Lior Eldan, Moburst COO on Reinventing the Marketing Agency Before AI Eats It

Michael Koenig: The concept that AI can now replace people is maybe a little bit backwards because AI makes people more productive.

Lior Eldan: The kind of bottleneck of humans is moving to what to build, what to work on. People run to automate processes which aren't fully fleshed out, and then it doesn't work, and then they blame the AI.

Michael Koenig: If it's running well, you don't think about it. When it's not running well, you do. Hello, and welcome to Between Two COOs. I'm your host, Michael Koenig, and today I'm joined by Lior Eldan, co-founder and COO of Moburst. Moburst started in twenty thirteen as a mobile marketing agency, the kinda shop that figured out app store optimization before most people knew it was a thing, and it grew into one of the fastest-growing agencies in the world, working with Google, Uber, Samsung, Reddit, and a long list of others. Now, I've turned down agency guests for years, and I'm not totally sure why. but I changed my mind with Lior, and that's because Moburst isn't really one company. It's been three or four, and it started out in app store optimization. It then became a full-service digital agency, and now it's moving into AI products and going into companies to rebuild their operations from the inside. And if you're tracking overall technological inflection points, Moburst has touched a bunch, from mobile to Web two point oh, and now potentially the most consequential of all, AI. Most agencies find one thing that works for them, and they ride it into the ground. This one keeps reinventing what it sells, which is a much harder thing to run. So we're gonna get into how you actually operate a company that is in constant reinvention, why buying other agencies is harder than it looks, and then what happens when the business most exposed to AI decides to become the disruptor instead of the disrupted. Lior, great to have you. Thanks for coming on.

Lior Eldan: Great being here. Thanks for the great intro.

Michael Koenig: Let's talk about reinvention. It's something that's on a lot of people's minds as AI is forcing them to think about their own jobs and their company's stability. Old paradigms are shifting around business, and when there's more competition, because now it's easier to start companies, you you know, things actually change. And so tracking back to what I said, you started as App Store optimization, which is for folks not familiar, because it was quite a while ago, it's SEO for apps. As I mentioned, most agencies find one wedge until it dies. Walk me through the first time you decided to move the company off the thing that was definitely working.

Lior Eldan: Yeah. So I think it's not necessarily moving off, but building on top of. and I think the the story with, App Store optimization and mobile, which is our origins, as you mentioned, one we get, we got, a request from clients, so our clients started asking us, "Hey, can you work on our website? Can you promote our website? Can you do SEO," et cetera. And also we kind of figured, okay, we have this niche. Mobile is a big niche, but it's still a niche of, that is in the US, companies that are looking for premium agencies, et cetera, and we figured we wanna go after a bigger market. So we started kind of catering to our clients' needs, and as you mentioned, started to acquire, a bunch of other agencies to add on these, capabilities, around digital and build up, a a bunch of, new services, from the ground up. And, and yeah, that's how we kind of did the evolvement, basically, catering to our clients' need and trying to understand where the market is heading.

Michael Koenig: So there's a build versus buy decision with any acquisition. You could build it internally, you could go out and buy it. Walk me through that decision.

Lior Eldan: Yeah. So I think that, one of the things that we realized with the buy is that, with the build you you kind of take it, you know, step by step. You need to hire someone that knows how to do, then you need to go and sell it, and kind of build it slowly but surely, and then that person, employee might leave, and then you' you kind of need to start all over again. I think the What we found with buy is that you get the founder mentality. They usually, or the ones that we acquire, come with their, you know, own set of clients and business. so besides already kind of not starting from scratch, you get another founder, in the company, so someone that you know, will stick with you for long. and also that founder mentality of, okay, how can we, how can we do more? Can we cross-sell, this service to our existing clients? Can we sell some of, Moburst's service to, into their portfolio? So, we found that in many cases it's a win-win to go on the buy route. the build I think is more with, new stuff that I think AI is a good example. Like it's very new. There aren't that many firms that really know, like, how to do it. So in this case, we're kind of opting for, okay, let's bring in the talent, let's invest in the team, and let's build it from the ground up.

Michael Koenig: Well, we're definitely gonna get into the AI part of this business. What I'm curious about, because you mentioned it, is founder mentality. And, you know, just to expand on what this actually means for folks, the overall concept, and correct me if I'm wrong, but no one's gonna care about your product as much as the person building it, and that's very hard to instill in others. Is that something that you've cracked the code about? Like, how do you get your team to care about the company and the service and the product as much as you do?

