CMSWire TV
The Gist
- AI orchestration. Becoming an AI-native marketing team requires coordinated, cross-functional adoption of tools rather than isolated experimentation.
- Zero-click SEO. Marketers must optimize for visibility in AI overviews and LLMs by prioritizing consistent language, earned media and semantic clarity.
- Collaborative change. Leading AI transformation means balancing innovation with empathy, bringing teams along rather than pushing them through disruption.
What does it mean to become an AI-native marketing team — and how do you build one while the tech is still evolving?
In this episode of The CMO Circle, Paula Mejia, VP of Marketing at Wix, unpacks the real-world challenges of leading AI transformation at scale. From orchestrating cross-functional workflows and rethinking SEO for a zero-click world to setting smart guardrails around AI experimentation, Mejia shares candid insights on what's working, what's hard and what's next.
Whether you're just starting your AI journey or scaling fast, this is a conversation packed with lessons for modern marketing leaders.
Table of Contents
- Episode Transcript
- Staying on Top of the Latest AI Trends & News
- Managing 400+ People Within Marketing Processes
- The New Rules of SEO
- Measuring the Impact of AI Overviews and LLM Mentions
- Unlearning Outdated SEO Habits & Funding AI Initiatives
- The New Marketing Team & Leadership Lessons Learned
Episode Transcript
Michelle: Hi everybody and welcome to the CMO Circle. I'm your host Michelle Hawley, the editorial director at CMSWire. And with me today I have Paula Mejia. She is the VP of marketing at Wix. Thanks for being here today, Paula.
Paula Mejia: Thank you so much for having me. It's really nice to be here.
Michelle: Jumping right in, you mentioned that one initiative that's top of mind for you right now is, in the long term, evolving into a fully AI-native marketing team. So when you say AI-native marketing, what does that actually look like? How does that change the work day to day?
Paula: So I think first, probably many people in leading marketing teams are facing this tidal wave of questions. What is AI-native marketing? What does AI-native marketing look in my specific organization? And I'm a marketer, so context really matters and the answer is always it depends. And it does really depend.
And I think this seismic change in technology for me, especially someone who originally started in operations and marketing, really opened up a lot of opportunities to think about is the way we work the best way to work today and will it be the best way to work tomorrow? And what I find really both motivating and challenging is that nobody really knows what AI will eventually look like. Like, when is that hard deadline? It doesn't exist. And so we find ourselves in a place where we know we need to adapt to this technology. It must be orchestrated, right? If you're doing it in silos as teams, you probably won't get the level of benefits that you can from this technology. And so you're in a situation where you have to build the train tracks as you're driving the train. So for me, the definition of AI native marketing, I know will change almost on a day-to-day basis because I'm constantly pushing teams to rethink.
We used to do, for instance, localization in a, honestly, very manual way, and then a slightly more automated way, because there are some AI translation solutions. But now the next step is how do we integrate that into our email marketing system and the way we spin up sites, being that it's, you they're all built on Wix.
I think what's great about AI is for sure it's made individuals more autonomous and more efficient if they're using AI. But that level of use of AI will very soon reach the limit until you fully orchestrate the whole machine. And the whole machine cannot be orchestrated in one day as you kind of build little pieces. And there will probably be some tech debt that you have to like scrap along the way.
I'm speaking maybe a bit too theoretically. So I'm trying to give you a good example of what that might look like. But I will give you one. So we have been using AI, some custom built solutions for generating landing pages that are structured in a similar way to each other. And that alone has been an unlock in terms of the number of pages we can build that are already SEO optimized, etc. But it's just one piece of many different types of content marketing that we do. So the next step is how do you make sure that that level of efficiency is also applied, for example, to videos that we promote on YouTube and not just the English ones, but also the ones that are localized into our main languages.
So right now the teams are working on very different platforms with different software. And even though product marketing kind of sits at the foundation and the messaging is coordinated, the technology is still fragmented. So we're looking at, are we ready to merge them? Is there such a tool that exists? Do we need to build it? What does it mean to build it? So this is what I mean by building the train tracks as we're driving the train.
