn, test and test.
- [16:14]
B
The world's changing too much, costs are changing too much that you have to have this true kind of innovation angle on your team.
- [16:22]
A
So with all this in mind then, so what are the first steps? What if we want to go away and actually say, right, how are we going to move the needle? What are we going to do about this? What should people go away and think about doing?
- [16:31]
B
If you're thinking about your marketing strategy, largely depends. It's like, are you a solo marketer? Are you a team of 100 marketers somewhere in between? Like, it largely depends. But what you're. At the end of the day, what you're saying is, I've got some emerging channels that I want to take advantage of and how am I going to do that? I would suggest starting with AEO because it has high growth trajectory, it's still early. How am I going to track my visibility and how I'm showing up in these tools? And we have a great free product AEO grader that gives you a baseline, but then you can go and there's a bunch of great tools out there to to do that and then you can actually start taking actions to improve how you show up there. That's awesome. Then you need to, if you are not doing like fully AI enabled content, generation of images, of videos, of text, that's going to greatly accelerate your ability to tell remarkable stories at a fast pace. That now needs to happen. Right. And then the next part of that is like, what are the agents or workflows I need to build to work in this new AI first world?
- [17:37]
A
I think just on that, that last point as well, that there needs to be some investment up front to change the working practices because otherwise you're doomed just to keep doing the same things than experimenting, but not really embedding it in.
- [17:48]
B
This is the classic innovator's dilemma. I don't have time, so I'm going to do it the old way. I'm going to optimize the old way. And you run out of Runway on the old way.
- [17:57]
A
Exactly.
- [17:57]
B
That's the problem.
- [17:58]
A
I just want to give you a chance to talk about what's kind of going on with HubSpot and the changes that are happening with the platform at the moment. I just want to give a bit of a plug to Inbound, because inbound, if you're not familiar with one, is HubSpot Conference comes up every September. It's in Boston again this year. And it is my favorite conference and those that know me is the one that I go to every year as well. And we always get a good few interviews and it always sparks my thinking. And it's the one I go to and don't speak at on purpose because I get to just sit there and absorb.
- [18:24]
B
There's so much to learn. It's such an amazing group of people. I came out of that week just so energized.
- [18:30]
A
It is brilliant. And it's where we met the lovely Jay Schwederson, who's been on the podcast a couple of times. To Jay and the team as well. So what's kind of happening at HubSpot? What should we people be looking out for?
- [18:39]
B
Yeah, I mean there's so much happening. We have, we've got some awesome agents out again. Customer agent, we've got a great prospecting agent. It's going to be some awesome marketing AI tools coming very soon. So excited to share more on that soon. And really how we're thinking about this is that like at HubSpot we are the context layer for AI. If you've used AI at all, you notice the AI is just as good as the context and information it has. And we are fortunate across, you know, 300,000 ish customers that we have a ton of data, a ton of context. We know what great looks like and we are just honestly sprinting and unlocking that in the platform, in our agents. If you want to connect to cloud or chat, GPT or Gemini and access your data there, all of that is like really what we are trying to do and trying to really be the context layer for the future growing companies. And I would say that the last big thing is like I think marketing is getting disrupted the most. We came out with this loop marketing playbook at Inbound in the fall. We've got a, we've got a bunch of fun stuff coming this year around loop marketing and really maturing and giving people like the tactical handbook of how to actually do marketing in this post AI world. And that's going to be coming this year and very excited about that.
- [20:00]
A
Yeah, we'll put all that into the show notes as well. I mean the loot marketing stuff I think has really inspired a lot of people around this as well. So we'll put that in show notes. So as ever, target Internet.com forward/podcast kit. Thank you so much for joining us. Loads of really good practical tips there. I'm off to start dumping videos of myself into Google Gemini, and I'm interested.
- [20:15]
B
To hear hear what you learned, how that works for you. So please let me know.
- [20:17]
A
I'll let you know and we'll get the listeners to do the same. So thank you for your time, Zoe. Really appreciate it.
- [20:21]
B
Awesome. Daniel, thanks so much for having me. Thanks for everybody for listening.
- [20:23]
A
Thank you so much. For more episodes resources to leave a review or to get in contact, go to targetinternet.com podcast.