Lior Eldan: Yeah, that's a good, that's a great question. So I think, A, thinking about agencies traditionally, and we can talk a bit more on, products and the stuff we're building, but traditionally agencies are essentially the product is the people, right? It's the knowledge, it's the skills of the team. So for us, the team was always It's, it was all about the team, and the story we like to say is that my co-founder and I, we actually, met in the Israeli army. and the typical kind of Israeli founders met in some intelligence unit. We were actually HR officers, which shows you that from the army we're, we were essentially very much geared towards caring for people I think a lot of how we structured the company from, the onset was to cultivate this environment that would be, that would, you know, give people, a place to do their, best work, make it a fun environ- environment, and have them being seen. So I'd like to think when, you know, when you get people to a place where they feel that they have an opportunity to shine and can do their best work and have them, feel that they're being seen, then they kind of, you build that commitment. It takes, it takes time, and obviously not everyone will stick with you for the long run. But I think we have, many employees in the company that have been with us five, six, seven, eight years, and more. so we see that this kind of approach of looking at employees, as more than team members, is, has worked well for us. and then on the founder side, as you mentioned, I think when you bring someone and it's their business and we structure our M&A in a way that is kind of, they have vested interest in continuing growing their business and as part of ours, then it just makes-- brings that kind of, mentality of not just doing the work, but really feeling that it's yours. So we're trying to cultivate it from the founders we bring in and from the employees we build through the way.

Michael Koenig: Agencies notoriously, especially the big ones, bring people in and grind them until they have nothing left. And it's a great experience because you learn how to hustle, but you also get exposure to so many different sectors and customers that you wouldn't otherwise get. But it generally is brutal. Now, coming from this HR background, probably in a pretty intense environment, the Israeli army, not for the faint of heart And then you also said that an agency business is largely about people. what breaks as you grow, and what did you have to build to stop breaking? How do you avoid grinding people into dust and keeping people around for seven years?

Lior Eldan: Yeah, that's a great question. I think, what breaks, I think as, the, a company or our company grew, I think at the one hand adding, you know, additional layers of management and mid-management, et cetera, I think every stage is kind of okay, like we at the one, at the one hand, we want to make sure, again, that everyone feels seen, and that, our We have an open door mentality or anyone can, reach out and speak with, any of the founders or managers. But on the other hand, we don't want, you know, all these hierarchies, in the company. So I think that's one thing, like how many, levels of management that you instill to, on the one hand, have, make sure that, the team is structured for success, but not make it a super corporate environment. So I think that's one thing that there's a nuance here in how it breaks and how to keep the team members for long. I think it's, as the company grew and, we have a few cases, of employees which started off, you know, at an entry level, and today are VPs in the company. So they kind of grew with us, to being a manager, as a team leader, as, and eventually a VP. So I think giving the the employees a place to grow and also, and you talked about this with regards to agencies and grinding, but I think because we transformed ourselves every few few years into, you know, a whole new set of clients' capabilities, then I think it's also, for the employees, it's kind of, it's a new company almost. Especially with all the businesses we acquired. So it's not just a mobile shop, it's not just a digital shop, but it's So I think also there's a lot of opportunities. We actually had team members from the teams we acquired that grew into roles at a totally different part of the company. and I think that's also, something that keeps, people with us for long, that there is that, those growth, opportunities, to even if it not, it's not the traditional manager, team leader, et cetera, but actually in different parts of the organization.

Michael Koenig: Well, this is interesting. Now we're getting into middle management, organizational design, and how you grow. AI has connected executives to the ground truth in ways w-we haven't necessarily seen before, and it's very easy. You don't necessarily need that layer in between that's translating action into what that means in terms of results. How do you think about this? I know it's on a lot of people's minds.

Lior Eldan: Yeah, so I think w-one of the things that was important for us, from the onset, and we-- I think it's a guiding principles in how we hire and grow our team is managers Do the work. They're hands-on. they're not just, you know, there to give orders and manage up and down. That's kind of our management philosophy. This is how we run the company. We're very, much hands-on with the clients, with the work. And, and I think once that is the mentality, then the middle management do not become just measurers. They're, you know, they're part of the team, they're part of the production. And I think actually in that kind of philosophy, AI, I think, only kind of improves the team and the output because you take, you know, rock stars, give them the, give them on the one hand team that can help them with, some of the nitty-gritty and take AI And, and add that to AI, now you can have a team of rock stars which just can do so much more. So, I think that there's, you know, also a difference in terms of, you know, sizes of companies and, but And, and the, once you get to a place where middle management is all about just managing up and down. But as I mentioned, I think what worked for us is making sure that, whoever manages also does, the work as well.