Related Article: CMO Circle: The Secret to Building a Brand That Lasts 100 Years
Staying on Top of the Latest AI Trends & News
Michelle: So with the state of AI changing day to day, like you said, how are you staying on top of the latest announcements, trends, and how are you deciding what's worth paying attention to and then what to kind of just put in the background for now?
Paula: I think that's probably the hardest part of what's happening now because there's an update every day and I, you know, I would be kidding myself if I said I always know exactly what is the most cutting-edge thing.
I spend a lot of time on LinkedIn and there are people that I follow who do this more or less for a living. I am very thankful to the CMO HubSpot for having really amazing content specifically on AI application uses for marketers, so I rely on them quite a bit. But we also have built-in infrastructure within the teams to make sure that the more subject matter specific AI tools are being explored.
So we have a group of... We call them "AI builders" who own for their teams the responsibility of exploring new tools and developing PoCs and understanding if the tool we have now is as good as the other tool. It's, I wouldn't say it's a full-time job, although it could easily be, but we found it really is important to involve as much of the organization as possible in the experimentation because I can't really know what is the best tool for a designer working in motion graphics because that's not what I live and breathe.
Michelle: What surprised you the most when you started to operationalize AI across a large team?
Paula: It's hard. It's really hard. Well, because AI promises, I think in the, you know, 2023, early 2024, everyone was like, AI will unlock it all. Like it's just, just use it. And I was like, yeah, I'm in, but I think, what I realized is the hardest part is this coordination between both the technology and the team. And if you don't have the right team structure, it doesn't matter if you have all the AI in place if the different AI tools don't connect to each other. Also, you don't get the most bang for your buck.
So it's just that it's a constant effort. It reminds me a little bit about the realization I had when I first became a manager, because I thought, you know, there will be a time when my team is built and we will just work and it will be great. We will be great at our jobs. And then you realize that's not the case because goals change and people move to other companies or they grow into different directions than you thought. So you're constantly, you know, adapting that team.
And I think, thinking about AI as the hybrid partner of marketers, it's very much that we are evolving with it all the time and will probably never be done.
Managing 400+ People Within Marketing Processes
Michelle: In a similar vein, with 400 people in your organization touching marketing, what did you change or what do do to help the work move together instead of in silos?
Paula: This is the constant battle. On the one hand, not to confuse silos with ownership, but there is something beautiful that happens when you have very clear owners, which is they tend, those things tend to run faster. The reality is that the, you know, our marketing team is very large and the organization itself is very complex. So it's very hard at Wix in marketing to do anything completely alone. And I think one of the things we've realized is that there were ways of working that maybe worked really well two years ago that don't necessarily work as well now, even kind of apart from the AI conversation, that's kind of always something that you're realizing is like, even product teams, so you kind of reflect a bit on the marketing side, what is in focus and what isn't. So I think this is an area that we're constantly trying to balance.
Right now, instead of thinking more about silos, for sure I am very much emphasizing ownership and orchestration. So if we ever run into a situation where because the team is so broad, or like in our particular marketing structure, for instance, the SEO team and the AEO team operate across many different brands, there can be situations where the processes don't quite match every single team. And so you get inefficiencies just from like the breadth of their work. So we're constantly trying to look at not just how you use AI, but how do you orchestrate the work in such a way that the humans that are behind it are still getting the efficiency benefit. It's, like I said, it's always a work in progress and I'm sure we can do better, but I would say we're really dedicated and always looking for the opportunity to improve.
Webinar
Sep
25
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Register
Webinar
Sep
25
Call Spoofing Trends Financial Institutions Can't Afford to Ignore
Don't let robocalls sabotage your customer relationships. Learn how to secure your voice channel.
Register
Webinar
Oct
15
Agentic AI Playbook Real-World Customer Service Use Cases You Can Deploy Now
Boost self-service by 30% and slash call volume by 63% with agentic AI.
Register
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
CX Reality Check: What Your Customers Are Really Thinking
Register now and get the insights you need to not just meet rising expectations, but exceed them.