Michael Koenig: Yeah, it's interesting, and I align with your thinking here, which is, maybe I'll hop on a soapbox for just a couple of seconds. But the concept that you AI can now replace people, and so go ahead and do that, is maybe a little bit backwards because AI, to your point, makes people more productive. So why would you replace people to accomplish the same end goals when if you just gave that AI to actual people, you can exceed those same end goals? am I putting words in your mouth or is that kind of how you're thinking?

Lior Eldan: Yeah. I And I think that once you take, you know, people that have the domain expertise and are really good at what they do, they And, and that's what I feel is missing in the AI conversation, that they can ask things even if they're working with AI, the vocabulary they have, the how, exact and precise they can be in the design or coding or whatever the domain may be. They ask things that another person just doesn't You know, you'll ask for a website, but you won't be able to go into the padding and shadows, like, and all the different elements which makes, a website, amazing and not just, you know, very flat,.. So I think that that's the how we see it, that you take people that are super professional, you give them You give them AI, and that gives them superpowers.

Michael Koenig: Separately on that, just real quick AI tip. There's actually some stuff out there where you can have Claude Code or Codex, whatever you're doing, takes snapshots of what it develops so that it can actually see it and analyze and evaluate, and it'll throw in all of those nice polishes. Doesn't mean that that, front-end engineer and product designer doesn't need to be there anymore. Eh, you know, just little-

Lior Eldan: Yeah, yeah heads It's, it's,.. No, it's definitely improving, and I, you know, maybe at some point it will figure out those things themselves, but I do think that, - Knowing what to ask, I saw, I, listened to a podcast from someone from Anthropic, and how it's, increasingly it's not Like building is solved. what do we build? The AI doesn't know what to do that, how to do that yet. So it's the kind of bottleneck of humans is moving to what to build, what to work on, and I think that's where we need to, make sure the, our brains are being worked on.

Michael Koenig: The way I think about it is just anything that requires human relationships, taste, or judgment is something that is still squarely within the human's, capab- or responsibilities right now. Yeah. Now, because we're talking about people, we're talking about AI, I'm talking to someone who comes from an HR background, how are you thinking about bringing on junior employees now? Because many folks are looking at this and saying, "AI is a junior employee. We don't need junior employees."

Lior Eldan: Yeah. so I'll be honest, I do think, it's tougher in this new world to be a junior. but on the other hand, I think that, a junior with deep or meaningful AI that speaks the language, I think is just so much more valuable, because all these companies are now looking, you know, to take the, AI pill, and bringing on an employee that speaks the language, that knows how to, is sometimes the make or break of a company. So I think, we, still hire juniors, but I can tell you that any junior we bring, we we, you know, validate that this person lives, breathes, and not just works with ChatGPT, but actually builds stuff works with Claude Code, works with, Cowork, kind of thinks in a way which is AI first.

Michael Koenig: It's a great point. And I think the juniors, to your point here, the juniors, right, that everyone thinks AI has replaced, well, they're not that same juniors anymore. Junior employees are coming in as experts with AI because this is what they've realized they have to be. Well, look, before we keep going down AI, I'm really curious about your acquisitions, the strategy, everything. You've acquired several agencies. We talked about how you expanded your business based on what your customers are requesting. You, you acquired Kitcaster, a PR shop, a dev shop, and you've integrated this founder-led agency. How I guess the question is here, like what did you get wrong on that first one that you learned for the second one, and so on and so on?

Lior Eldan: That's a good question. I think one, the integration part is, and this, I don't think this is news, but you-- the, there is the exciting part of, okay, you know, we found this, we found this great agency and, you know, everything adds up and they want to work with us, we want to work with them. But then there's the actual inter- like how does, the day after look? Does, does their team kind of stay standalone? What kind of support do they get? and I think that's something that we are still working on to improve from each M&A and are kind of building, sort of a methodology on how to do that, 'cause there are so many elements from IT like, okay, are they on Microsoft? Are they on Google? How do we migrate? Do they, do they, correspond, continue corresponding with their clients with their Kitcaster email, or do they move to a Moburst email? What does that mean for client relationship? What does that mean for their brand? -hmm. So lots and lots of kind of second order effects, which requires some thought. And also many times there's some overlap, like, okay, so but they also do SEO. How does that look like? Do they use our SEO team? Do they have their own? How do we make sure that we're not, you know, fighting for the same, business? so I think those are the integration is probably the tricky piece, to get right because there are so many nuances.

Michael Koenig: This, so, you know, be straight with me here. If AI's writing the copy, is building the site, is testing the creative, and, you know, is only getting better and better and better at it, right? What, what, you know, what's the agency actually for in five years? Because you are forward-looking. That is what we've kind of, you know, concluded here Based on how you reinvent yourself. What, what changes? Yeah. How are you thinking about that?

Lior Eldan: So I think there will be, services or activities which, yes, probably an AI, will do and will do well. We're seeing that already, with, you know, some of the work of a campaign manager, for instance, Google and Facebook and TikTok and all the rest, increasingly are launching all these AI, campaigns, which do a lot of the daily work of those folks, for them. So you don't have that many, levers to pull. We-- You basically tell the algorithm what you want. I still think that there's, and to our point earlier, that the judgment here, - On, okay, what's-- given my target audience, what's the messaging? What's the brand? What, w-what kind of-- what do I want them to feel, when they interact with my website or my brand or my ad? Those are things that, once you get a-- you want to test a few direction, yes, AI can amplify it can test it, but I think there's a lot of the more strategic work. And also to understand, okay, now that I have this data, what's kind of next? How do I build upon it? So I think that the-- that that's where, I think the agency of the future lives in the more strategic work, the planning work, looking on, okay, we have, the AI campaign running on Facebook, on TikTok, et cetera. How much budget do I give each? 'Cause I need even an AI, and there is no AI currently, to look on my whole marketing, campaigns and see how I optimize, that and be able to also understand what I'm seeing. Like, what's the signals I'm giving it? Maybe I need to change my KPIs a bit. So, so I think that there's that element, and then there are certain services which we see, you can automate them. But for instance, influencer marketing, is still increasingly being done by human influencers. Yes, there's a certain niche of AI influencers but it, but we know that the majority and people still relate to human. so going after these, influencers and cutting deals with them, yes, AI can amplify that, but you need people to do that. And it's not something that a machine can just do because there are so many influencers out there. And, so there's also that part of actually influencer marketing and PR. You need to interact with journalists, and journalists hate AI pitches and kind of will disregard you, if they understand, if they see that you pitch them with AI. So I think in the larger world of marketing and communications, there's still lots to be done, and the orchestration part of making sure, again, that the whole machine functions, and that you're getting the results, and if not, like why or what do I need to so yes, you probably won't need, as big of a team. You'll, you can work with smaller teams, with the AI leverage, but you definitely need all those components we talked about, the strategy, the judgment, the taste. those are still not replaced by AI.

Michael Koenig: Well, what's also interesting right now in terms of business models and in terms of, differentiation and value proposition, is that we're seeing product companies moving towards services. Yeah. We're seeing consulting, agencies having just a golden era right now where everyone wants to know, "How do I implement AI in my business?" And you have all of these, you know, of these air quotes, experts coming in, who are very smart, and understand the technology, but they're having just this incredible boom right now. And as I mentioned, these product companies, AI companies, are now leveraging more and more into, or expanding more and more into services. Now, you are doing the opposite. You are a services business and you're launching AI products. A marketing agency selling AI tools on top of services. That's a services to product pivot, and those are famously hard. What, what's actually hard about it and, from where you sit, but also making you think, "Ah, we should get into the product game"?

Lior Eldan: Yeah, those are great questions. I And, and you-- I think you pretty much said it. There's Everyone is seems trying to get into services because everyone is realizing that for every dollar on service- on, software, there's $6 on services. So they want to capture that. and I think, we, as you mentioned, are trying to kind of look backward on, okay, if we can It's not just about not buying software, but it's about let's kind of give the holistic end-to-end solution. And if we have those, all of these AI capabilities, A, we can make sure that our team, does the strategic taste, kind of judgment work and less the nitty-gritty. And also, we can just do so much more. We can, run much more optimizations. We can, monitor so much more things. We can just analyze a bigger amount of data. So, and some of it we're packaging as internal tools, some of it we said, "Okay, it actually makes sense to start to sell into clients." so we have a bunch of, different I think we have now around 10, different products which we offer as part of our, you know, stack. So you get the services. You can also You can take the products without the services. Obviously, the ideal is to work in tandem and to build on top of or give the service on top of the products we build. And we're still in the early phase. I mean, we have a, we have a vision of how we want this. We're thinking about this as a growth OS, like an operating system for growth. That-that's kind of where we want to evolve in. but we're still in the early stages of actually understanding what pieces of the stack do we build, what do we buy, what do we do. But early, engagement from customers is quite amazing.

Michael Koenig: That's fabulous. well, now we're talking about internal tools. you and I spoke previously, you set up an AI champions program, which is essentially a person in every department, plus you have an, a central AI build team. one, how do you keep that from becoming theater? That's the first thing, because I've seen that quite a bit. And then two, I told you we'd get back to the build versus buy in terms of AI. How are you thinking about what do we build versus what do we buy, keeping in mind all of the things that come with building versus buying? how do you continue to, maintain and innovate on something you build when you have a 200-person company that can do that for you if you just buy it?