Watch Now
Webinar
On demand
Content Leaders Collective: What a Good CCMS Actually Looks Like
Discover how content leaders transform clunky systems into growth engines — and why your competitors are ahead.
Watch Now
Michelle: When it comes to more of the safety side of things, what kind of guardrails have you put in place around AI that allows your team to still move fast but without as much risk?
Paula: Yeah, this, it's funny, I was talking to someone from my team who's kind of owning AI transformation and we were discussing where, at what point does an AI initiative require broader organizational oversight? Because even just like data management, before you even get into like risks to users and data privacy, and you can, I really think it's a philosophical debate, like how far do you go in one direction or another? And what I decided essentially was I just need to know on a case by case basis, what initiatives, which software we want to take to our security and IT team versus what is something that's so specific to maybe even the individual who developed their own whatever custom web app which you are able to do on platforms like Base44 and kind of letting them do that. You know what I mean?
I will say that as an enterprise company, Wix has very strong guardrails in terms of the types of software we buy as an organization. So that I find already a comfort. Security and IT work with procurement to make sure that if we buy, as an example, like ChatGPT accounts, that they've checked all those P's and Q's. But there are some things that we do, like I said, build custom and it's part of the complexity of working on this transformation.
Michelle: I don't know if you're involved in the buying process at all, but what do you look for as far as green flags, we want to use this? Or even red flags, like let's stay away from this vendor?
Paula: I do, I am very much involved in the purchasing of software.
So I used to work in B2B marketing. So green flags for me are always how quick the turnaround is from lead creation to outreach. So always keep an eye out for that. I am very sensitive to the sales conversation because I would love a partner. I really think of those types of deals as partnerships that is really transparent about what their software can and cannot offer. And that will usually come with disclaimers on, you know, we're really good at this, but maybe not the best at that. And if that's what you're looking for, maybe we're not the right brand for you or whatever. And we're even recommending a competitor.
The reason that that is a green flag apart from the obvious ones is that it just sets everyone up for success if the expectation and the product kind of meet each other exactly where they are rather than overselling what could be or what's in the roadmap that we all know how roadmaps work. They're never exactly delivered necessarily to the T. So I look for that level of transparency. And in my conversations and those demos, I'll often ask, you know, usually they start with what's so great about their product, what makes it unique. And I like to ask what isn't so great about your product, and if it's nothing, that's a big red flag because no product is perfect and it's okay, you just have to know what you're getting and what you're not getting.
Related Article: CMO Circle: The Real Cost of Marketing on a Shoestring Budget
The New Rules of SEO
Michelle: Shifting gears a little bit. What are the new SEO — or, I don't know if GEO is being used anymore still — but what are the new SEO best practices that you follow for LLMs?
Paula: So I think it's called AEO (answer engine optimization) now with GEO (generative engine optimization) being a subset of, of, of this. So GEO and, LLMEO and — once we all agree on the terminology, I'm like more than happy to get on board with what that is. There's so much and it's such an interesting place, I think, to be playing in, right? Wix as an organization that has a very strong organic growth team and watching that team adapt to these changing dynamics, but also that are still quite similar to SEO and balancing the two.
A couple things I think are very, very established at this point, and it's the value of earned media over owned media as a change. So being mentioned on product review sites like Captera, being discussed on Reddit. I believe Reddit and ChatGPT have this partnership agreement that cost ChatGPT like $60 million and it was closed last December because they're crawling Reddit to inform the model. And that alone tells you that if you are not playing that field, you're missing big opportunities in terms of where and how your product will be mentioned.
Because I think one thing that I guess most CMOs and marketers, especially in digital, are aware of is much of the buyer discovery journey is, and if not yet, will move to these conversations within LLMs. So it's not a question of whether your audience will ask, you know, what is the best website building solution for me or even what is the best, you know, running shoe for women working on a 10k, whatever that might be. It just means, you know, if you're not actively trying to train that, your brand won't show up, or if it does show up it might not be in the way that you want. So using that earned media really wisely and not just relying on owned media is super important.