Lior Eldan: Yeah. I think the way that we're thinking about it is it in the, is this capability strategic, and is it in our best interest, as you mentioned, to build it and to build on top of it? if the answer is yes, it's, you know, a capability that we want to make sure we can control and also certainly in the world of AI, does it exist, right? If it doesn't exist, maybe there's an opportunity. We'll probably opt, to build it. but I can tell you that like a CRM, for instance, we ask ourself, "Is it strategic for us to build our own CRM?" No. Like there's these amazing companies, so we ended up, taking HubSpot and, you know, investing in building that as the the place where all the the client context sits and all the other different AI tools go to HubSpot to understand, okay, what's top of mind, what was in the recent discussions, email, Slack, et cetera, and that's essentially the place where the customer context lives. So th-that's how we think on build versus buy with regards to software.

Michael Koenig: So y-you think about it in terms of, essentially building the connectors that are connecting the internal tooling systems so that you can get a better visualization around this. You mentioned something very interesting, which is something strategic enough that we should build this versus buy? We previously talked about founder mentality. How do you give people agency? There's a thing here, which is how do you get people thinking strategically Where, they can themselves start to think, "Is this worth building versus buying?" because the strategic aspect, this is a very senior aspect that comes with experience, that comes with being the founder, that comes with getting punched in the face a lot. So how do you do that? like, it's so difficult to train people to think strategically.

Lior Eldan: I think what worked for us, and you mentioned the AI champion, so we did a few things in our kind of journey of, implementing AI internally. And one of the things, so the AI champion was one to have a team member which we train, which we upskill towards AI in every department. But then on top of that we started to do, AI hackathons, which is essentially to take a whole team, put them in a virtual or physical meeting room for a day and tell them, "Guys, you have a day to build an MVP." some training, planning, et cetera beforehand, but essentially like one day come up with,.. And we did a few of those, and the first one was, you know, people were really uncomfortable and didn't know what to do. And we we made sure they have some, technical resources on call, so if they, you know, hit any roadblocks, we can help them with that. But after doing a few of those, we really saw the kind of AI pills, like people starting to understand, okay, this is-

Michael Koenig:

Lior Eldan: What I can do with this technology. And I think there is no way you can conceptualize, but until you actually, you know, try and building something with Claude Code and kind of hit some walls, but then see that you can overcome it, and what's the potential here? I think that's what got the really the, inertia to start having people come up with their own ideas and building their own stuff, and their own kind of, products. And what we are trying to think about now is, okay, what You're talking about data. What is the data we wanna make sure we have and can control and can have governance, and do it in a way which gives the capability to people to just build on top of whatever they want. So if it's, the CRM data, if it's campaign data, if it's AEO data, PR data So make sure we have all that in built in a way which is very accessible, has all the context, so an AI can do whatever, the team wants with it. But. And the teams can build on top of it. So, that's kind of our journey in how to instill that, AI thinking into the team.

Michael Koenig: Interesting. when we spoke previously, you lit up, as did I, about the forward deployed idea. for listeners who aren't familiar, this is a old model. Think your mainframe computers back when they first came out, IBM sending out these forward deploy engineers. The more, recent is Palantir setting up these forward deploy engineers for companies that really need to integrate their data into highly technical systems. Now, obviously that's not yours, but you lit up about this, embedding, a team to transform a client's whole operation, not just its marketing. Now, you did a proof concept with a mortgage company. Walk me through what that actually looked like.

Lior Eldan: Yeah. so one of the things that got us to start to think about this is we, you know, we-- I spoke about our, transformation to being AI first and for us it's kind of we understood that it's existential and it's we're all in AI, right? But then we come with all this enthusiasm to some of our clients, and they were like, " what's AI have to do with us?" and that kind of got us to think, okay, like there's a gap here in this industry. They know And these clients know that AI is a thing, and it's good, but they just don't see how it applies to anything outside of marketing or to the business or day-to-day. So we we did a few of those, but the one I told you about, we ended up doing a few on-site days with, this, mortgage company. and, doing an on-site of actually, you know, walking through what they do, how they do with their different teams. and it was amazing because you see the, these, you know, businesses which are, built on processes that have existed dozens and dozens of years, and having people do stuff which AI can do, and free the team members to do other stuff. So all these from, thinking about like there's actual, there, the the company we talked about has team members which look through forms to make sure it was filled properly and thinking about customer experience and what context do you want to have the account executive speaking with a customer, and what happens if he calls during the weekend and he's at home. And all these kinds of, like real world problems, but problems that there are-- there's a solution, we just need to figure out how it lives in the context of, okay, there's, you know, certain regulations and everything has to happen on-prem, et cetera. but that kind of got us to understand, okay, there's so much potential and so much, solutions we can build, from, just from what we know. And I think one of the things as marketeers, and we build websites and stuff like that we got used to thinking on how can we What's the experience here? Okay, like what's the Somebody needs to, like that mortgage company, a big pain was getting the users to actually like fill in the forms and sign it properly, and then filming on their iPhone, and then it's blurry and like they said it like can be a 15 minutes back and forth between the account And this sounds like, okay, let's build, you A very, easy tool to give, to the customer to make sure that they scan and to give him the info if it's blurry and how to do it right. So I think just coming from a digital perspective on how to do this right and layering AI on top of it, and from, we did this with a few different clients. There's also many similarities with, you know, call centers, and intake processes and what kind of experience, what happens when a client calls after hours, what experience do you give them? so yeah, we're very excited about this kind of AI transformation service, which we're just starting to set up because when-- we're not the only ones. You mentioned, like, all the AI companies themselves are starting to, launch these because this is this is the, I think, the barrier to actually using AI is getting it, getting into the hands of the people and building on top of it the solutions that they can use in their daily, working lives.

Michael Koenig: Yeah. It's interesting, within the tech bubble, everyone's thinking, "Oh, the world is using AI." And with all of these advanced use cases. And yeah, ChatGPT has nine hundred million active weekly users, but the vast majority of those are like, you know, "How, how should I help my sixth grader learn, you know, this math concept?" Or, you know, rather maybe that's even too advanced. But, it's so-- I take a lot of, or I get a lot of excitement from talking to other verticals that, have not actually implemented AI. So I love the way the business is going. Let's shift gears here. when we talked, I got really excited because the core business of demand gen is 100% changing. So at this point Like, let's talk about AEO. your AEO bet is that controlling the brand's whole digital footprint, not just the site, is the edge in an LLM world. So how do you run a service line whose rules change every month?

Lior Eldan: So maybe I'll just preface by telling it how-- telling you how we kind of stumbled into it, where we Started looking at, you know, how we as Moburst are coming up. And as you mentioned, we do a bunch of different services. and one of the things we realized is that, it's not just the it's not just your website, not just your SEO, the LLM is looking on all your digital footprint. It's looking on a podcast, it's looking on the articles written about you, it's looking on LinkedIn and the articles and social and Reddit and, Wikipedia and all these, and videos, YouTube. So it's all of your digital footprint. And then we realized, hey, like we basically touch all of those elements which the LLM is looking at, and we're actually in a uniquely positioned to actually move the needle. So it's not just, let's go in and put LLMs TXT and help improve your website. it's let's do these, you know, holistic approach to what-- how do we get an LLM to understand the brand. And one of the interesting we discovered again, from, the work we did for ourselves, if you look at best mobile marketing agencies, Moburst is probably top two, top three, however you ask that, because we have a lot of, content, a lot of, relevant But if you ask on the later, like podcast agency or PR agency, we don't have-- it's-- we don't have that much of historic, content for us. So one of the things we're now doing to set ourselves up for success is moving towards a house of brands. We'll have, different brands, which is the acquired brands, doing like doing the stuff that they're, best known for. So Uproar would be the PR agency. So I think- it's not just the mechanics of how the machine works, it's how you, approach it strategically, where you understand that it's-- the LLM will spit out, like, it understands, like, one thing about you, what you want it to understand, what you want, it to comprehend you're good at or you're best at, and how can you have, a cohesive digital footprint, essentially. So have the LLM look, and if it sees that there's, you know, mixed signals, it will recommend someone else. so that-that's how we stumbled into it, and one of the data plays we're doing is, and I mentioned this earlier, of connecting all the different data points, is that if you, do the PR of a brand and do, the podcast and, YouTube, et cetera, and on the other hand, we monitor daily, how a brand is ranked on ChatGPT, Claude, Perplexity, et cetera, we can start and identify trends. Hey, a PR just went live on Forbes. We're seeing that move the needle on ChatGPT. Okay, maybe there's some connection. so we're actually building now the kind of, a connectivity and doing all these tests to understand what actually moves the needle across each of the LLMs. And, and we're seeing that, like ChatGPT and Gemini have totally different algorithms and different things, affect them. and then that's one of the reasons, talking about kind of how we reinvent ourselves. So this is one of the ways we reinvent ourselves because there's a whole new, world, a whole new ecosystem for us to test and play in.

Michael Koenig: Well, let's talk about, COOs specifically. Drawing on your experience, w-what should a COO listening to this take away in terms of running a services business or even a product business where the ground keeps shifting? you know, what did you get wrong? What do you know now, and how should we be thinking about it?