And then in terms of your own assets, the proper use of the semantic description of your product and the consistent use is critical. So making sure that you, your brand team, your messaging team, whoever happens to own that responsibility, has agreed this is the way we talk about certain aspects of our product and doing the best you can, going back to the orchestration piece, to make sure that others are talking about your product with the same language, whether that's in user-generated content or when a journalist covers your product, that they're using those exact phrases, because that's how those brand vectors are built on LLMs and that's how they learn. When you say website builder, Wix is going to be mentioned in this particular way. But I think it's super exciting.
Measuring the Impact of AI Overviews and LLM Mentions
Michelle: How do you measure or track the impact of AI overviews in Google or answers from assistants like ChatGPT when it comes to referral traffic?
Paula: So we are able to some level of traffic coming directly from Perplexity, ChatGPT, etc. What we definitely know to be true is that there are many situations where that audience is not clicking on the link that's coming up.
And so we can track the link being clicked, but this zero-click growth is very much part of the world we live in. And attribution is not. It is not there. It is not directly attributable. And I think that it is okay. And, I love data as much as any marketer and if I could measure it all, I would. But I think in some ways we're going back to the type of, you know, billboard marketing, TV advertising, where you couldn't always know, but you need to be able to understand the lift that your LLM optimization efforts are bringing overall.
So instead of obsessing over which percentage of the clicks that I'm not able to attribute to any particular effort, they kind of are now falling in the brand bucket. And we do see LLMs and that discoverability path as similar to when someone watches a YouTube commercial about Wix, that is very brand oriented. And very few times that leads to an instant click. It's just kind of happening behind the scenes.
So when it comes to how we measure our success in this particular effort, I care more that Wix shows up in the answer to all relevant prompts in the way that we want, as opposed to directly measuring the specific traffic that's going. If I can get it one day, I'm sure we'll all certainly pay for that, whatever cookie or whatever the solution will be. But for now, given that it's not what I have, we're taking it a step back and making sure we are showing up and showing up in the way we want.
Michelle: What strategies or, content-wise, what are you changing to cater to that zero-click environment rather than, like you said, most people aren't clicking and going to the webpage?
Paula: So what we talked about, emphasizing consistent language use on owned media, ensuring that it's broader than maybe it used to be. We might not have done the page for a smaller feature, and now that's something that does help inform our lines on the full capabilities of the product. Making sure that on earned media, we're working very closely with our PR team to make sure we're covered in tier-one publications, because we know those tend to get more weight. Same for ensuring that from a product review perspective, we're encouraging happy users to go and leave positive reviews, which I think are areas that maybe in the past, marketing in any company would have been, that's important, but it's not the most important thing I need to do today. I think we understand now that Wikipedia, super important, Reddit, super important, specific affiliates are more valuable than others were in the past. And emphasizing where we know you can have a stronger impact on these models, which in and of themselves are always evolving. So you can't take your eyes off what's changing.
Related Article: CMO Circle: Behind the Scenes of a Multi-Million Dollar Customer Marketing Program
Unlearning Outdated SEO Habits & Funding AI Initiatives
Michelle: For established brands with deep SEO equity, what's the hardest habit to unlearn in the transition?
Paula: I think it's that orchestration element. I think really, really good SEO teams used to more or less own all of the real estate that would have made the biggest difference and could control every aspect of it. Now a lot of the impact, not all of it, right? A lot of it is still built on SEO principles. It's not that that goes away, but it's how do you as an SEO get your PR team to make sure that they pitch to this, you know, publication, and use this language at this level of frequency across all the GEOs where you operate. You know what I mean?
It's like SEO became a cross-functional team, whereas I think before it was a discipline within marketing, a really important one, but they kind of could be on their own doing things no one really got. I don't know, the Google algorithm always felt like very, very theirs. And so now they have to really integrate across and be spokespeople for those efforts that they are responsible for, but don't necessarily control.
Michelle: How do you fund AI initiatives? Is it like a separate innovation budget? Is it in functional budgets? Is it something completely different?