Lior Eldan: So I think from a, operations perspective, I feel it's a lot about the people and the processes. and especially in the context of AI, people run to automate, processes which aren't fully fleshed out, and then it doesn't work, and then they blame the AI. but making sure that whatever process you have in your business works manually well, and that you have the different team members, you have the outcome you want, and only then going ahead and automating it. And, and thinking about the automation piece, context is everything. And, we're seeing that especially in the agency context where there's, deliverables that need to go through multiple departments, so strategy and creative, and media and digital, creating the landing page, et cetera. So you wanna make sure you have that shared context layer, amongst team members, but obviously in the AI context to make sure it understands the different perspectives. yeah, so thinking about the, a COO's role today, I think AI opens so much possibilities, but on the other hand, you need to make sure that, you are-- you have the right, people and processes in place, before you go ahead and, you know, and automate it.

Michael Koenig: I told you we'd get into actual COO-ing. Yeah. You know, one of the things that you said to me was, and I'm paraphrasing here, but it's sort of the ugly process, the ugly grinding, how the, you know, the hotdog gets made type of thing, that happens, with the COO aspects and that are the least appreciated, at least in your world. Tell me a little bit about that. what are some of those things, at least, you know, in the agency world?

Lior Eldan: Yeah. I think, and I don't know if this is a phenomenon outside of the agency world, but, you know, I studied in uni industrial engineering, and when I Thinking generally about the, you know, operations and COO role, you think about supply chains or, you know, where there's like physical operations. And yes, an agency, specifically a digital agency, obviously, operations are all about people and computers and systems, right? There, there is no, supply chain other than that. and I think there, yeah, the fact that stuff just works and happens and, you know, there's the systems in place that,.. And, and, in some conte- in some way, I think it's almost invisible, which is a good thing, that stuff works. but yeah, as I mentioned to you, I think sometimes it's underappreciated the fact that, things just seem to, work. - And, and in my specific case, it's through a lot of trial and error of, okay, these are things that we do or not do. These are folks that we hire or do not hire based on, certain personality traits, and how we structure our teams. And, and this is something we think about a lot now with But, and I'm sure that AI will, change that as well. How should teams be structured once you have AI team members or AI doing, part of the work? so the, that piece of the puzzle is something that is, A, ongoing because a business changes and evolves, and is I don't know. I'd like to think of a COO sort of like a project, a product manager for a company thinking about different, you know, processes as features and how can you improve them and put them in a sort of roadmap and understand where things might fall, and now let's build a new process for it. So that's something that I'm constantly, working and thinking about.

Michael Koenig: It's one of those things where if it's running well, you don't think about it. When it's not running well, you do. I actually got a, when I was COO of Tucows, which, public company, and you have an investor day with all of your institutional investors, I got up there, the last line I said was, "Listen, if things are going well, you're never gonna hear from me. So hopefully you never hear from me again." And then, you know, mic drop type of thing, walk off the stage. so, well, Let's talk about the operating system of the business. you know, what does that look like? What's the cadence? Do, do you all use some sort of, you know, methodology? Like how do you think about this? How do you set goals? How do you execute? How do you track execution, align expectations, hold people accountable?

Lior Eldan: That's a great question. and I think we tried different things through the odd, through the years. We tried OKRs, we tried a bunch of things. the way that we're increasingly doing, especially as we're growing more, is leaning more and more on like bringing strong, managers and bringing that founder mentality, giving them the support they need, putting the the guardrails they need, but basically let them run the business as they see fit. And we're seeing that if, you know, a few years ago we had one way of doing stuff for all the departments, today there are nuances. There are teams which have a different methodology in terms of how many times they meet as a team or how the some of their internal processes. and that seems to have worked well. We have the way that we operate as a company in terms of, okay, this is the cadence of our company meetings, this is what we do, this is how we run employee evaluations, et cetera. But, but all the stuff in between, how the work is actually being done, is up to the, specific department. And, that has been working really well. And I actually saw, this year, I attended at some conference, a panel from Supercell, which is like a huge app developer. and interestingly enough, that's how they run. They have like every game is like a company Which, and they told like it's even a different, like a size of studio is depending on the who manages it and how he wants to structure his team and all of that. And I think that's, just bringing that founder mentality of telling the VPs that we, you know, it's We trust you enough. It's your domain. You get the guidance. You get, again, the framework. You get support from HR and finance, whatever you need. But, letting-- giving them the freedom to just to And also just on the goal stuff, different managers are treating that differently. We have a goal in terms of this is our revenue goal, this is how it applies to every department. But how they actually go and ahead and, you know, achieve that is really up to every, VP and every department's own, methodology, discretion, and however they see fit.