Paula: So from a product perspective, think everything we're doing is AI-led at the moment, AI native. So all of our product R&D has this strong focus. We don't have a separate budget in terms of marketing for really any initiative. We just adapt to what is going to bring us the most value. So I'm not sure if that answers your question.
Michelle: What's the biggest hurdle would you say when it comes to then proving the return on investment as far as your AI initiatives go?
Paula: I don't think about it that way so much. I think about it more, if you do proper procurement work, you should get the value you expect. And if you don't, it's more to do with what about the way you're using it changed. Maybe you're not using it enough, right? This happens a lot with software where, especially in martech, there's like 10 different tools and all of them do a little bit of something. So is there a way to consolidate it just because even the breadth of what people are expected to use is too much and that creates a lack of use kind of by default.
So I think about it more from, what is the team getting the most value from? Not necessarily because the product is or isn't the right one, just like as a whole, the combination of the structure or all the other software we're using, the goals that we have. Let's say, for example, localization of video is really important. Do we have that tool? Or are we still working on some other thing that is not quite the cutting edge version of this? And not to bring everything back to AI, but I think part of what's so challenging is that the landscape changes every day, so there's always a new tool to look into, explore, and yeah, I think it's a of question how you make sure you always have the best one.
The New Marketing Team & Leadership Lessons Learned
Michelle: Looking at your team, are there any new roles that you've hired for or upskilled employees into that you didn't previously?
Paula: New roles that we've hired for? I don't know that I would necessarily call them new roles. The marketer with like very broad expertise is fundamentally the type of marketer that works really well at Wix and also works really well with AI because AI is allowing people to, if before you only did XYZ as part of your role, expanding that because the tool is enabling you to take more ownership of the creation of an asset, like that will be very advantageous.
So Wix obviously acquired Base44 a couple of months ago and for sure this team is exploring much more kind of the full-stack marketer profile that I think most marketing teams are going to quickly adopt because part of what makes marketing teams a little slow is the handoff of, you know, from me, this person, to this person, so if you're able to kind of capture more ownership and not necessarily spread it across disciplines, I think it's a very interesting way to do that.
But I also think there are some disciplines that are going to be very hard to fully collapse into a broader marketing role. And I'll take like motion design as an example that is still quite technical. I think if you want to do it at a very amateurish level, you can with AI, but you know, the way we, for instance, do ads, no, we wouldn't do that.
Michelle: You've touched on this a bit already when talking about AI, but even if you're taking AI out of the equation, what are you learning right now that's changing how you lead?
Paula: What am I learning that's changing how I lead? So I think it has a lot to do with me personally, both as a manager and as a human. I do very well with change and I like things to go really fast, but I work with other people who are not built the same way I'm built. And because my role is AI transformation, I feel both the responsibility to push them into an area of discomfort, but also I don't want to kill them. Like I don't want them to be unhappy when they come to work. I don't want them to feel like I don't understand why we're using this software when that one was okay.
So a lot of the way I lead now, and I'm not saying I do it perfectly, by no means am I there yet, but I'm really trying to bring people with me rather than push them through the AI transformation machine that if it was just me and 400 other people that were exactly like me, we could have gotten around. But we're still working with humans and colleagues and people who have real and legitimate concerns. Sometimes I'm very convinced about something we should use or a different way we should work. And then I talk to someone and they're like, but why? And I'm like, well, why not? You know what I mean? And that's not necessarily always the best answer.
So just making sure that if I'm implementing a significant change, that people really do understand the specifics of why, and it doesn't mean they will always agree, but I do feel as long as we work with humans, and I think that we will for a very, very long time, even if we incorporate AI more into the equation, bringing someone with you and making the experience feel collaborative will make for a better work environment and output anyway.
So that's definitely something I'm working on.
Michelle: That's all the time we have today. Thank you, Paula, so much for being here.
Paula: It was my pleasure. Thank you for having me.
Michelle: Folks, thank you for joining us. Don't forget to check back next month for an all new episode of the CMO Circle. And in the meantime, we have other CMSWire TV shows that you can check out as well, Beyond the Call and The Digital Experience.