Michael Koenig: Yeah. So I was actually just gonna say that. If I'm listening correctly, it means that KPIs rule the day. Yeah. And you set overall KPIs, you know, key performance indicators, whether they be revenue goal, whether they be new client acquisition, product development, whatever it may be. You set those at the executive organization level. You let your managers know, "Here's what we're targeting. Here's what you're going to be held accountable to, getting these numbers. Now go off and figure it out." And then I'm assuming because you said finance is there to support you, et cetera, you're there to also consult with them. but, let me ask, you know, those revenue figures, it's art and science to figuring out what reasonable goals are. How do you go about that?

Lior Eldan: That's, that's a good question, and I think it's, yeah, it's, as you mentioned, it's art and science, and we've, - were, evolving, I think, through the years. we actually just, brought in an FP&A to help us with some of that, like to help us With making sure that we, okay, if we're seeing where that a certain department is missing the marks, why, and does that mean we need to change, the goals, et cetera? so I don't say that I have all the answer then, but I also think that, AI here can do a great job of helping you surface, some of the reasons why, and maybe the trends that otherwise would, you know, take an FP&A department to do. Yeah, so it's kind of an ongoing thing. We see that, like this, I think is the one of the first years, that we're seeing that we're, on pace to on track for our goals, which is shows me that, okay, we, you know, we had years where we overachieved, which means maybe we didn't put aggressive enough goals. There were years where we underachieved, which shows me that we're kind of calibrating right.

Michael Koenig: I don't know. I would challenge you on that. If you're on track with your goals, they're not ambitious enough. I'm kidding. I'm kidding. Well, you know, Lior, it's time for my last and favorite question, which is, we've all been in the seat, and the craziest things get bubbled up to us because no one else knows how to deal with them. And sometimes you see those things where you just go, "Wow, I never thought I'd see that." Do you have one you can share?

Lior Eldan: I have a few. but, yeah, a few that comes to mind. I think one, and this is, you know, just a, personal anecdote I think of being an Israeli, company is the fact that, as, I think most, you and most of your listeners know, Israel is kind of on an ongoing, wars in recent years. We obviously work with US, based clients and being in a situation where, okay, you're in the midst of a call, there's a siren. Like, do I go off? Do I And, and seeing, A, that, you know, kind of the majority of clients obviously are very receptive, but but B, that the resilience of the team and taking on calls from bomb shelters and taking on calls, during sirens. So I think that, is something that, yeah, well, I didn't see coming 13 years ago when we started the business. but I think, yeah, after you've, you're kind of working through those experience and, you know, managed to overcome it, you feel that, the team It, it, I think it really brought out, the resilience of the team and being supportive to one another 'cause people live in different areas that are affected differently. so that's one of the things which I think, grew us, made us, I never thought I'd see, but definitely made us, stronger after

Michael Koenig: I never thought I'd see that either. I was in a board meeting, and one of our board members is Israeli, located in Israel of course, and halfway through just said, "Hang on, I have to relocate to my bomb shelter." Yeah. And all of us were just like, "Boy, is this a different world." And, and now all of us are kinda like freaking out. he he comes back on, he's in his bomb shelter, and he's like, you know, "Okay, let's keep going." And we're like, "Wait, wait, are you sure? Like, you are under active bombing. this board meeting doesn't seem like the most important thing anymore." And he's like, "Eh, no. Just, just keep going." Yeah. so the resiliency, but also, yeah, that was something I never thought I'd see either.

Lior Eldan: I think it goes nationwide, but, yeah, person- like from the personal experience, that's definitely And, and I think even, you know, clients and US-based colleagues, et cetera, like initially that was the reaction. But at some point, yes, obviously they care a lot, but it kind of got normal that, okay, there is a siren and we're just continuing the conversation from another, place, which is also crazy in and of itself.

Michael Koenig: Yeah. it's crazy. That's crazy. Well, it's, I'm there with you. Lior, well, this has been awesome. I really appreciate you taking the time, I appreciate you coming on and, you know, and breaking sort of my prejudice against agencies here.

Lior Eldan: I hope you'll bring more. I'm happy to continue being the sole one. No, but this was very fun. I really appreciate it.

Michael Koenig: No, we'll have more come on. well, look, thank you so much and, thank you to you all for listening to this week's episode of Between Two COOs. I'm your host, Michael Koenig. A very special thank you to Lior Eldan, COO of Moburst. And tune in next time for another fabulous chat. go on, you know, Spotify, Apple, wherever you listen, and give us a review, give us a rating. We'd love to hear it. Share it with other people so that they can get great advice from other COOs. And until next time, so long.